Monday, 8:00 AM - 9:30 AMStream: Combinatorial Optimization Invited session Chair: Dominique de...

TECHNICAL SESSIONS

Monday, 8:00 AM - 9:30 AM

� MA-01Monday, 8:00 AM - 9:30 AM - Convention Centre: Ballroom 1

Some Models of CombinatorialOptimisation

Stream: Combinatorial OptimizationInvited sessionChair: Dominique de Werra

1 - A Multi-Start Heuristic Approach for the General-ized Traveling Salesman ProblemPaolo Toth, Albert E. Fernandes Muritiba, ,

The Generalized Traveling Salesman Problem (GTSP) is a variantof the Traveling Salesman Problem where the node set is partitionedinto clusters. GTSP calls for a minimum-cost elementary cycle vis-iting one node for each cluster. We propose a multi-start heuris-tic algorithm, which iteratively starts with a random set of nodesand applies a decomposition approach, combined with improvementprocedures. Extensive computational experiments on a set of bench-mark instances show that the proposed approach is competitive withthe most effective heuristic algorithms presented in the literature.

2 - On Fairness in Fair QueuingWieslaw Kubiak

The requirement of a fair bandwidth allocation to flows is a recur-rent topic in the design of fair queuing algorithms. These algorithmsresult in fair and stable networks. The min-max fairness, the ab-solute fairness and the relative fairness are used in the design ofthe fair queuing algorithms. We show that the min-max criterionis closely linked to the well known parametric apportionment meth-ods, the relative fairness is linked to another apportionment methodwhich however is not house monotone, and the absolute fairness isthe minimization of the bottleneck of a just-in-time sequence.

3 - Advances in the AutoGraphiX research programPierre Hansen

The AutoGraphiX (AGX) system for automated and computer-assisted graph theory has been developped at GERAD since 1997.We present new features of the AGX 2 version and their applicationto discovery of conjectures of various forms and of new families ofgraphs.


Optimization Under Uncertainty

Stream: OptimizationContributed sessionChair: Toru Nakai

1 - Risk-Sensitive Average Reward Optimality inDiscrete-Time Markov Decision ProcessesKarel Sladky

Employing properties of a collection of non-negative matrices weshall characterize risk-sensitive average reward optimal policies inMarkov decision chains, i.e. if the stream of obtained one-stage re-wards is evaluated by an exponential utility function. Conditionsguaranteeing independence of optimal values on initial conditionswill be discussed. Policy and value iteration algorithms for findingaverage reward optimal policies will be presented.

2 - Full information best choice problem when thenumber of observations has a negative binomialdistributionmitsushi tamakiWe consider a version of the so-called full information best choiceproblem with a random number of observations. N i.i.d. randomvariables with a known continuous distribution are observed sequen-tially with the objective of choosing the largest. No recall of previ-ous observations is allowed and one observation must be chosen.The case in which N is uniform, Poisson or geometric was studiedby Sakaguchi and Porosinski. Our main concerns here are to ex-amine the case where N has a negative binomial distribution. Anoptimal stopping rule and the probability of best choice is studied.

3 - Meta-Heuristic Approach For Training HiddenMarkov ModelsTarik AL ANI, Yskandar HAMAMAn algorithm for the training of Hidden Markov Models (HMMs)by simulated annealing is presented. This algorithm is based on afinite coding of the solution space based on the optimal trajectoryof the state. The initial and final temperatures are derived automat-ically from theoretical considerations. The cost evaluations of thealgorithm are made independent of the problem size in order to min-imize the computation time. A comparative study between conven-tional algorithms and our algorithm shows that our algorithm givesbetter results and overcome the problem of HMM initialization.

4 - A Sequential Decision Problem on a Partially Ob-servable Markov Process related to an EvaluationToru NakaiWe will consider a sequential decision problem on a Markov pro-cess, and the state of this process is closely related to this outcome.This state can be changed by expending an additional amount, and italso changes according to a Markovian transition rule based on thetotal positivity of order two. The dynamic programming formulationimplies a recursive equation about the expected value obtainable un-der the optimal policy. Finally, we treat this problem on a partiallyobservable Markov process with Bayesian learning after observingsome properties since the state can be changed by decisions.


Railway Network Design

Stream: Railroad ApplicationsInvited sessionChair: Juan A. Mesa

1 - Stochastic Urban Rapid Transit Network Design, Bouza Gemayqzel, Carlos Bouza

The rapid transit network design problem considers at upper levelthe list of potential transit corridors and stations to design the net-work. At lower level the networks are evaluated based on the routeand mode user decisions. The optimization objective is to maximizethe transit demand considering the user’s behaviour and the networkdesign constraints. The stochastic extension considers the demandas a random variable. "A priori’ and "a posteriori’ formulations areconsidered. The model is proved experimenting the formulationsusing Branch and Bound in test networks.

2 - Using Flow Reliability Conditions in Robust RapidTransit NetworkFederico Perea, Antonio J. Lozano, , Juan A. MesaIn this work the problem of designing rapid transit networks is ad-dressed. The robustness of such network is tackled by means ofreducing the effect of possible disruptions. To this aim the modelis constrained to three different kinds of flow conditions. Such newconstraints impose that the demand should be routed through dif-ferent paths so that in case of an arc failure the number of affectedpassengers is bounded from above.

1

MA-05 IFORS 2008 - Sandton

3 - Use of the Weibull distribution to predict rail defectoccurrence ratesChris RothThe Weibull distribution is frequently used for the prediction of therate of occurrence of defects in the steel rails in a railway network.This paper presents the rationale and outline of the method and ap-plication to rail data from rail networks in South Africa.

4 - Customer-oriented Robust Railway Network De-signJuan A. Mesa, Federico PereaRobustness in Network Design is a relatively well known topic insome fields such as Telecommunication Networks. However, rail-way networks present specific features that give rise to differentproblems. Usually these systems are in competition with othermodes of transportation. Railway systems consist of a set of lines.When a disruption happens, passengers must be transferred to a dif-ferent line or transportation mode and travelling times increase. Inthis work an Integer Linear Programming model to design a net-work minimizing the maximum total travelling when an arc fails isproposed.

� MA-05Monday, 8:00 AM - 9:30 AM - Convention Centre: Boardroom1

Lessons from OR Consultancy Practice1Stream: OR and PracticeInvited sessionChair:

1 - Water Resources Planning : a pragmatic approachto optimisation modellingLiz ArchibaldThis session is based on the author’s recent experience with a UKwater company. The project aims to ensure sufficient and appropri-ate investment in water schemes to meet forecast demand under var-ious climate scenarios. Valuable lessons were learned about balanc-ing the required pragmatism to meet regulatory reporting deadlineswith the employment of good OR consultancy practice.

2 - Surviving encounters with the clientSue MerchantThe talk will highlight some successes and failures of the author’spublic sector (Ministry of Defence and Metropolitan Police) careerin OR, pointing up lessons for aspiring OR analysts. The projectsmentioned will include planning the siting of dockside facilities forwarships and the conversion of the entire UK criminal record officelibrary to microfiche in the 1970s.

3 - Lessons from OR Consultancy Practice

This discussion draws insights from the presenter’s 20 year-experience in OR work as part of a corporate internal group andas an external consultant for the government and the private sectors.Topics covered include selling to internal customers, charging forservices, centralization vs decentralization, and position in the or-ganization. Difficulties and successes with government and pri-vate sector projects are discussed. The presentation concludes withlessons learned, including understanding client resistance and keycontributors to a successful OR group.


Decision Analysis

Stream: Decision AnalysisInvited sessionChair:

Chair: Ahti Salo

1 - Robust Risk-based Resource Allocation forCounter Terrorism

A method for allocating funds to counter terrorism measures un-der incomplete information is described. The method uses robustportfolio allocation methods that identify efficient portfolios wheninformation on risk, benefits, and costs is limited. An example of al-location of funds to infrastructure protection measures is presented.

2 - Predicted and actual utility in consumption-payment decisions, Silvia Bellezza

In standard decision analysis, any payment scheme having the samenet present value would be deemed indifferent by the consumer.However, the timing of the payments with respect to the consump-tion can produce different hedonic effects. We propose a variationof the model by Prelec and Loewensein (1998) based on hyperbolicdiscounting and the absolute magnitude effect. We argue that ourmodel is an improvement of theirs, and calculate the optimal pay-ment scheme for non-durable and durable goods. Our model helpsunderstand the hedonic evaluation of general flows of positive andnegative events.

3 - A Portfolio Model for the Allocation of Resourcesto Standardization ActivitiesAhti Salo, Antti Toppila, Juuso Liesiö

High-tech companies seek to strengthen their technological compet-itiveness through standardization activities which contribute to en-hanced interoperability and market growth. We present a model thatwas developed for a major telecommunications company to supportthe allocation of resources to standardization activities. This modelcaptures relationships between market growth and standardizationactivities, links these activities explicitly to key strategies and corecompetencies, and guides the determination of optimal resource lev-els for the entire portfolio of standardization activities.

4 - Portfolio and Scenario Analysis in the Cost-Effectiveness Evaluation of Weapon SystemsJussi Kangaspunta, Ahti Salo, Juuso Liesiö

Cost-effectiveness evaluation of weapon systems is challenging dueto multiple mission objectives, interactions among weapon systemsand the large variety of operating situations where these systemsmay be used. We develop a portfolio-optimization model to iden-tify cost-effective combinations of weapon systems at different bud-get levels and in view of different objectives. The model — whichbuilds on results from combat simulations — also accounts for in-complete information about the importance of mission objectivesand the probabilities of operating situations.


Simulation 1Stream: Simulation (INFORMS Simulation Society)Invited sessionChair: Michael Fu

1 - A New Statistical Procedure for Validation of Simu-lation ModelsRobert Sargent

We describe a statistical procedure for validating simulation modelswhere the amount of model accuracy is specified. This procedureprovides for the model to be accepted if the difference between thesystem and the model is within the specified model accuracy. Datafrom both the system being simulated and the simulation model isrequired for this procedure.

2

IFORS 2008 - Sandton MA-16

2 - Derivative Estimation for Stochastic Activity Net-worksMichael Fu

We consider the problem of estimating derivatives of performancemeasures in stochastic activity networks when Monte Carlo simula-tion is employed. We review various approaches, including infinites-imal perturbation analysis, smoothed perturbation analysis, likeli-hood ratio or score function method, and weak derivatives. We con-sider various performance measures and report on some numericalresults.

� MA-11Monday, 8:00 AM - 9:30 AM - Convention Centre: CommitteeRoom 2

Berth Allocation in Container PortsStream: Container PortsInvited sessionChair: Ek Peng Chew

1 - Berth Allocation and the Quay Crane ResourceFrank Meisel, Christian Bierwirth

In this talk we consider the Berth Allocation Problem and the QuayCrane Assignment Problem in seaport container terminals. They de-termine berthing positions, berthing times, and crane assignmentsfor the calling vessels. The decisions are strongly interrelated. E.g.the crane assignment determines vessel handling times for the berthallocation. To respect the interrelations we combine both problemsin a new mathematical formulation where focus is put on modelingcrane productivity effects. We present two meta-heuristics whichperform well on a set of real world like problem instances.

2 - Berth allocation at multi-berth container terminalsZheng Tong, Jiyin Liu

We study the berth allocation problem (BAP) at multi-berth con-tainer terminals with equipment-dependent vessel processing times.An integer programming model is formulated to simultaneously de-termine the berth and the position for each incoming vessel to min-imize the maximal tardiness. Service priority, nonoverlapping con-straints and physical restrictions such as water depth and equipmentcapability are all considered.

� MA-12Monday, 8:00 AM - 9:30 AM - Convention Centre: CommitteeRoom 4

Finance: Dynamic models

Stream: FinanceInvited sessionChair: Gerd Infanger

1 - The Optimal Timing of Maintenance Investmentsunder UncertaintyThomas Yeung

Billions of dollars are spent every year on maintenance in all facetsof industry including manufacturing, service, and health care. Tra-ditionally maintenance has been treated as a cost that is a necessaryevil, however, we treat maintenance as an investment that can leadto increased revenue and reduced costs in the future. In doing so,we borrow from the finance literature and utilize real options the-ory and contingent claims analysis to determine the optimal timingof maintenance when there is a great deal of uncertainty about thedegradation process and the costs of maintenance.

2 - Stochastic Dynamic Model of Transmission Invest-ments in Radial NetworksEnzo Sauma

In this paper, I formulate a stochastic dynamic model of transmis-sion investments and use this model to compare the behavior of anetwork planner when considering and when ignoring the dynamicnature of transmission investments. I also characterize the structureof an optimal transmission investment policy in the case of a par-ticular two-node network. I illustrate the results with a numericalexample.

� MA-15Monday, 8:00 AM - 9:30 AM - Sandton Sun: Maroela South

DEA Tutorial and Software DemoStream: DEA and Performance ManagementTutorial sessionChair: Ali Emrouznejad

1 - Evaluation of research in efficiency and productiv-ity: Past and futureAli Emrouznejad

Since the original Data Envelopment Analysis there has been rapidand continues growth in the field. As a result, a considerable amountof published research has appeared, with a significant portion fo-cused on DEA applications of efficiency and productivity in bothpublic and private sectors. A comprehensive listing and analysis ofDEA research covering its first 30 years of history is collected. Thispaper thus presents an extensive listing of DEA research coveringtheoretical developments as well as â EUR œreal-worldâ EUR . Fur-ther a standard DEA process for large projects is discussed.

� MA-16Monday, 8:00 AM - 9:30 AM - Sandton Sun: Jacaranda

Production and Inventory Management

Stream: Production and Inventory SystemsInvited sessionChair: Moses Dowart

1 - Two-Commodity Perishable Inventory System withBulk Demand for One CommodityGunaseelan Arivarignan, Balasubramanian Sivakumar,Venkata Yadavalli

This article considers a two-commodity inventory system with threetypes of customers which follow follow a Markovian arrival process(MAP). A modified ordering policy is adopted . The lead time isassumed to have phase type distribution and the demands that occurduring the stock out period are assumed to be lost. The joint prob-ability distribution for both commodities is obtained in the steadystate case. Various measures of system performance and the totalexpected cost rate in the steady state are derived. The results areillustrated with numerical examples.

2 - Optimal order quantity for newsvendor modelswith net utility function that incorporates shortagecostsHonest Chipoyera

A number of criteria are used when an order quantity is recom-mended for a single-period inventory model. This paper exploresa new dimension to the Newsvendor Problem: maximization of anet utility function that incorporates shortage costs in the profit perinventory cycle. The order quantity maximizing expected profits perinventory cycle is almost always greater than that for models whosenet utility function omits shortage costs.

3

MA-19 IFORS 2008 - Sandton

3 - Theoretical Analysis of Improving ProcurementPerformance of Small to Medium Scale Enterprises(SMEs): The Zimbabwean CaseMoses Dowart

This paper is an analysis of the inventory levels in small to mediumorganizations in Zimbabwe to come up with an optimal and cost ef-fective inventory model.

� MA-19Monday, 8:00 AM - 9:30 AM - Sandton Sun: Acacia

Condition Based Maintenance andPrognostics

Stream: Maintenance and ReliabilityInvited sessionChair: Daniel BriandChair: Craig Lawton

1 - Conditional residual time modelling using bothmonitored responsive and reflective variables in oilsamples

This paper presents a model for estimating the residual time of asystem subject to oil based condition monitoring. In particular weassume both responsive and reflective variables are sampled at regu-lar intervals for monitoring the system’s condition. Responsive vari-ables are those which may cause the change of the underlying resid-ual time while reflective variables are those variables which onlyreflect the residual time but cannot change it. The model establishedwas fitted to the real data of diesel engines monitored by an oil anal-ysis program and the results are presented.

2 - Modeling the Application of Prognostics andHealth Management in Large-Scale EnterpriseLevel Logistic SimulationsKimberly Welch

Prognostics is a growing concept in system design and is a require-ment for military systems, such as the JSF F-35 Fighter Jet and theFuture Combat Systems. Algorithms have been developed by San-dia National Laboratories in order to model the effects of prognostics(e.g. autonomic) in large-scale enterprise level reliability and logis-tics simulations. These prediction algorithms allow for an intelligentmodel of the impacts of prognostics on the logistics enterprise of asystem over the life of that system.

3 - PHM Simulation in the Context of a System-of-Systems

Both industry and the military have established Enhanced Diagnos-tic / Enhanced Prognostic capabilities as a mandated area for im-provement. Objective requirements in this area, for both newlydeveloped systems as well as current force systems that are beingmodernized, will be ambitious. This presentation will discuss re-search and development of diagnostic and prognostics simulation inthe context of a system-of-systems such as a brigade.

Monday, 9:45 AM - 11:15 AM

� MB-01Monday, 9:45 AM - 11:15 AM - Convention Centre: Ballroom 1

Official opening of IFORS 2008 TriennialConferenceStream: PlenariesInvited sessionChair: Hans W. Ittmann

1 - The World and South Africa in the 2010s: the latestClem Sunter

Clem Sunter is well known for Scenario Planning in South Africa.Before 1994 he developed two scenarios for South Africa - the "HighRoad" and the "Low Road" scenarios and influenced the events inthe country. The talk at IFORS 2008 examines possibilities for theglobal ’game’ over the next 12 years and South Africa’s optionswithin that framework.

4

IFORS 2008 - Sandton MC-03

Monday, 11:45 AM - 1:15 PM

� MC-01Monday, 11:45 AM - 1:15 PM - Convention Centre: Ballroom 1

Combinatorial Optimization: VehicleRouting 1

Stream: Combinatorial OptimizationInvited sessionChair: Michel Gendreau

1 - A well-scalable metaheuristic for large-scale fleetsize and mix vehicle routing problem with time win-dowsOlli Bräysy, Pasi Porkka, Wout Dullaert

In this presentation we consider the fleet size and mix vehicle rout-ing problem with time windows. A new metaheuristic that combinesideas from threshold accepting and guided local search with four lo-cal search neighbourhoods is presented. Several new search limita-tion strategies are also discussed. Computational results are demon-strated with the 168 benchmark problems of Liu and Shen and 600new large-scale benchmark problems. Comparison with the previ-ous state-of-art methods shows that the new procedure is efficientand competitive, reporting several new best-known solutions.

2 - A tabu search for an inventory routing problemM. Grazia Speranza, , Luca Bertazzi, Alain Hertz

A product is shipped by capacitated vehicles from a supplier to sev-eral retailers, where it is depleted in a deterministic way over a giventime horizon. Each retailer can be visited several times. Shipmentsare performed at discrete time instants. The problem is to determinea shipping policy minimizing the sum of transportation and inven-tory costs both at the supplier and at the retailers in such a way thatno stock-out occurs. We present a sophisticated tabu search heuris-tic that is very effective on all instances where an exact solution isavailable.

3 - Dependence in scheduling problemsPatrick Schittekat

A sub-problem in logistic planning is determining the optimal timeschedule of each resource in order to maximize its utilization. Often,optimal schedules for each resource are determined independently.In practice, however, this may not be sufficient, e.g. VRP problemswhere different drivers have to arrive simultaneously at a customer.

4 - A Hybrid Monte Carlo Local Branching Algo-rithm for the Single Vehicle Routing Problem withStochastic DemandsMichel Gendreau, , Patrick Soriano

We present a new algorithm that uses both local branching andMonte Carlo sampling in a multi-descent search strategy for solv-ing 0-1 integer stochastic programming problems. This procedureis applied to the single vehicle routing problem with stochastic de-mands. Computational results demonstrate the effectiveness of thisnew approach for solving hard instances of the problem.


Optimization: Gradients and Constraints

Stream: OptimizationContributed sessionChair: Daniel Wilke

1 - Gradient algorithms with fast asymptotic rates forquadratic optimizationAnatoly Zhigljavsky

We consider a general class of gradient algorithms for quadratic opti-mization and show that these algorithms can be equivalently writtenas multiplicative procedures of constructing optimal designs on thereal line. We demonstrate and discuss some theoretical and numeri-cal results concerning the convergence rates of selected algorithms.We only study the rules for choosing the step-lengths under the as-sumption that the dimension of the problem is very large. We showhow to get better asymptotic rates than the rates of the celebratedBarzilai-Borwein algorithm.

2 - Predictor-corrector method applied to the ACactive-reactive optimal power flow problem usingnonlinear corrections in all optimality conditions, Aurelio Oliveira, Secundino Soares,

A variation of the predictor-corrector interior point method is devel-oped to the AC active and reactive optimal power flow problem. Thenonlinear corrections are developed not only for the complementaryconditions, however also for the primal and dual feasibility ones.This is possible due to use of Cartesian coordinates.

3 - Fractional Objective Function and Sensitivity Anal-ysis for Different Solving AlgorithmsMarija Cileg, Tibor Kis

Fractional objective function is appropriate for presentation of eco-nomic effectiveness for several reasons (the quality of parameters:relative value of linear relations and the chance of including con-stant values). There are several solving algorithms for optimizationmodels with fractional objective function. In this paper we analyzesensitivity analysis concerning constraints and the objective func-tion for different solving algorithms, depending on changes of inputdata. An application of the procedure onto an actual problem will bepresented.

4 - Gradient only optimization for discontinuous func-tionsDaniel Wilke,

We present a gradient only optimization algorithm that is capableof ignoring discontinuities in the optimization problem. Such an al-gorithm is applicable to optimization problems based on numericalsimulation, that in some instances introduces numerical discontinu-ities. We apply the method to an optimal control problem.


Robust and On-line Problems inRailways

Stream: Railroad ApplicationsInvited sessionChair: Paolo Toth

1 - A mathematical approach for building robust rail-way rolling stock circulationsDavid De Almeida, Christian Weber

Railway transportation is usually operated under a precise schedulethat takes into account several constraints. However, this scheduleoften needs to be modified because of practical disturbances. In thiscontext, we will present a mathematical approach for building robustrolling stock circulations. We consider the so-called "absorption ro-bustness’ and study how to improve it by minimizing the total prop-agated delay. Our approach relies on a multi-commodity networkflow model previously developed at the R&D department of SNCFto build engines circulations.

5

MC-04 IFORS 2008 - Sandton

2 - Approximation of generic Locomotive AssignmentProblemssadki jalila, Laurent Alfandari, Anass Nagih, Agnès PlateauThe locomotive assignment problem is known to be NP-hard. Thiscomplexity is confirmed by experimental results on real-case datausing the column generation method, where the master problem ismodelled as a Covering Integer Programming problem. We givea reformulation of the master problem as a Set Covering problemwhich can be solved by a greedy heuristic achieving a logarithmicapproximation ratio.

3 - Evaluation of green-wave policy in real time railwaytraffic managementDario Pacciarelli, Francesco Corman, Marco Pranzo,In this work, we analyze the effectiveness of a new railway trafficcontrol policy called green wave, which consists in letting trainswaiting at the stations, thus avoiding speed profile modificationsin open corridors. We use an optimization algorithm to resched-ule trains in case of perturbation, and compare the delays obtainedwhen trains are allowed or not to change their speed profile in opencorridors. We report on our computational study for two practicaldispatching areas of the Dutch railway network.

4 - On the Online Track Assignment ProblemGabriele Di Stefano, Marc Demange,We consider a track assignment problem in a train depot where eachtrack can store more than one train. The problem is to assign a trackto each train as soon as it arrives and such that it can leave the de-pot on time without being blocked by any other train. We modelthis problem as an online bounded coloring problem on permuta-tion graphs or on overlap graphs. For permutation graphs we studythe competitiveness of a First Fit-based algorithm and we show itmatches the competitive ratio of the problem. For overlap graphs,even the unbounded case does not admit a constant competitive ra-tio.


Applications of OR in Mining

Stream: OR in Agriculture and Natural ResourcesInvited sessionChair: Alexandra Newman

1 - Navigable Decline NetworksDoreen ThomasA major infrastructure component required to develop an under-ground mine is a decline, which is a system of tunnels used foraccess and haulage. A key design consideration is that the declinemust be navigable to trucks and mining equipment, hence must sat-isfy a gradient and turning circle constraint. I describe the DeclineOptimisation Tool (DOT) that designs a minimum cost network ofdeclines which avoids barriers and satisfies given gradient and cur-vature constraints. DOT finds near optimal locations for junctionswhere one decline breaks out from another.

2 - Long Term Optimization of the Production Chain inOpen-pit and Underground Copper MinesAndrés Weintraub, Felipe CaroBased on our experience in modeling Chilean open pit and under-ground copper mines, we discuss the importance of modeling in anintegrated way the production chain from the initial extraction to thedelivery of final products. In particular, we show the impact of amodel used to optimize long term mining plans that has been imple-mented at Codelco, Chile.

3 - Open Pit Mine SequencingAlexandra Newman, Kevin WoodWe describe mathematical techniques to increase the tractability ofthe open pit mine sequencing problem, which, generally speaking,consists of maximizing net present value subject to block sequencingconstraints and operational constraints such as minimum and max-imum production and processing capacities. We discuss how thesetechniques might help address a stochastic version of our model inwhich ore grades are variable.

4 - Challenges in the Practical Implementation of Plan-ning Models in the Mining IndustryPatrick Veldhuizen

This is a case study of the obstacles encountered with the practicalimplementation of a large complex planning and simulation modelin a coal value chain operation. Items such as model robustness,input data integrity and super-user capabilities will be discussed.

� MC-05Monday, 11:45 AM - 1:15 PM - Convention Centre: Boardroom1

Lessons from OR Consultancy Practice2Stream: OR and PracticeInvited sessionChair:

1 - Intellectual Property and Operations ResearchAshok K Mittal

Operations research mainly deals with development of algorithmsand method to solve business related problems.In recent years thecoverage of patents, in some of the patent offices have been extendedto cover methods of doing business as well computer software. Thispaper deals with the issues related with copyright and patents whichare likely to effect operation research practice and profession. Wealso discuss how to effectively search and map this information toprovide directions of movement of a Technology.

2 - The OR toolbox used within Climate Change Adap-tationKarin Mossberg

Within Military OR, a number of analytical tools are used, such asscenarios, simulations and decision analysis. I present experiencesfrom applying this toolbox in another context, adaptation to climatechange within Swedish municipalities. Similarities and differencesare described between the settings when developing an applicationspecific, user oriented toolbox.

3 - Some Key Factors for Successful OR Consultancy

Research into success factors for internal OR Groups will be as-sessed against previous consultancy experience and the needs of cur-rent practice. With reference to case studies, some important lessonswill be offered for negotiating projects, communicating with clientsand delivering implementable solutions.


OR-in-Development 1

Stream: OR-in-Development Prize CompetitionInvited sessionChair: L. Paul Fatti

1 - Introduction to the OR in Development Prize Com-petitionL. Paul Fatti

A brief introduction to the prize and its history will be given, thisyear’s submissions will be reviewed and the method by which theshortlisted papers were selected will be described.

6


2 - Forecasting demand using a model based on theapplicationDavid Fernando Muñoz, Omar Romero, Jose E.Detta-Silveira, David G. Munoz

We present the development of a Bayesian model to forecast thedemand for educational material in the Mexican Institute for AdultEducation. Our model incorporates particular characteristics of thecase under study and provides not only a better interpretation butalso a more precise forecast than traditional methods.

3 - Robust multi-objective optimization for waterdistribution system design using a meta-meta-heuristicJan van Vuuren, Darian Raad, Alexander Sinske

The design of water distribution systems (WDS) is a challengingproblem involving numerous conflicting objectives. A multiplemeta-heuristic framework (AMALGAM) is applied to find the op-timal trade-off between WDS cost and reliability. Superior perfor-mance is achieved versus several competing algorithms and largesavings demonstrated for a South African case study


Facilitated Multi-Criteria DecisionAnalysis

Stream: Decision AnalysisInvited sessionChair: Gilberto Montibeller

1 - The importance of problem structuring for MCDA:avoiding lost CAUSEsValerie Belton

It is well recognised that good problem structuring is an essen-tial component of effective MCDA practice and the key consider-ations in problem structuring for MCDA are captured in the CAUSEmnemonic. We draw on case study experiences over a number ofyears, using a simple multi-attribute value function to support deci-sion making in a wide range of organisations, to highlight the poten-tial consequences of overlooking elements of CAUSE and to illus-trate the potential benefits of MCDA as problem structuring.

2 - Applied Value-Focused ThinkingRalph Keeney

The purpose of making any decision is to achieve something. Value-focused thinking specifies that something in terms of detailed ob-jectives structures that are useful for creating alternatives, recogniz-ing decisions opportunities, and evaluating alternatives. This pa-per describes the relevance and usefulness of the key concepts ofvalue-focused-thinking that are often missing in other approaches.It also discusses the opportunities and challenges in applying theseconcepts. Several cases illustrate the application of value-focusedthinking to public, business, and personal decisions.

3 - Facilitation in Problem Structuring Methods:lessons for MCDA?Jonathan Rosenhead

The hard modelling approaches of operational research rarely if everuse facilitation. By contrast Problem Structuring Methods are rarelyused in any other format. Between them lie approaches, not onlyMCDA but also System Dynamics, which can be and are used inboth modes. This paper will explore the structural reasons under-lying these differences, and some of the specificities of the use offacilitation in PSMs, for the light they may shed on MCDA practice.

4 - Action-Researching MCDAGilberto Montibeller

Action-Research is a research strategy that permits the systematicinvestigation of an issue while intervening in an organisation. Thisresearch strategy may be useful for MCDA researchers who want totest MCDA methods in real-world problems, and for MCDA con-sultants who are keen in reflecting upon and improving their prac-tice. This paper reviews the literature on Action-Research, with aparticular emphasis on those Action-Research designs which seemsuitable for MCDA interventions. It also reflects on my experiencein conducting Action- Research studies for real-world MCDA inter-ventions.


Discrete Event SimulationStream: Simulation (INFORMS Simulation Society)Tutorial sessionChair: Thomas J. Schriber

1 - Inside Discrete-Event Simulation Software: How ItWorks and Why It MattersThomas J. Schriber, Daniel T. Brunner

This paper provides simulation practitioners with a grounding inhow discrete-event simulation software works. Topics includediscrete-event systems; entities, resources, control elements and op-erations; simulation runs; entity states; entity lists; and entity-listmanagement. The implementation of these generic ideas in Auto-Mod, SLX, and Extend is described. The paper concludes with sev-eral examples of "why it matters’ for modelers to know how theirsimulation software works, including coverage of SIMAN (Arena),ProModel, and GPSS/H as well as the other three tools.

� MC-11Monday, 11:45 AM - 1:15 PM - Convention Centre: CommitteeRoom 2

Container Port Design and Optimization

Stream: Container PortsInvited sessionChair: Akio Imai

1 - A Simulation Optimization Framework for Con-tainer Port OperationEk Peng Chew

In this talk, we will present a simulation optimization framework totackle the layout design problems for container port by consideringthe actual port operation. First a simulation model which representsthe basic port operation is created, and then through the optimiza-tion framework, we will iteratively generate and improve the layoutdesign.

2 - Integrating a design optimizer in a container portsimulatorSzu Hui Ng, David Lee

Container port simulation is needed to accurately process all of thedifferent factors that may be in play in a port on a given day. It isessential for the design and evaluation of possible port design andlayouts. Given many design factors and hence many possible layoutdesigns, a study is conducted on various optimization algorithms ona newly developed detailed container port design simulator. Thisincludes design search algorithms and allocation procedures for afixed computing budget. We further discuss integration issues ofthese techniques into the simulator.

7


3 - Uncertainty in container terminal operationsPanagiotis Angeloudis, Michael Bell

An accurate visual microsimulation model has been developed andused to analyse the interaction of various equipment units commonlyfound in automated container terminals. We attempt to quantify thefactors that contribute to operational uncertainty, and discuss theirconsequences for algorithms that optimise horizontal and verticalcontainer movements in such terminals

4 - Port container terminal design perspectiveand relevant operational issues for the mega-containsership eraAkio Imai

In the context where the market has already deployed mega-containerships with a carrying capacity of more than 10,000TEUs,container liner have been choosing more limited ports to call. Asa result, the selected hubs likely have more container traffic to behandled. In this paper, some port container terminal designs aredemonstrated and compared each other in terms of various evalu-ation factors for a short turnaround of calling ships. Also, efficientberth allocation strategies to enhance the productivity of those ter-minals are presented.


Finance: Optimisation models

Stream: FinanceInvited sessionChair: J. E. Beasley

1 - Using Mathematical Programming to optimise re-covery placementsMargarete Bester

In this presentation we will be focussing on the application of amathematical model in the recovery environment, used to maximisethe recovered amount by optimising the placements of accounts inrecovery.

Fair Isaac developed a product called the Placement Plus with amodule called Placement optimiser that use Mathematical Program-ming to optimise the placement of accounts with different agencies.The mathematical programming model takes into account, capacityof the different agencies, budget for legal fees and fixed distributionpercentages.

2 - From Scenarios to Lattices: Flexibility in PortfolioOptimisation,

Managers often model the uncertainty on technology investmentsthrough alternative cash-flow scenarios. Our approach transformsthe cash-flow scenarios into trinomial lattices, which enable the useof general cash-flow distributions as well as the valuation of vari-ous real-options (typically abandonment or additional investment).The optimal strategy is solved with the dynamic programming algo-rithm. The model is extendable for instance to the optimisation ofseveral technology portfolios with competing technologies. A realpatent-portfolio valuation case demonstrates the approach.

3 - Multinational tax planningJ. E. Beasley

Multinational tax planning involves designing a corporate structure(across different countries) so as to remit profits from a number ofsubsidiaries to a single parent company, whilst minimising the taxpaid (maximise the amount received at the parent company).

We present a nonlinear mixed-integer formulation that is linearisableas well as a tabu search heuristic. Computational results are given.


Finance: Applications

Stream: FinanceInvited sessionChair:

1 - South African Banking - Striving Toward an OpenMarket EnvironmentDave Masterson

Commercial Banking in SA followed British Colonial banking Mod-els but was a tightly-controlled cartel that restricted open competi-tion. OR played key role in breaking the cartel and assisted com-petitive market development. Early models employed to gain sup-port of banking community removed pricing restrictions in 1980’s.The pricing model is again in need of review and in addition, banksare faced with developing secondary markets for affordable-housingsector of the mortgage market. This problem is not dissimilar to thatencountered by other African countries.

2 - Assessing Impact of Liberalization on the Effi-ciency of Indian Public InsurersSubir Sen

For a long time the Indian Insurance Industry was insulated fromprivate participation and hence devoid of any competition. Themonopoly of LIC, a life insurer and four non life insurers was endedfollowing liberalisation of the industry in 1999. Although the entryof private sector companies seized a small market share of the pub-lic insurers in both the segments, yet public firms are supposed to bemore efficienct. The present study is an attempt to analyze the effi-ciency of public insurance firms during the period 1975 to 2006-07using DEA and identify factors affecting their efficiency.

3 - An Econometric Study of Currency Crises in Devel-oping Economies: The Zimbabwean CaseCaston Sigauke

In this paper we investigate the causes of currency crises in emerg-ing markets. We estimate the probability of a currency crisis using alimited dependent variable model. Past banking crises are significantdeterminants of the probability of a currency crisis.

� MC-14Monday, 11:45 AM - 1:15 PM - Sandton Sun: Maroela North

DEA: Applications 1

Stream: DEA and Performance ManagementInvited sessionChair:

1 - Efficiency of Indian Marine Fishery using DEAAjit Kumar Ray, Neelangshu Ghosh

Indian marine fishery exploited only 69% of potential even after sig-nificant enhancement of fishing fleet during last decade. Hence per-formance measurement of India’s maritime states (as DMU’s) is re-quired to find the scope of improvement. Their causes of inefficien-cies are identified by applying DEA, BCC-O model with respect tokey contributing resources which come under three broad categories-geographical, human and fishing fleet & gears and fish production asoutputs. Necessary conditions of independence of inputs are verifiedby stepwise regression.

2 - Managerial and Economic Efficiency of Date Pro-cessing Industry in Saudi Arabiakhalid alrwis

8


Measurement of economic and productive efficiencies is consideredone of the most important planning methods that can be used to de-termine the efficient amount of resource usage, thus reducing costsand increasing competitiveness ability of production units in localand global markets. The study concentrates on productive and eco-nomic efficiency. The objective is to measure performance, TE, AEand EE efficiencies of date processing plants. Also, significance dif-ferences in average TE between large, medium, and small firms willbe estimated. DEA & SFP Function Theories will be applied .

3 - Data Envelopment Analysis and OrganizationalPerformance

This study uses Data Envelopment Analysis for measuring organiza-tional performance which reconciles some financial and nonfinancialmeasures in a single index. Model results show that only about 30%companies were operating on the best-practice frontier. Finally, thisstudy offers an alternative perspective and characterization on theorganizational performance with relation to the Balanced Scorecardframework. This alternative is based on Data Envelopment Analysiswhich truly measures organizational performance to improve effec-tiveness and efficiency in any company.

� MC-15Monday, 11:45 AM - 1:15 PM - Sandton Sun: Maroela South

DEA Theory: New developments 1

Stream: DEA and Performance ManagementInvited sessionChair:

1 - On Data envelopment analysis of both real andinteger-valued inputs and outputsReza Kazemi MatinConventional Data Envelopment Analysis (DEA) models assumereal-valued inputs and outputs. In many occasions, some inputsand/or outputs can only take integer values. This paper deals withDEA in the case of both real and integer-valued data from axiomaticfoundation and also computational points of view. An empirical ap-plication is used to illustrate the model.

2 - Ranking of Units on the Overall Profit Efficiency inDEA with Fuzzy DataMohsen Rostamy-MalkhalifehData Envelopment Analysis (DEA) is the nonparametric method ofmeasuring the efficiency of a Decision-Making Unit (DMU) such asa firm or a publicsector agency. This paper describes a new DEAranking approach that is according to overall profit efficiency. Thisranking is also used for DMUs with fuzzy data. The proposed DEAranking approach defines DMUâ EUR TMs ranking and their effi-ciency or inefficiency contemporaneously that is an advantage. fi-nally, the method is illustrated by solving numerical example.

3 - Sensitivity analysis of the proportionate change ofa subset of outputs or/and inputs in DEA, Juergen Guddat, Oliver Stein

Sensitivity analysis in data envelopment analysis (DEA) is studiedfor the case of the proportionate change of a subset of outputs or/andof a subset of inputs of an efficient decision making unit (DMU)according to the Charnes-Cooper-Rhodes (CCR) ratio model. Suffi-cient conditions for an efficient DMU to preserve its efficiency underthe proportionate change of a subset of outputs or/and of a subset ofinputs are obtained. An illustrative numerical example is provided.

� MC-16Monday, 11:45 AM - 1:15 PM - Sandton Sun: Jacaranda

Production LinesStream: Production and Inventory SystemsInvited sessionChair: Yun Fong Lim

1 - Lean Manufacturing Implemented at a Fish Can-ning Plant, Henk Hunter

Lean Manufacturing is a methodology for processing products orservices as fast, cheap and efficient from order to customer as vi-able. The principles of Lean Manufacturing were applied to a fishcanning plant, the largest of its kind in the Southern Hemisphere,resulting in the reduction of labour cost by 40 percent and decreas-ing downtime by 20 percent. The holistic effect was a reduction inmanufacturing cost of 35 percent. The scope of the project, analysisdone and the solutions implemented are presented.

2 - Optimizing U-Lines: Use of Single-Pass HeuristicsChun-Hung Cheng

In JIT implementation, firms use U-shaped production lines to re-place straight-lines. The U-line problem is also NP-hard. In order tomake the U-line model applicable, we have to consider heuristic pro-cedures. Talbot et al. [14] review and evaluate various single-passheuristic decision rules for the straight-line problem. In this work,we would like to examine the extensibility of straight-line heuris-tics to U-lines. Further, we would like to evaluate the performanceof these heuristic decision rules under different problem characteris-tics.

3 - Assignment of cross-trained workers to tasks on aproduction lineLieschen Venter, Stephan Visagie

The assignment of permanent and temporary cross-trained workersto tasks is presented. A multi-level assignment model is formulatedwhich maximizes a production utility function and the improvementof workers’ skill levels, while determining an optimal sequence oftasks. A real life case study is presented to illustrate the model.

4 - Reducing Labor Requirement in Bucket BrigadesYun Fong Lim

Workers in a bucket brigade production line perform unproductivetravel when they walk to get new work from their colleagues. Ex-cessive unproductive travel could hurt the throughput of the line sig-nificantly. We propose a new way to configure a bucket brigade pro-duction line so that this unproductive travel is reduced and so lesslabor is required to attain the same level of throughput.

� MC-17Monday, 11:45 AM - 1:15 PM - Sandton Sun: Oleander

OR Software Applications

Stream: OR/MS SoftwareContributed sessionChair: Bjarni Kristjansson

1 - New Stochastic Programming features in the MPLModeling LanguageBjarni Kristjansson

Formulating stochastic models in a modeling language has long beena challenge in practice. In this presentation we will demonstratenew features in MPL that allow stochastic models to be effectivelyformulated and solved. Several formulations of SP models will bedemonstrated, including both independent variables and scenario-based models.

We will also show examples of SP models in OptML, an XML basedlanguage. OptML is designed to represent problem instances of stan-dard LP/MIP/QP/NLP/SP models and is intended to be an alternativeto the well-known MPS and SMPS problem instance formats.

9


2 - Expected Shortfall Portfolio Optimization IncludingTransaction Costs (Using SAS/OR)Machiel KrugerA very important part of portfolio construction is the generation ofscenarios for future asset prices. We fit GARCH models to shareprice data and use the fitted models to generate future share pricescenarios. The implementation and solution of the portfolio opti-mization problem are done using PROC OPTMODEL.

3 - Walking the Expression Tree for Interfacing Model-ing Languages to Nonlinear (or Not-Linear) SolversRobert Fourer, David M. GayFor most of their existence, optimization modeling languages haveworked with the same kinds of algebraic expressions and the samekinds of solvers. Lately the pace of change has accelerated, however,as languages have been extended to keep up with solvers for globaloptimization, conic programming, optimization under uncertainty,and various hybrids with discrete optimization problems. This pre-sentation will review language extensions for such purposes, usingmainly the AMPL language for illustrations.

� MC-18Monday, 11:45 AM - 1:15 PM - Sandton Sun: Syringa

Graphs and Networks 1

Stream: Graphs and NetworksContributed sessionChair: Luís Vieira

1 - The Euclidean Steiner ratio

In 1992 Du and Hwang published a paper confirming the correctnessof a well known 1968 conjecture of Gilbert and Pollak, regarding theEuclidean Steiner ratio for n points in the plane. We take a criticallook at this paper and provide counterexamples to expose a funda-mental mistake. We conclude by applying their strategy to prove thevalidity of the Gilbert-Pollak conjecture for 6 points.

2 - On Hamiltonian Properties of Locally ConnectedGraphs with Bounded Vertex DegreeYury Orlovich, , Chris Potts, Vitaly StrusevichThe existence of Hamilton cycles for locally connected graphs withbounded vertex degree is considered. We show that every connected,locally connected graph with maximum vertex degree equal to 5 andminimum vertex degree greater or equal to 3 is fully cycle extend-able. We prove that the HAMILTON CYCLE problem for locallyconnected graphs with maximum vertex degree less or equal to 7 isNP-complete.

3 - New results on an operation of closure in claw freegraphsBenmeziane ZinebThe concept of closure for claw free graphs has been defined byZ.Ryjacek in 1997. The closure cl(G) of a claw free graph is ob-tained by recursively completing the neighborhood of any locallyconnected vertex of G, as long as this is possible. The closure cl(G)is well defined, remains a claw free graph and both graphs G andcl(G) have the same circumference. In this article, we proove thatunder certain conditions, the graph cl(G) is a complete graph.

4 - Euclidean Jordan algebras and necessary condi-tions for the existence of a strongly regular graphLuís VieiraLet G be a (n,p;a,c) strongly regular graph, A the adjacency matrixand let V be the Euclidean space spanned by the powers of A overthe reals where the inner product is the usual trace of matrices. Vis an Euclidean Jordan algebra when one introduces in V the usualproduct of matrices. One defines the power of exponent x of themodulus of A. Working inside the Euclidean Jordan algebra V withthe Hadamard product of matrices one establishes necessary condi-tions of existence over the parameters and the spectra of the stronglyregular graph G.

� MC-19Monday, 11:45 AM - 1:15 PM - Sandton Sun: Acacia

Reliability

Stream: Maintenance and ReliabilityInvited sessionChair: Freddy K. Nailana

1 - Reliability Equivalence of a Series SystemSanjay Chaudhary

The concept of reliability equivalence is discussed and extendedfrom simple series system to a complex system with n independentand identical components. The lifetime of each component is as-sumed to be exponentially distributed random variable. The reduc-tion method, hot duplication method and warm duplication methodare used to improve the reliability of the system. MTTF of the sys-tem is measured for the different methods and compared.

2 - Software Reliability ModellingFreddy K. Nailana

Software failure data collection is usually carried out by, recordingtimes between succesive failures, or by specifying testing intervalsand recording the number of failures during each interval (groupeddata). The collected data can be synthesized to estimate the SoftwareReliability Growth Models (SRGMs) parameters using techniquessuch as maximum likelihood method and the least square method.It is asumed that the number of failures follow a non-homgeneousPoisson process. In this paper, the validity of the model will be doneusing various system measures

10

IFORS 2008 - Sandton MD-07

Monday, 2:00 PM - 3:30 PM

� MD-01Monday, 2:00 PM - 3:30 PM - Convention Centre: Ballroom 1

Combinatorial Optimization: VehicleRouting 2

Stream: Combinatorial OptimizationInvited sessionChair: Michel Gendreau

1 - A Branch-and-Price Approach for Pick-up and De-livery with TransfersPablo A. Rey, Cristián Cortés, Michel GendreauThe Pick-up and Delivery Problem is routinely encountered in a va-riety of application contexts, including local express mail or parceldelivery and dial-a-ride transportation. Recently, a variant of thisproblem in which the persons or objects being transported can betransfered from one vehicle to another, was introduced and solvedusing Benders decomposition.In this talk, we explore an alternative formulation of this new prob-lem based on the generic set covering model for routing problems.We propose to solve this alternative formulation using column gen-eration and branch-and-price.

2 - Resources Requirement and Routing in CourierServiceKa Yuk Carrie LinA pickup and delivery problem with time windows (PDPTW) andnon-identical delivery resources is studied with the objective of min-imizing total cost. Special characteristics considered include thatone resource (e.g. van) can transport both customer items and anon-identical resource (e.g. courier); and that item transfer is al-lowed between resource units.

3 - An Exact Solution Approach for the Prize Collect-ingMohamed Haouari, Hanif SheraliWe address the Prize Collecting Steiner Tree Problem (PCSTP) on agraph, which is a generalization of the well-known Steiner tree prob-lem. We present 0-1 programming formulations and use them to de-velop preprocessing procedures for reducing the graph size. Also,we design an exact branch-and-cut approach. Our computationalstudy reveals that the proposed approach can solve relatively large-scale PCSTP instances having up to 1000 nodes to optimality with areasonable effort.

4 - Hybrid Predictive Control for the Dynamic Pick-up and Delivery Problem with Variable Fleet SizeBased on an Evolutionary Multi-objective Optimiza-tion ApproachCristián Cortés, Doris Sáez, Alfredo Núñez, MichelGendreauWe present a hybrid adaptive predictive control approach for the dy-namic pickup and delivery problem that relies on an EvolutionaryMulti-objective Optimization (EMO) algorithm for balancing theopposite interests of users and operators included in the objectivefunction. Operators are allowed to rent additional vehicles to solveunexpected demand problems. User costs are related to the wait-ing and travel time of passengers. EMO is used to find the dynamicPareto front, and make the real dispatch decision based on differentoptimization criteria. Preliminary simulation results are reported.

� MD-03Monday, 2:00 PM - 3:30 PM - Convention Centre: Ballroom 3

Optimisation and Railroad SchedulingProblemsStream: Railroad ApplicationsInvited sessionChair: David Ryan

1 - A Set Packing Model For Train RoutingRichard Lusby, Jesper Larsen, David Ryan, MatthiasEhrgott

We present a set packing approach to solve the problem of routingtrains through railway junctions. The proposed model is fundamen-tally different to conventional node packing approaches in that a re-source based constraint system is adopted. We demonstrate that themodel has an improved LP relaxation and can be applied to differ-ent (strategic, tactical, and operational) variants of the problem. Abranch-and-price solution method is adopted and computational re-sults for a European Train Operator are presented.

2 - A Decomposed Model for the Rolling Stock Recov-ery ProblemJulie Jespersen Groth, Jesper Larsen,

We address the problem of real time recovery in a railway rollingstock plan. We decompose the problem and first determine typesand number of units assigned to each train task and secondly routesthe physical train units incorporating maintenance reguirements. Wepresent the MIP models of the decomposition and computational re-sults.

3 - The Train Driver Recovery Problem - a Set Partition-ing Model and Solution MethodDavid Ryan,

Using data from a Danish passenger railway operator, a solutionmethod to the Train Driver Recovery Problem (TDRP) is developed.The TDRP is formed from a set of duties and a set of train tasks af-fected by the disruption within a certain recovery period. The LP re-laxation of TDRP is solved with a column generation method basedon a limited subsequence strategy for generating recovery duties. Ifa feasible solution is not found, other drivers are added or the recov-ery time period is increased. Fractions are resolved using constraintbranching. Solutions are found within a few seconds.

� MD-07Monday, 2:00 PM - 3:30 PM - Convention Centre: Boardroom 3

OR-in-Development 2


1 - A Genetic Algorithm Metaheuristic for a Multi-stageHostel Space Allocation ProblemAderemi Adewumi, Ojo Ayeni, Ebunoluwa Fasina, MontazAli

We present a genetic algorithm metaheuristic to solve the NP-hardhostel space allocation problem under domain specific hard and softconstraints. The study is based on a real-life case obtained fromone of the largest universities in Nigeria. Results obtained show theapplicability of metaheuristic to real-life optimization problem

2 - Research on measuring and controlling fishing ca-pacity for Chinese inshore fleets by DEA methodZheng Yi, Zhou Ying-qi

The fishing capacity and capacity utilization for Chinese inshorefleets in the latest 13 years were measured by DEA method. Thefunction of collecting tax for controlling fishing capacity was sim-ulated quantitatively, which indicated that the tax system would beeffective if the tax rate is not too low.

11

MD-08 IFORS 2008 - Sandton

3 - Farming system performance and water use effi-ciency in the Tunisian semi-arid region: Data En-velopment Analysis ApproachFraj ChemakThe objective of this investigation is to analyze the performance ofirrigated farms based on a survey data carried out in the region ofSidi Bouzid. Various irrigation modes are considered and comparedin farms using collective and private resources. Empirical findingsrevealed a technical inefficiency use of inputs and more importantwater overconsumption under private than under collective watermanagement system. However the profitability of the activity, basedon the use of wells for irrigation, appears to be much higher thanthat based on the collective mode of irrigation.


Multiobjective Optimization Methods

Stream: Decision AnalysisInvited sessionChair: José Rui Figueira

1 - Primal and Dual Approximation Algorithms for Mul-tiobjective Linear ProgrammingMatthias Ehrgott, Yufen ShaoWe present two algorithms to approximate the nondominated set ofa multiobjective linear programme. The primal algorithm is basedon Benson’s algorithm, the dual algorithm uses geometric dualityof multiobjective linear programmes and a dual variant of Benson’salgorithm. We prove that both find a set of epsilon-nondominatedpoints and give numerical results.

2 - Solving the bi-objective integer minimum cost flowproblemAndrea Raith, Matthias EhrgottWe present a two phase algorithm to find a complete set of efficientsolutions of the bi-objective integer minimum cost flow problem.The algorithm uses the network simplex algorithm in phase 1 and analgorithm to rank k-best flows in phase 2. Numerical results on testnetworks are given.

3 - Multiobjective Educational Timetabling: Modellingand Recent Developments

Educational timetabling problems are difficult to solve due to theirNP hard structures. Besides 0-1 variables make them more complex.Exam and course scheduling, invigilator assignment are examplesin this context. Multiobjective mathematical models of some realcases are introduced. Some methods to prioritize different objec-tives and/or criteria of the problems are discussed. Besides, recentdevelopments on educational timetabling problems, modelling is-sues, considering preferences are discussed as ongoing studies andmotivations for further researches.

4 - Branch and bound algorithm for the 0-1 knapsackproblem with multiple objectives, Xavier Gandibleux

Our talk concerns a branch and bound algorithm for the multi-objective 0-1 knapsack problem. We discuss different orders inwhich the items are selected by the branching procedure, all basedon the so-called efficiency of each item. Then, we present severalapproaches for the bounding part of the algorithm.


Simulation 2Stream: SimulationContributed sessionChair: Lea M. Wakolbinger

1 - A bridge between simulation theory and practice ina process industry environmentLynette Kotze

A simulation study was done of a catalyst processing loop. It provedto be difficult to apply simulation theory regarding. This paperpresents solutions to overcome these limitations and suggests waysof making feasible recommendations when there is no straight for-ward answer.

2 - Arena in an Integrated Growth Program at SasolAnette Van der Merwe

Sasol is embarking on a growth program integrating new and exist-ing technologies into its current facilities. The impact on the inte-grated refinery and propylene value chain operations are tested usingan Arena model to determine infrastructure and capacity constraints.Interactions between units and across value chains are taken intoconsideration.

3 - An agent-based computer simulation of new prod-uct diffusionLea M. Wakolbinger, Christian Stummer, Markus Günther

Determining effective marketing measures that further the diffusionof a new product is a challenge. We present an agent-based computersimulation that supports managers in deciding which measures to ap-ply when introducing a new product on the market. The applicationof our tool is illustrated through a real-world scenario.

� MD-11Monday, 2:00 PM - 3:30 PM - Convention Centre: CommitteeRoom 2

Resource Scheduling and Dispatching inContainer PortsStream: Container PortsInvited sessionChair: David Lee

1 - Strategic terminal allocation and time schedulingof cyclically arriving container vesselsMaarten P.M. Hendriks, Jan Tijmen Udding

We present an MILP formulation to construct a strategic, cyclic ter-minal allocation and time scheduling for a set of container vessels,which are processed by a multi-terminal port operator. Our casestudy is about balancing the workload over time, minimizing inter-terminal transport and minimizing deviations from the existing timeschedule.

2 - Scheduling of Quay Cranes in Container TerminalsChristian Bierwirth, Frank Meisel

We consider the scheduling of quay cranes used in container ter-minals. A new set of constraints is presented to avoid the craneinterference problem. A Branch-and-Bound based heuristic is pro-posed to search a subset of above average quality schedules. For awidely accepted benchmark suite, the new heuristic produces muchbetter solutions in considerably shorter run times than other algo-rithms known from the literature.

3 - An implementation of prioritized path planning fora large fleet of autonomous straddle carriersHaye Lau, , Dikai Liu, Shoudong Huang, Daniel Pagac

The productivity of automated vehicles in congested container ter-minals can be best realised when individual vehicles can freely movein any allowable direction as appropriate at the time. This howevercomplicates the coordination required between the vehicles. We dis-cuss some results from an implementation of prioritized path plan-ning in a bi-directional network for a large fleet of autonomous strad-dle carriers in a fully automated terminal in Australia.

12

IFORS 2008 - Sandton MD-14

4 - Vehicle Disptaching for Transshipment Hub Opera-tionsDavid Lee

In this talk, we present a problem which is motivated by a transship-ment container hub operation. We aim at determining the fleet sizeas well as the vehicle dispatching by considering both the quay craneand yard crane operation constraints. We formulate the problem as aMIP model, and develop efficient approaches to tackle the problem.


Finance: Financial modelling

Stream: FinanceInvited sessionChair: Galina Andreeva

1 - Modelling the collection process for unsecuredconsumer loansLyn Thomas, , Adiel de Almeida Filho, Christophe Mues

The Basel accord highlights the need to model the collection processso as to estimate LGD. This paper describes how the recovery rateon unsecured consumer debt is affected by the actions of the collec-tor and the circumstances of the defaulter. It models the collectionsprocess at a strategic level by a decision tree , while at the opera-tional level it can be modelled as a stochastic dynamic programmingproblem. The properties of the optimal operational collection strat-egy are developed including which actions to take and how long toperform them for so as to minimise loss given default

2 - A new system of multivariate distributions: fittingmodels to financial dataAlexander Belgorodskiy, Rose Baker

Baker (2008) uses an order statistics based approach to derive a newclass of multivariate distributions with fixed marginal distributionsof which Farlie-Gumbel-Morgenstern distribution is a special case.We fit these distributions to high-dimensional financial return data.There are plenty of models to choose from since the form of the dis-tribution depends on the choice of marginals. A number of goodnessof fit tests can be used then to determine whether these distributionsprovide a good description of a population as compared to other mul-tivariate distributions.

3 - Financial Model to Efficient Water Allocation be-tween the Municipal and Agricultural SectorsZilla Sinuany-Stern, Erez Braude,

In this study, financial models were used to present, analyze andsolve existing problems in water economies that experience a watershortage. We found a strong similarity in the relationship betweenthe residual claim of the agricultural sector for water (in a nationalwater economy) and the residual claim of shareholders in a lever-aged firm for dividends (in the world of finance). In both cases, allo-cations of water/dividends are made after the municipal sector/debtowners first receive their share of the water/interest payments.

4 - Merton models or credit scoring: modelling defaultof a small businessGalina Andreeva, Jake Ansell, Tachun Lin

Modelling credit risk of a small business is under-researched area.The presentation compares two approaches by applying Merton-typeand credit scoring models to predict financial health of the U.K.small businesses. The comparison is extended to cover several cut-off points, corresponding to different acceptance policies and riskappetites.


Finance: Stocks and bondsStream: FinanceInvited sessionChair: Hitoshi Takehara

1 - An Investigation in the Autoregression of ImpliedVolatility in Selected South African StocksOlufemi Adetunji, Venkata Yadavalli

GARCH and implied volatility models are the currently pervasivevolatility models after the black Monday. An investigation is con-ducted in the autoregressive properties of at the spot implied volatil-ity in South African equity market, combining GARCH and impliedvolatility for the prediction of volatility evolution is less liquid op-tion markets.

2 - Expected Return, Liquidity Risk, and the Contrar-ian Strategy: Investigation of Tokyo Stock Ex-change FirmsHitoshi Takehara

We try to extend the Fama and French(1993)’s 3 factor model byadding contrarian (Loser-Minus-Winner, LMW) factor and Liquid-ity (Illiquid-Minus-Liquid, IML) factor. After adjusting for the fac-tor risks with the conditional asset pricing model, book-to-marketratio and past realized return of the individual firms cannot predictabnormal returns any longer. Our findings suggest that anomaliesfound for value stocks and momentum in Japan cannot be used asevidence against the basic notion of efficient market hypothesis.

� MD-14Monday, 2:00 PM - 3:30 PM - Sandton Sun: Maroela North

DEA: Applications 2

Stream: DEA and Performance ManagementInvited sessionChair: Emilio Galdeano

1 - A new approach for classifying the non-homogeneous decision making units using DEAMohammad-Reza Alirezaee, Arash Aliakbari, AhmadMakui

Homogeneity assumption,as one of the most important assump-tions of DEA,is sometimes ignored.This may cause some unreason-able results.This paper redefines the homogeneity assumptions andpresents a method for comparing the efficiency of non-homogeneousDMUs.A classifying method based on the Multiple Criteria DecisionAiding,the ELECTRE TRI method has been used which coincidesperfectly with the method of Banker and Morey. This model hasbeen tested by a case of Banking which ranked by using ELECTRETRI method and then,their efficiency has been computed and com-pared with banker and morey method

2 - Enhancing Decision-Making in SMEs with DEAUlrich Bodmer, David Roger Vaughan, Peter Wagner

Decision-making usually is regarded to be a matter of rationality. Atleast in SME, however, social and psychological factors often alsohave a decisive influence upon decision-making. DEA will be pre-sented as a tool to enhance decision-making with quantitative andqualitative facts on the single enterprise level compared to traditionalapproaches.

13

MD-16 IFORS 2008 - Sandton

3 - Productivity and Environmental Performance:An Analysis Based on Boostrapped DEA andMalmquist TFP IndicesEmilio GaldeanoThe aim of this paper is to offer a decomposition of productivity in-dices incorporating environmental changes. The bootstrapped DEAis applied and the best practice frontier obtained corresponds to de-cision making units showing the best environmental behaviour. TheMalmquist total factor productivity (TFP) is decomposed into tech-nological change, efficiency and environmental change. In the em-pirical analysis, the model followed is an input-oriented approachrelated to undesirable output (environmental performance attribute)by using a set of horticultural cooperatives.

� MD-16Monday, 2:00 PM - 3:30 PM - Sandton Sun: Jacaranda

Military Optimization Models

Stream: Military Applications of ORInvited sessionChair: Hoong Chuin Lau

1 - A Hybrid Approach to Convoy Movement Planning, Lucas Agussurja, Hoong Chuin Lau

The Convoy Movement Problem considers the coordination androuting of convoys within a transportation network. An effectivehybrid algorithm has been formulated using the standard Dijkstra’salgorithm and constraint programming techniques. The effective-ness of the algorithm is reinforced with testing on varying problemsizes and complexity.

2 - Heuristic and exact solution methods for the dy-namic weapon target assignment problemFrancois du Toit, Jan van Vuuren, Jaco RouxVarious generalisations of the well-known Dynamic Weapon TargetAssignment (DWTA) problem are considered, accommodating sev-eral temporal stages over which weapon assignment proposals haveto be made. Exact and heuristic methods for solving these modelsare also investigated and compared, both in terms of solution timesrequired and solution qualities obtained.

3 - Application of the Resource Allocation problem inAmmunition ReplenishmentGeertien VenterResource allocation describes a general problem where a limited re-source must be allocated to competing activities, each with its ownyield function and lower and upper bounds. The objective is to op-timize the total combined yield. In this paper resource allocation isapplied to Ammunition Replenishment for the SA Navy.

� MD-17Monday, 2:00 PM - 3:30 PM - Sandton Sun: Oleander

Optimization Software

Stream: OR/MS SoftwareContributed sessionChair: Lutz Westermann

1 - SYMCLIP - System for Multi-Criteria Linear Pro-grammingVeronika Skocdopolova, Josef JablonskyThe paper presents system SYMCLIP - a MS Excel based softwaresupport for solving multi-objective linear programming problems.SYMCLIP is a MS Excel add-in application written in VBA thatuses internal MS Excel optimization solver which enables its uti-lization by any interested professionals. SYMCLIP enables solv-ing MOLP problems by several approaches including goal program-ming models and other methods with information a priori, interac-tive methods, etc. The application is a freeware and can be easilydownloaded from author’s web page.

2 - OPTEX Mathematical Modeling SystemJesus Velasquez

OPTEX Mathematical Modeling System is the technological toolthat supports all informatics projects developed by DecisionWare, atechnology based Colombian company oriented to provide DecisionSupport Systems, based on large scale optimization models, integrat-ing mathematical models with information systems. OPTEX’s solu-tions are competitive, comparable to world quality software avail-able for the development of DSSs.

3 - Global Optimization with GAMSLutz Westermann, Michael Bussieck

The aim of Global Optimization is to find the best solution of anonlinear problem which might have several local optima. In thiscontext GAMS made advances in the recent past by introducingnew solvers with both academic and commercial background intothe system. We discuss requirements for interfacing these solvers.Besides algebraic model information access to a wide range of lo-cal solvers increases the reliability of a GO code. Furthermore,we present tools for benchmarking global and local optimizationsolvers.

� MD-18Monday, 2:00 PM - 3:30 PM - Sandton Sun: Syringa

Graphs and Networks 2

Stream: Graphs and NetworksContributed sessionChair: Jan van Vuuren

1 - The Best Upper Bound For the Randic Index ofTrees, Marina Stojanovic

The Randic index of a graph is defined as the sum of weights as-signed to every edge, where the sum goes on all edges. The weightof an edge is equal to the product of degrees of the end points ofedge, raised to the power of -1. We found maximal value of thisindex and the Max Tree for all trees of order greater or equal to 103.The structure of the Max Tree is a little bit unexpected. For someorders it was predicted by Clark and Moon, for some orders it waspredicted by conjecture of Hu, Jin, Li , Wang and for some orders itdiffers.

2 - Solution methodologies for the classical assign-ment problemGrant van Dieman, Jaco Roux, Jan van Vuuren

The classical assignment problem first appeared in the literature inthe landmark 1952 paper by Votaw and Orden. Over the years manysolution methodologies have been developed for this problem andits many variations. In this paper a number of exact and heuristicsolution methods are discussed and comparisons of the efficienciesof these solution methods are presented. The exact solutions arecompared in terms of the solution time and the heuristic solutionmethods are compared in terms of the solution quality.

3 - Comparing Metaheuristics for the Quartet MethodSergio Consoli, Gijs Geleijnse, Kenneth Darby-Dowman,Jan Korst, Steffen Pauws

Given a set of objects and their pairwise distances, we wish to deter-mine a visual representation of the data. We use the Quartet Methodparadigm to compute a hierarchy of clusters of the objects. As thisproblem is NP-hard, we present and compare metaheuristics to ap-proximate the optimal hierarchy.

14

IFORS 2008 - Sandton ME-02

4 - New upper bounds for the metric dimension of hy-percubes, , , Mirjana Cangalovic

We develop genetic algorithms for computing upper bounds for themetric dimension as well as for the cardinality of minimal doubly-resolving sets of hypercubes. For large scale hypercubes whichcan not be handled by the computer, we approximate the metricdimension using a dynamic programming approach based on thefact that a hypercube can be represented as the Cartesian productof two smaller hypercubes. Now an upper bound can be derivedfrom the metric dimension of one and the cardinality of minimaldoubly-resolving sets of the other factor.

� MD-19Monday, 2:00 PM - 3:30 PM - Sandton Sun: Acacia

Maintenance: Spare parts and logistics

Stream: Maintenance and ReliabilityInvited sessionChair: Aris Syntetos

1 - Distributional assumptions for parametric forecast-ing of intermittent demandMohamed Zied BABAI, Aris Syntetos, Nezih Altay

Forecasting and inventory control for spare parts is a very challeng-ing exercise. Most research in this area focuses on the control ofinventories assuming that an appropriate estimator is in place to fore-cast demand that in turns is represented by one of the standard sta-tistical distributions. The choice of a demand distribution has animportant impact on the stock control performance. In this paper,an empirical investigation is conducted, using a real data set fromthe US Navy and RAF (UK) that assesses the empirical effects ofdistributional assumptions on stock control.

2 - Inventory Control of Spare Parts in the Final PhaseRommert Dekker,

We consider an appliance manufacturer’s problem of controlling theinventory of a service part in its final phase. That phase begins whenthe production of the appliance is discontinued and ends when thelast service contract on that appliance expires. During this perioddemand can be satisfied by repairing the defected product or be re-placed by a new one. We propose two policies two determine theoptimal final order quantity and time to switch to the alternative pol-icy.

3 - Periodic stock control heuristics for spare parts:analysis and empirical investigationAris Syntetos, Mohamed Zied BABAI

Typically, the inventories related to spare parts are managed throughperiodic stock control solutions, though the specific policy selectedfor application will depend upon various parameters including thedegree of intermittence (slow/fast intermittent demands) associatedwith the SKUs under concern. In this research, the performance ofsome periodic stock control heuristic solutions is examined in detail.Those heuristics have been shown to perform well for differing de-mand categories and the investigation under concern allows insightto be gained on demand classification related issues.


� ME-01Monday, 4:00 PM - 5:30 PM - Convention Centre: Ballroom 1

Combinatorial Optimization: RoutingProblemsStream: Combinatorial OptimizationInvited sessionChair: Richard Eglese

1 - An Improved Tabu Search Heuristic for the OpenVehicle Routing Problem with Soft Time WindowsLeon Li, Zhuo Fu, Richard Eglese

An improved tabu search heuristic for the two basic types of OpenVehicle Routing Problem with Soft Time Windows (OVRPSTW) isproposed. Computational results on the benchmark problems areprovided and compared with another method for the open vehiclerouting problem with hard time windows (OVRPHTW) in the liter-ature.

2 - A metaheuristic approach to the routing of securityguardsElias Jakobus Willemse, Johan Joubert

One of the most essential security services offered by security es-tates is the continual patrolling of the estates’ roads and paths. Thispaper presents a Tabu Search algorithm capable of generating patrolroutes for an estate’s security guards. The paper shows that the prob-lem of designing these routes is an Arc Routing Problem (ARP), butunlike most ARPs the objective is not to minimize distance travelled;rather the objective is to traverse all the arcs as evenly and regularlyas possible. The algorithm is illustrated with a real problem instancefrom an estate in Gauteng, South Africa.

3 - An Efficient Time and Space K Shortest Path TreesAlgorithmAntonio Sedeño-Noda, Carlos González-Martín

We address the problem for finding the K best simple paths con-necting a source node with any other non-source node in a directednetwork with arbitrary lengths. The main result in this paper is theproof that the kth shortest path tree is adjacent to at least one ofthe previous (k-1) shortest path trees. Consequently, we design anO(f(n,m,C)+ Km) time and O(K+m) space algorithm to determinethe K shortest path trees, in a directed network with n nodes, m arcsand maximum absolute length C , where O(f(n,m,C)) is the best timeneeded to solve the single-source shortest simple path.

4 - Green Logistics: Avoiding Congestion in VehicleRouting and SchedulingRichard Eglese, William Maden

Time-varying travel times for links in a road network provide in-formation on when and where congestion slows the flow of traffic.A heuristic algorithm is developed that uses this data for a vehiclerouting and scheduling problem and results of computational exper-iments are presented indicating the environmental impact.


Linear Programming

Stream: OptimizationContributed sessionChair: Claudio Sodini

1 - A fast and robust parameter choice for interiorpoint methodsAurelio Oliveira, , Clovis Perin

15

ME-03 IFORS 2008 - Sandton

The choice of the perturbing parameter and the primal and dual steplengths are delayed allowing to build a merit function of these pa-rameters from the predictor corrector search directions. The polyno-mial merit function has degree 12 and is solved in a mixed grid andanalytical approach. The merit function aims to achieve robustnessand simultaneously indicates where linear system solutions can beavoided leading to faster iterations. Numerical experiments revealthat the proposed approach is robust and fast in comparison with thetraditional parameter choice performance.

2 - Computer experiments: Synergies between linearprogramming, experimental design and statisticalmetamodelsDiki Langley, Roelof CoetzerStatistical experimental design methodology was applied to predictproperties of petrol component blends, using a linear program whichrepresented the actual system from which the blends originated. Theresultant statistical metamodels could be used to improve the linearprogram accuracy of blend property prediction, among other bene-fits.

3 - Global optimization of a generalized linear programClaudio Sodini, Riccardo CambiniThe aim of this paper is to propose a solution algorithm for a classof generalized linear programs having a polyhedral feasible region.The algorithm is based on the so called "optimal level solutions’method. Various global optimality conditions are discussed and im-plemented in order to improve the efficiency of the algorithm.


Timetabling and Scheduling in Railways

Stream: Railroad ApplicationsInvited sessionChair: Christian WeberChair: FEDERICO BARBER

1 - Disruption Simulator in Railway timetabling for Em-pirical Robustness AssessmentFEDERICO BARBER, MIGUEL A. SALIDO, LAURAINGOLOTTI, ANTONIO LOVA, PILAR TORMOSSeveral approaches have been developed to analyze the robustnessof timetables. Simulation methods are related to "what-if" analysis.It consists of simulating incidences in a timetable, re-planning, andevaluating the effects on the final timetable. We present a railwaysimulator able to perform flexible empirical analysis of effects ofdisruptions on parameterized and on real-world timetables. Throughdifferent statistical analysis (where, when, and what disruptions oc-cur), we empirically show how the main parameters (heterogeneity,capacity and speed) affect to robustness.

2 - Analytical Assessment of Timetable RobustnessMIGUEL A. SALIDO, FEDERICO BARBER, LAURAINGOLOTTI, PILAR TORMOS, ANTONIO LOVAThe robustness of railway timetables depends on several factors,as train heterogeneity, used capacity and average speed of runningtrains. We also could minimize knock-on delays with appropriatetime supplements in journeys and commercial stops, security mar-gins in traffic operations or buffer times between consecutive trains,which leads to a trade-off between optimality and robustness. Thiscontribution deal with analytical assessment of timetable robustness:How robust a timetable is? Which the main timetable data that affectto their robustness are? How can increase timetable robustness?

3 - Periodic timetable scheduling tool used in FrenchrailwaysChristian WeberPeriodic timetable scheduling between main railways stations is to-day an important topic to increase railways market. To design suchperiodic timetables, French railways use an optimization tool basedon the description of the wished trains services on the network, anobjective function based on the value of trains and of passengerstraffic. It calculates also the robustness of the optimized timetable.


Applications of OR in Agriculture

Stream: OR in Agriculture and Natural ResourcesInvited sessionChair: Alejandro Mac Cawley

1 - A Visualization Aid for Informal ReschedulingSergio Maturana, Victor Perl

Production schedules frequently suffer disruptions in the productionfloor, which may force a reschedule of the remaining jobs. In manysettings, this rescheduling is carried out without the aid of the sys-tem that generated the schedule. We propose a visual representationbased on constraint program to aid this type of rescheduling, whichwe call informal. The representation allows decision makers to usetheir cognitive abilities to quickly generate a new schedule. A visu-alization tool to support the scheduling of a wine bottling line wasdeveloped to test the feasibility of the proposal.

2 - A decision support model for the choice of culti-vars of deciduous fruit on a Western Cape farmWim Gevers

One of the prime agricultural export products of the Western Caperegion of South Africa is deciduous fruit. Over the past numberof years there have been major changes in popularity of cultivars.A model was built to determine how existing farms with a currentportfolio of fruit trees should change over time to become more prof-itable by systematically replacing old trees with new cultivars, takinginto account cash flow and labour constraints. Many of the farms inthe region can become more profitable by converting current stockto newer cultivars.

3 - A Bulb Acquisition Planning Model for a ChileanLily Flower EnterpriseAlejandro Mac Cawley, Sergio Maturana, Manuel Aspillaga

The acquisition process of raw materials in the flower industry in-volves a large number of technical, biological and economic vari-ables. We develop a Bulb Acquisition Planning (BAP) mixed inte-ger optimization model (MIP). The model incorporates aspects suchas substitutability, precocity, risk diversification, seasonality, and in-ventory among others. The model was implemented using enterprisedata from production planning executed in the past season. Finallythe results of the BAP model and the executed plan were comparedwith respect to cost, container programming and inventory mainte-nance.

� ME-07Monday, 4:00 PM - 5:30 PM - Convention Centre: Boardroom 3

OR-in-Development 3


1 - Visitor profile, satisfaction levels and clustering oftourists for decision making in Michoacan, MexicoMarco Serrato, Juan Tello,

A survey was conducted on tourists that visited Michoacán, México,to determine their profile and satisfaction levels. A cluster analysiswas developed to categorize their lifestyle. A regression analysiswas then performed to identify variables that influence satisfaction.Results are satisfactory since consequent actions have increased rev-enues reported by organizations

16


2 - Integrated Aircraft Fleeting and Routing at TunisAirMohamed Haouari, Najla Aissaoui, Kamel Berrima, HanifSherali, Farah Zeghal Mansour

We address models and solution approaches for integrated airline op-erational planning problems, that arise at the Tunisian national car-rier, TunisAir, where the fleeting and aircraft routing problems areconsidered simultaneously. Computational experiments conductedon real-data show that the proposed approaches are effective, fast,easy-to-implement, and robust.


Muliple Criteria Decision Aiding: Sortingproblems

Stream: Decision AnalysisInvited sessionChair: José Rui Figueira

1 - Extension of the FlowSort method to multiple refer-ence profiles for the definition of the categoriesPhilippe Nemery

Generally, in sorting problems the categories are defined by uniquecentral or limiting profiles for each category. In this paper we willpresent some extensions of the sorting method called FlowSort to thecase in which the ordered categories are defined by several limitingor central profiles. Therefore, we will consider the definition of thereference profiles as uncertain and propose a stochastic extension ofFlowSort. This approach can under certain conditions be applicablefor the sorting problem in a group-decision making context.

2 - Relational Multicriteria Clustering : A new modelStefan Eppe, Yves De Smet

In multiple criteria, as the number of the alternatives increases, in-terpretation of the results gained through classical multicriteria out-ranking methods however becomes difficult. This paper will fur-ther explore the spoor of relational multicriteria clustering methods,which address the aforementioned drawback by synthesizing the re-sults into clusters, yet giving valuable information about the rela-tions between them. After reviewing the latest results achieved inthe case of binary outranking relations, the extension to valued pref-erence relations will be analysed.

3 - Electre Tri-C: A Multiple Criteria Ordinal Classifica-tion Method BasedJosé Rui Figueira, Juscelino Almeida-Dias, Bernard Roy

In this paper we propose an outranking method for ordinal clas-sification or sorting problems within the Electre framework. Thisextension considers that the pre-defined and ordered categories arebased on central actions instead of profile-limits. Some propertiesfor an assignment rule, structural requirements, and natural condi-tions should be fixed before designing the assignment rule. Then,two different assignments rules, one ascending and the other de-scending are proposed. The paper contains also the theoretical foun-dations of the new method and other extensions.


Simulation 3Stream: SimulationContributed sessionChair:

1 - Reliability modelling as a tool to assist plant design

Production plants are designed all over the world to create productand throughput to make profit. Usually, once it is commissioned thethroughput promised and what is seen differs dramatically. . . why?Was availability realistically assessed during design, can it be im-proved with simulation models?

2 - Refinery Stochastic ModellingMarlize MeyerStochastic modelling (Arena) is used in two Petrochemical Refiner-ies (coal to liquid and crude oil) to model the complete value-chainusing several semi-integrated models. This is an unconventional useof stochastic models. This presentation elaborates on the integrationfocus across the value chains and how these models link to othermodels in the integrated system.

3 - Determining tactical operations policies for an autocarrier using discrete-event simulation, James Bekker

We report on a simulation study that was done to assist an auto car-rier with tactical planning. The objective is to identify a fleet com-position and operational policies that would maximise the numberof cars delivered, while being on time. Carrier utilization must bemaximised, adhering to staff and maintenance schedule constraints.

� ME-11Monday, 4:00 PM - 5:30 PM - Convention Centre: CommitteeRoom 2

Container Shipping Flow

Stream: Container PortsInvited sessionChair: Meng Qiang

1 - Container Shipping Network Problems with Con-tainer Management and Competition among Ship-ping Companies for Mega-ship EraKoichi Shintani, Akio Imai, YingLai WangThis study addresses the issue of designing service networks for con-tainer liner shipping while explicitly taking into account empty con-tainer repositioning and competitive circumstances. The liner ship-ping industry is faced with the problem of allocating empty contain-ers. Two typical kinds of shipping networks are examined by math-ematical modeling: Multi-port calling and Hub-&-Spoke. Fromviewpoint of viability of mega-ship, it is expected to be the signifi-cant issues that how to reduce the transshipment cost (handling cost)in hub ports for an efficient distribution of empty containers.

2 - Optimal Fleet Planning with Cargo Demand Uncer-tainty for Liner ShippingMeng Qiang, Tingsong Wang, Tien Fang FwaThis paper first builds a stochastic programming model for the op-timal liner fleet routing problem that aims to maximize the profitthrough the assignment of containerships to a finite set of candidateroutes, and the determination of the required service frequency andthe number of containers to be transported by each ship betweeneach pair of ports. Based on the model, it proceeds to develop astochastic dynamic programming model for the optimal fleet plan-ning with cargo demand uncertainty. Numerical examples are usedto demonstrate the proposed models and solution methods proposed.

� ME-12Monday, 4:00 PM - 5:30 PM - Convention Centre: CommitteeRoom 4

Finance: Portfolio optimisation

Stream: FinanceInvited sessionChair: emmanuel jurczenko

17

ME-14 IFORS 2008 - Sandton

1 - A Mixed Integer Programming Approach to Dy-namic, Multi-Criteria Portfolio Optimization

MIP formulation and weighting approach are proposed for dynamicportfolio optimization over a rolling planning horizon. The objectiveis to allocate wealth on assets to maximize the weighted differenceof the portfolio expected return and the threshold of the probabilitythat the return is not less than a required level.

2 - FSD portfolio optimality testMilos Kopa,This study develops a test for first-order stochastic dominance (FSD)optimality of a given portfolio relative to all possible portfolios. Tosimplify the search over all increasing utility functions, we reformu-late the problem in terms of piecewise-constant utility functions. Weprovide a linear programming criterion for implementing the test. Toidentify the input for the linear programming problem, we may usemixed-integer linear programming or subsampling techniques. Weshow that the US stock market portfolio is significantly (98%) FSDnon-optimal relative to benchmark portfolios.

� ME-14Monday, 4:00 PM - 5:30 PM - Sandton Sun: Maroela North

DEA: AHP / MCDMStream: DEA and Performance ManagementInvited sessionChair: Valério Salomon

1 - A Comparison among Dominant AHP, similar Dom-inant models and ANPEizo KinoshitaAHP was proposed by Saaty and he extended AHP to ANP as net-work structure with Super matrix. This paper compares DominantAHP (proposed by Kinoshita Nakanishi), similar Dominant mod-els (Belton/Gear Model,Referenced model,Linking pins model) andANP. Also presents numerical example and process of above meth-ods. Next we introduce the basics of AHP/ANP and explain princi-ple rule of Dominant AHP. Moreover we describes numerical exam-ples of Dominant AHP,similar models.

2 - An Application of the AHP Method to Analyze RoadPaving-Routing Alternatives, Christiane W. Nogueira

We present the results of an application of the AHP method in-volving qualitative and quantitative criteria for ranking road paving-routing alternatives, in the Itajai Valley Region, Santa Catarina,Brazil. The main outcome of the study is its contribution for thedevelopment of a unified methodology intended to define the bestroutes for road paving in the hinterland, ensuring that public re-sources are applied in the best way, and therefore providing higherreturns for the community.

3 - Tests of Column Averaging for MCDM RatiosWilliam Wedley, Eng ChooVarious methods are available to elicit preference ratios from apaired comparison matrix. Columns of the matrix and other poten-tial columns are separate estimates of the ratios. This study investi-gates a new method, column averaging. From simulated true values,perturbated matrices are used as data to test the efficacy of variousmethods.

� ME-16Monday, 4:00 PM - 5:30 PM - Sandton Sun: Jacaranda

Operations Management: Inventory risk,assortment, and pricing

Stream: Operations ManagementInvited sessionChair: Frank Y Chen

1 - Comparison of Models for Managing Profit Risk inInventory SystemsCandace Yano, Houmin Yan, Hanqin Zhang

Many different approaches have been suggested for managing riskin inventory systems. We consider a situation with uncertain de-mand in which the decision-maker wishes to manage his risk withrespect to meeting profit targets in each of several time periodsvia inventory-related decisions. We explore differences among so-lutions from mean-variance, conditional-value-at-risk and chance-constrained representations of these problems. We illustrate the ad-vantages of using the chance-constrained model, both in terms ofprofit considerations and in ease of implementation.

2 - A Censored Newsvendor Model with UnnormalizedProbabilitiesSuresh Sethi, Alain Bensoussan, Metin Cakanyildirim

This paper revisits the finite-horizon model of a censored newsven-dor by Ding et al. (2002). An important result claimed there with-out a proper proof is that the myopic order quantity is always lessthan or equal to the optimal order quantity. Lu et al. (2005b) havesupplied a correct proof of the result. We analyze the same modelusing the concept of the unnormalized probability, which simplifiesthe dynamic programming equation and facilitates the proof of theclaim. Moreover, it produces the proof of the existence of an optimalsolution for an infinite-horizon setting of the problem.

3 - Dynamic Assortment PlanningFelipe Caro, Rene Caldentey

We investigate optimal assortment planning strategies for a retailerwith limited shelf space. The retailer can choose among basic andfashion items with low and high risk (and return) respectively. De-mand learning is allowed and depends on the amount of space al-located to the volatile products. Our motivation is in the apparelindustry, and we explicitly model the vogue as a stochastic processthat the retailer tries to follow. The objective is to maximize thelong-term value of the retail business by dynamically adjusting themenu of products on display.

4 - Dynamic Pricing and Inventory Control in a Make-to-Stock Queue with Information on the ProductionStatusFrank Y Chen

This paper addresses the dynamic pricing problem of a single-item,make-to-stock production system. Demand arrival rates depend onthe selling price, and item processing times follow an Erlang dis-tribution to capture the information on the production status. Wecharacterize the optimal policy which coordinate the production andpricing decisions. We also use our model to quantify the benefit ofemerging technologies such as RFID that can provide informationon the production status or the location of the pipeline inventoryalong the supply chain.

� ME-17Monday, 4:00 PM - 5:30 PM - Sandton Sun: Oleander

Modeling Languages and SolverInterfacesStream: OR/MS SoftwareContributed sessionChair: Robert Fourer

1 - Introducing CoinMP: A Simple C-API Windows DLLand Linux implementation of CLP, CBC, and CGLBjarni Kristjansson

The COIN Open Source Initiative has become very popular in therecent years. To make life easier for users that simply want to solvemodels and not compile C++ applications, we have developed a stan-dard C-API Windows DLL CoinMP.DLL that implements most ofthe functionality of CLP, CBC, and CGL. A Linux/Unix version us-ing AutoMake is also available.

18


2 - Extended Mathematical Programming in GAMSJan-Hendrik Jagla, Michael Ferris, Alex Meeraus

Modeling systems support a wide collection of established mathe-matical programming classes through their MP solver clusters. Theclassical interface between a modeling system and a MP solver con-sists of passing on information about the constraint matrix and func-tion and derivative evaluations for non-linear problems. With emerg-ing new model types and advanced solver technology the need in-creased for exchanging more structural model information. We willoutline recent additions to the GAMS system that allow solving newmodel classes using established methods.

3 - The Optimization Services Project on COIN-OR:Progress and PlansRobert Fourer, , Kipp Martin

We describe Optimization Services (OS), a unified framework for anew generation of Internet optimization systems, which has advan-tages for both developers and users of optimization software. OSincorporates XML-based standards for representing and communi-cating optimization problems, so that each solver needs only oneinterface. Components of OS include an XML-based representationfor problem instances and a corresponding in-memory representa-tion. An open source library, hosted on COIN-OR, provides utilitiesfor reading and writing these representations and for converting be-tween them.

� ME-18Monday, 4:00 PM - 5:30 PM - Sandton Sun: Syringa

Survivable Telecommunications NetworkDesign

Stream: TelecommunicationsInvited sessionChair: Luís Gouveia

1 - Path Generation Issues in Survivable Network De-sign ProblemsMichal Pioro, Tomasz Sliwinski, Michal Zagozdzon,Wlodzimierz Ogryczak, Mateusz Dzida

The paper deals with network design problems where, in the case ofa failure, the affected primary flows are restored using assigned pro-tection paths. The problems are formulated as linear programmesusing the non-compact link-path notation of multi-commodity flow.The paper presents and compares path (column) generation algo-rithms enabling efficient resolution of the considered design prob-lems for networks of realistic size.

2 - Survivable Network Design with Uncertain TrafficRequirementsFanie (SE) Terblanche, Roland Wessäly, Giel Hattingh

Obtaining optimal solutions for the survivable network design prob-lem with fixed traffic requirements has proven to be computationallychallenging. The situation becomes worse when uncertain traffic re-quirements are considered. For obtaining good solutions we con-sider a scenario based formulation of the problem and explore aniterative polyhedron expansion technique as a solution approach

3 - The K-edge-connected 3-Hop constraint networkdesign polyhedronBernard Fortz, Quentin Botton, Luís Gouveia

The K-Edge-Disjoint Hop-Constrained Network Design Problemconsists in finding a minimum cost subgraph such that there exists atleast K edge-disjoint paths of length at most L between given pairsof nodes. This problem was considered in the past using only designvariables. We consider an extended formulation, introducing flowvariables to model the paths. New classes of valid inequalities areadded to strengthen the model leading to the complete description ofthe associated polyhedron for L = 3.

� ME-19Monday, 4:00 PM - 5:30 PM - Sandton Sun: Acacia

Maintenance 1Stream: Maintenance and ReliabilityInvited sessionChair: Adiel Teixeira de Almeida

1 - Managing Warranted Systems Sold with AdditionalService and its Optimal Maintenance Policy

We investigate a system with free warranty involving minimal repair.Additional service is offered whereby repair level of first failure canbe chosen. What is the optimal choice? How does inclusion of ad-ditional service affect consumer’s system maintenance policy? Westudy optimal strategy structures and provide numerical examples.

2 - Preventative maintenance regimes under machineobsolescenceDiego Ruiz-Hernandez, Christopher Kirkbride

We consider a variant of the machine maintenance problem in whichthe decision-maker may continue to maintain ageing equipment ordecide to upgrade. Our aim is to develop maintain/upgrade heuris-tics that attempt to minimise expected costs incurred in maintaininga facility/fleet of machines and numerically evaluate their perfor-mance.

3 - Multicriteria decision approaches for selection ofpreventive maintenance intervals in the context ofservice production systemsAdiel Teixeira de Almeida, Cristiano Cavalcante

Multicriteria approach is generally necessary for selection of preven-tive maintenance interval in service production system. In this kindof system the product is produced while the client is being servedand the product may be received with disturbs caused by failures.Then, the losses regarding failures can not be simply accounted forin a monetary way. In these systems the preventive maintenance hasto account for several different objectives such as: cost, reliability,downtime and quality. The paper this problem is analysed throughmultiattribute utility theory and PROMETHEE methods.

19

TA-01 IFORS 2008 - Sandton

Tuesday, 8:00 AM - 9:30 AM

� TA-01Tuesday, 8:00 AM - 9:30 AM - Convention Centre: Ballroom 1

Combinatorial Optimization: RecentAdvances in Scheduling

Stream: Combinatorial OptimizationInvited sessionChair: Ibrahim H. Osman

1 - A Variable Neighbourhood based Heuristic for theHeterogeneous Fleet Vehicle Routing ProblemNiaz Wassan, , Said Salhi

The VRP with heterogeneous fleet is investigated using some adap-tations of the basic VNS by exploring several types of neighbour-hoods and local searches. The initial solution is obtained by the Di-jkstra’s algorithm based on a well defined cost network. The methodis tested on existing and new (larger) data sets.

2 - Evaluation of Heuristics for the Weighted TardinessSingle Machine Scheduling Problem with PeriodicMaintenanceIbrahim H. Osman, hocine belouadah, Krzysztof Fleszar

We consider the problem of scheduling a set of jobs on a single ma-chine with its own schedule of periodic-maintenance operations withthe objective of minimizing the total weighted tardiness of jobs. Thisproblem can be shown to be NP-hard. A number of heuristics basedon well-know dispatching rules in addition to new ones are devel-oped. Computational experience are reported on a set of benchmarkinstances with different characteristics to evaluate the effectivenessof the proposed heuristics.


Integer Programming

Stream: OptimizationContributed sessionChair: Fabrice Talla Nobibon

1 - Integer programming models for the optimizationof promotion campaignsFrits Spieksma, Fabrice Talla Nobibon, Roel Leus

This paper presents different related models for the optimization ofpromotion campaigns based on integer programming. We describedifferent algorithms (exact and heuristics) for solving the developedmodels and discuss the results on the comparison of their efficiencybased on computational experiments.

2 - Solutions diversification in a column generationschemeNora Touati, Lucas Létocart, Anass Nagih

It is well known that the addition of many columns to the masterproblem at each iteration of column generation, allows to character-ize the optimal base more quickly, so, we get a decrease of the num-ber of iterations. Unfortunately, this intensification quickly over-loads the master problem and generally, the optimal base contains avery restricted subset of the generated columns. Thus, we compute abig proportion of bad columns. We study in this work some caracter-istics of the generated columns and try, using different algorithms,to select good ones to insert into the master problem.


Optimization Modeling for the PublicSectorStream: OR and the Public SectorContributed sessionChair: Mali Sher

1 - Network flow methods for best approximation ofexact quotas in biproportional apportionmentPaolo Serafini,

We apply mathematical programming techniques to the problem ofassigning seats in the Italian parliament with the goal of respectingdouble proportionality as much as possible. Minimizing the maxi-mum error among seats assignments can be done polynomially. Weaddress also the problem of finding a certificate proving optimalityto the layman.

2 - Applying Mathematical Modeling Approaches forInvestigating Japanese Food Supply Security Sys-temTatsuo Oyama

We investigate the past decreasing trend of the food self-sufficientratio (SSR) in Japan with their causes, then try to decompose thesedecreasing changes into two main factors (i) the change in caloriessupplies, and (ii) the change in the SSR itself. We show a Japanesefood supply and demand network in order to build an optimiza-tion model representing Japanese food supply system. Optimizationmodel analyses can be applied for this food supply network system,thus numerical results for improving Japanese SSR are shown withour strategic policy anaysis.

3 - A locate-allocate model for interurban traffic policevehiclesMali Sher, Nicole Adler, Shalom Hakkert, Jonathan S.H.Kornbluth

Interurban traffic police vehicles (ITPV) both provide assistance todrivers and uphold the law. The police working method locates ITPVon critical sections of the network. The ITPV handle calls for ser-vice on the network and attempt to be conspicuous on the chosensections in order to prevent illegal behavior and to issue tickets fortraffic violations. Consequently, coverage of the network is impor-tant as is their location in order to maximize utility. A multi-criteriadecision problem has been defined with a flow-capturing objectivefunction under a legally-defined coverage constraint.


Sports and OR

Stream: OR and SportsInvited sessionChair: James Cochran

1 - A DEA approach to selecting line-ups of a baseballteamNobuyoshi Hirotsu, Tohru Ueda

We propose a DEA approach to selecting line-ups of a baseball team.In this approach, we evaluate baseball players from the view pointof uniqueness, and show the evaluation of players by means of DEAefficiency together with Scoring Index, a measure of batting abil-ity. We then evaluate line-ups by the number of the DEA effectiveplayers and the expected runs scored in a game. We illustrate thismethod using real data of MLB, and demonstrate how this approachmay help to select the line-ups, by showing the actual line-ups withthe expected runs scored and the number of effective players.

20

IFORS 2008 - Sandton TA-07

2 - Referee assignment in sports leagues: Exact andapproximate multi-objective approachesCelso Ribeiro, Alexandre Duarte

Amateur sports leagues face the problem of assigning referees tothe games of a competition. The number of referees may vary, de-pending on the sport, league, and tournament. Referee assignmentis subject to a number of constraints and optimization criteria. Theproblem consists in assigning referees to all games of a tournament.We propose exact and approximate algorithms for a bi-objectiveproblem variant: the exact algorithm based on dichotomic searchmakes use a vehicle routing formulation, while the approximate oneis based on the ILS metaheuristic combined with an embedded MIPstrategy.

3 - Bender’s Approaches to Sports SchedulingMichael Trick

Sports scheduling has proven a rich area for exploring alternative ap-proaches for discrete optimization. Even relatively small instancescan be hard to solve, and there are a wide variety of problem types tochoose from. I will outline results on league and umpire schedulinginvolving the use of logical Benders’ constraints. These approachescan be orders of magnitude faster than direct integer programmingmethods.

4 - Is the March Madness Seeding Process Consis-tent?James Cochran, Kevin James

In the NCAA Division 1 men’s team basketball tournament, we ex-pect the proportion of wins by lower seeds to be highest in earlyrounds and decline as rounds progress. However, these proportionsare not realizations of simple binomial processes; in each round theprobability a lower seed wins varies greatly, producing a mixture ofbinomial processes. We model a round as a binomial process aug-mented by a beta prior, set a parameter equal to the round, and findthe MLE of the other parameter to assess reduction in variation ofthe probability a lower seed wins as the tournament progresses.

� TA-05Tuesday, 8:00 AM - 9:30 AM - Convention Centre: Boardroom1

OR Education 1Stream: OR EducationInvited sessionChair: Peter Bell

1 - Preparing students for the world of OR work: Lan-caster’s experienceGraham Rand,

Lancaster’s 12-month MSc course has been established for over 40years and concludes with a 4 month project for an external organisa-tion. The aim is not only to give students real world OR experiencebut also to deliver real benefits to the client organisation. We reporton our experience, including preparing the students and the issuesinvolved in setting up the projects, supervising and assessing them.

2 - Six sigma in a tertiary education environmentIlze Du Plooy, CJ Swanepoel, MF Immelman, WillemienVan Hoepen

The University of South Africa is a distance education institutionwith more than 250 000 registered students. It relies mainly onprint-based study material that is collated and sent by mail, carrier orhanded to the student when s/he registers. The collation and pack-ing process was taxed to its limits, and the delays in despatchingmaterial started affecting the success rate of students. Members ofthe Department of Decision Sciences applied Six Sigma principlesin the store of the Department of Despatch to optimise the process.

3 - Learning Operations Research Online: Benefits,Challenges and ExperiencesJavier Faulin, Sara Veronica Rodriguez-Sanchez, LluisMPla, Angel A. Juan, Pau Fonseca i CasasSome benefits, challenges and experiences related to the teachingand learning of operations research (OR) in online environmentsare discussed. Information technologies offer new ways to teachand learn OR in agriculture studies. In particular, they have driventhe growth of distance learning opportunities, as students who aretime bound or place bound can now access OR courses at theirconvenience. The use of computer software and collaborative e-learning to increase students’ motivation for OR are also discussed.Finally, some real experiences in different online environments aredescribed.

4 - Cases for teaching revenue managementPeter BellThis presentation reviews and discusses a number of real-worldcases that bring a degree of realism to the teaching of revenue man-agement topics.


Economic Development: Modelling andFinanceStream: OR and Economic DevelopmentContributed sessionChair: Paul Randall

1 - Development of resource utilization of construc-tion production in LatviaIneta Geipele, Janis VanagsWith the fast economic development of Latvia, overfull demand ofconstruction production has increased that has lead to the price riseof construction production. In the research the main attention isgiven to construction resources - workforce, building materials andavailable fixed assets of companies producing building materials.Due to the augmentation of shortage of manpower in the economics,the issue of wider usage of civil engineering resources and technicalequipment will become more topical in future.

2 - An Empirical Analysis of Capital StructureAlain Chevalier,In this paper, we investigate empirically the determinants of capitalstructure of 293 listed companies in Jakarta Stock Exchange (JSX).There is a change of financing behaviour in pre and post-crisis periodin Indonesia. In pre-crisis period profitability has a positive corre-lation with debt ratio but during the crisis and in post-crisis periodthe relation is negative. Before crisis, trade-off theory is favourablewhereas in post-crisis period pecking-order theory is more relevant.

3 - DEA Methods in the Moldavian EconomyTkacenko AlexandraThe Moldavian economy has recently gone through a very hard pe-riod. Events such as political independence and the introductionof a national currency led to great economic perturbations. That iswhy defining directions of development for the national economy isimportant. In this work we present an analysis of the situation ofsome sectors of the Moldovian economy. This was possible usingthe DEA technique based on a mathematical programmer approach.An analysis of some data concerning the efficiency or inefficiency ofthe sectors using the Efficiency Measurement Programs is provided.

4 - Performance and BudgetingPaul RandallNamibia embarked on a programme to link resource allocation tooutcomes, culminating in the formal launch of a new National Bud-get system. The way that system was developed is described alongwith the initial results from its use and the means by which the nec-essary change in mindsets is being entrenched.

21



Multicriteria Applications

Stream: Decision AnalysisInvited sessionChair: Theodor Stewart

1 - Using MCDA to measure povertyVivien Kana Zeumo, Alexis TsoukiàsIs it possible to measure poverty in a meaningful way, suitable to thepursuit of an effective policy of poverty reduction? We present anapproach based on the use of ordinal measurement techniques de-rived from classification and multi-attribute decision analysis meth-ods. The hypothesis is that there is no objective measure of poverty,but only measures instrumental to a poverty reduction policy. Theissue of meaningfulness is analysed both from a theoretical pointof view (measurement theory) and from an operational one (policyeffectiveness). The case of Burkina Faso is discussed.

2 - Participatory Decision Processes and DecisionSupportAlexis Tsoukiàs, Katherine Daniell, Chabane MazriParticipative decision processes are becoming of increasing impor-tance in modern societies due to the demand of legitimation and thedifficulty to handle complex problem situations. The result is a de-mand for decision support tools and methodology able to capture thecharacteristics of such processes and to be relevant for the participantstakeholders. In the paper we present a number of methodologicalissues characterising participative decision processes and we showhow these influence the type of decision aiding provided. Two casestudies are used in order to support our claims.

3 - Modelling Biobjective User Equilibrium on a TollHighway NetworkJudith Y.T. Wang, Andrea Raith, Matthias EhrgottWe present a route choice model for a toll highway network based onbiobjective user equilibrium defined as the condition that traffic ar-ranges itself in such a way that no individual trip maker can improveeither his toll or travel time without worsening the other componentby unilaterally switching routes.

4 - Goal Programming for benchmark selection in DEATheodor StewartBenchmark performance levels are an important output of DEA, butare typically based on a simple input or output orientation (either set-ting input targets for fixed outputs or vice versa). Management maywish to set performance targets which imply adjustments on bothinputs and outputs, in order better to achieve broader organizationalgoals. A combined DEA/GP model is proposed for this purpose, andillustrated with numerical results.


Facility Location and Routing

Stream: RoutingInvited sessionChair: Celeste Pizarro

1 - A location model with preferencesSergio García Quiles, Alfredo MarínThe Simple Plant Location Problem is a well-known problem in lo-cation literature: facilities are opened and customers allocated to theopen facilities to optimize some kind of profit. This model can beextended by taking into account customers’ preferences and embed-ding them into the model. How can this be done and some propertiesof the resulting model will be shown.

2 - Using reduced formulations with modern heuris-tics for solving large scale hub location problemsEnrique DominguezHub location problems are difficult to solve exactly. The best meth-ods available so far cannot solve instances with a large number ofdemand points unless the number of potential hubs is significantlyrestricted. The fact that both the number of variables and constraintsincreases rapidly with problem size means that these formulationsare not amenable to exact techniques. In order to make hub locationproblems more tractable, it is important to reduce the problem size.In this work a framework for making reduced hub location formula-tions is presented.

3 - The Facility Location Problem with Stochastic De-mands, Elena Fernandez, Francisco Saldanha-da-Gama

We consider a facility location problem where we suppose that thecustomers demands follow a Bernouilli distribution. The demandof each customer must be satisfied from the same facility, and eachopen facility has a capacity that represents the maximum number ofcustomers that can be allocated to it. Thus, for a given realizationit is possible that some of the customers with demand will not beserved, which implies a penalty cost. We define a recourse functionand we formulate the problem in terms of finding an a priori solutionthat minimizes the expected cost of the recourse function.

4 - A computational comparison of several formula-tions for the multi-period incremental service facil-ity location problemCeleste Pizarro Romero, Antonio Alonso-Ayuso, ElenaFernandez, Laureano Fernando Escudero, , YolandaHinojosaWe present three 0-1 equivalent formulations for solving the strate-gic problem of timing the location of facilities and the assignment ofcustomers to facilities in a multi-period environment, where all pa-rameters are known. They are based on the so-called impulse vari-ables and step variables. An extensive computational comparisonis performed. As a conclusion, the hybrid impulse-step formulationprovides better computational results than any of the other two for-mulations.


Project Scheduling: Resource Allocation

Stream: Operations ManagementContributed sessionChair: Boaz Golany

1 - Upon One Strategic Model in Project ManagementAvner Ben-Yair, Dimitri Golenko-GinzburgThe project of designing a new product is considered. The problemis to reallocate the company’s expenses within the product’s life cy-cle to maximize the company’s profit. We deal with optimal budgetreallocation within product’s life cycle, including the sub-period ofdesigning the new product, as well as the sub-period of distributingthe manufactured product on the market. The problem deals mostlywith determining the product’s selling price.

2 - Selection and planning of downsizeable projectsunder uncertaintyRoel Leus, , Willy HerroelenProject selection and planning involves choosing the right projectsand allocating resources to these projects. We consider the case ofstatic project selection, where the set of projects available for exe-cution during the planning horizon is known in advance. Since allprojects have uncertain requirements of renewable resources, whichare available in limited amounts, resources may become overcom-mitted. If a resource shortfall should occur, several remedies exist:abolishing projects, reducing the scope of some projects, etc. Westudy the case where projects are downsized in such an event.

22


3 - Allocating resources to R&D projects in competi-tive environmentsBoaz Golany

R&D firms select in which projects to invest and at what levels incompetitive and uncertain settings. We build a model that deter-mines how to allocate a fixed budget among several projects so asto maximize the total expected profit for any number of firms andprojects. We analyze cooperative vs. competitive situations andidentify the gap between the respective outcomes. Mechanisms areproposed to induce firms acting in competitive settings to select amodified Nash solution which coincides with globally optimal solu-tions.

� TA-11Tuesday, 8:00 AM - 9:30 AM - Convention Centre: CommitteeRoom 2

Operations Management: Health Care

Stream: Operations ManagementInvited sessionChair: Yehuda Bassok

1 - Improving Patient Experience: ETC Model for De-signing Service SystemsSriram Dasu

To design service delivery systems that go beyond objective processmeasures we need to understand how patients process experiences.Behavioral sciences suggest that our assessments of a service en-counter are based on three related factors: Emotions, Trust and Con-trol - the ETCs of service. This talk will describe how patient expe-riences could be managed using the ETCs model. We will discussan ongoing study involving patients undergoing radiation therapy.

2 - Improving Supply Chain Performance by ProcessInnovation in a Health Care SystemChang W. Lee, N. K. kwak

Supply chain performance improvement has received a great dealof attention in the health-care industry. The paper considers sev-eral scenarios to innovate existing health-care processes. In orderto reconcile this strategic process innovation matter, primary busi-ness processes are considered to modleing. The solutions to main-tain effective supply and other business functions are identified anddiscussed. The study compares the proposal system with the currentsystem. Results suggest that management seeks strategic advantagesin health care supply chain system.

3 - Rationing of HIV treatment in resource constrainedsettings under supply uncertaintySarang Deo, Charles Corbett

HIV clinics operating in developing countries have to contend withan uncertain supply of drugs. This creates a trade-off between en-rolling new patients and providing uninterrupted treatment to thosealready enrolled. We formalize this trade-off using stochastic dy-namic program, derive the optimal policy for the number of newenrollments and compare its performance with that of policies usedin practice. We discuss the relevance of our model to the resourceallocation decision in non-profits.


Finance: Revenue management andpricing

Stream: FinanceInvited sessionChair: A. D. Amar

1 - Retail Revenue Management under CompetitionJuan-Carlos Ferrer, Pedro Leon

We analyze a stochastic retail revenue management problem consid-ering competition. We obtain an optimal pricing policy for periodicrevisions in a non-cooperative duopoly market with perfect infor-mation (inventory levels are common knowledge at all times). Weobserved that there are price zones which depend upon the level ofinventory of each firm. When these levels are high, the Nash equilib-rium prices are fixed during the season and depend on the valuationthat consumers have for each product. We also studied the effect ofthe reservation prices variability on the optimal pricing policies.

2 - Revenue Management for Professional ServicesCornelia Schoen

For professional service providers (PSP), people skills are key as-sets. Work is typically project-driven, with high levels of client in-volvement. To remain competitive, a PSP has to match its resourceswith client needs. We present a MIP that combines revenue- andproject-management to determine for a given period 1. the projectrequests to carry out along with their prices; 2. how, by whom,and when to execute key activities, considering the impact on clientvalue; 3. the composition of the workforce with distinct capabilities;to maximize profit. A case study shows the applicability.

3 - How Firms Succeed: A Longitudinal Study of Re-source Allocations in Firms from their Birth on-wardsA. D. Amar

From the patterns of behavior of more than fifty-seven firms drawnfrom a period starting from their birth to the first fifteen years oftheir lives, this paper reports results on how firms succeed in grow-ing their revenues. The results are derived by relating success tothe allocation of resources to four major revenue enhancing activ-ities: R&D; selling, general and administration; capital expenses;and property, plant and equipment. Strategies flowing out of deci-sions that contributed to their success and can be transferred to otherfirms are related in the paper.


Finance: AuctionsStream: FinanceInvited sessionChair: Aurélie Casier

1 - Optimal Release and Empirical Analysis of OnlineAuctionsFredrik Odegaard, Martin Puterman

We consider a seller that wishes to sell a fixed number of items us-ing a sequence of, possibly overlapping, single-item online auctions.The decision involves a trade-off between a holding cost and a higherexpected final price the fewer the number of simultaneous auctions.The problem is formulated as a discrete time Markov Decision Prob-lem, and conditions to ensure a control limit policy is optimal arediscussed. An empirical analysis based 5000+ ebay auctions fromone of the largest consumer electronics firms is also included.

2 - About the combination of multi-attribute offers inan English auction context.Aurélie Casier, Yves De Smet

We consider Combinations of Multi-Attribute offers in English Auc-tions. This combination can be performed either by the auctioneer atthe end of the auction either by the bidders during the auction itself.We study the impacts of combining on these two designs through,among others, auctioneer’s and bidders’ expected revenue.

23


� TA-14Tuesday, 8:00 AM - 9:30 AM - Sandton Sun: Maroela North

DEA: Applications in transport

Stream: DEA and Performance ManagementInvited sessionChair: Nancy Eleria

1 - A hierarchical programming model for the evalua-tion of transit network configurations using DEAEsteve Codina, Mohamed Salim Guenatri

The design of transit transportation networks is carried out in prac-tice by engineers who take into account a big number of factors.Usually, the design alternatives are very distinct from each other andno binary variables design models can be used. The question is thento choose a solution. In this paper the concept of DEA is illustratedas a basis for the formulation of Nash game models problems whoselower level models are parametrized and congestion sensitive equi-librium transit assignment models.

2 - Measuring the efficiency of Brazil’s interstate buscompanies with DEAAntonio G.N. Novaes

The paper describes an application of DEA to measure the technicaland scale efficiencies of interstate bus transport in Brazil. A sampleof 129 companies is analyzed. Strong output asymmetry is observed,with 20% of the firms responsible for 82% of the production. Eightinputs and one output, represented by annual passenger-kilometers,are assumed. Since total weight flexibility may produce misleadingresults in DEA applications, weight restrictions were introduced intothe model. Finally, the benchmarking process of non-efficient firmsis discussed for some DMUs.

3 - Comparative Study on Performance Measurementof Philippine SeaportsNancy Eleria

The Philippines is an archipelago and because of her geography, wa-ter is a major means of transport. Seaports are constructed to servemany inter island vessels plying regularly from/to islands. At presentabout 114 seaports are located throughout the country. The authorwas motivated to make a study with the main objective of measuringtheir performances using five factors such as cargo and passengertraffic, number of ship call, berth length, and service time. The mainoutput is a comparative analysis of their efficiencies using data en-velopment analysis.

� TA-15Tuesday, 8:00 AM - 9:30 AM - Sandton Sun: Maroela South

DEA: Applications in industry

Stream: DEA and Performance ManagementInvited sessionChair: Babeth Isa Fernando

1 - Evaluation of Customer Impact on InnovationStrategies, Gulgun Kayakutlu

Information gathered through customer collaboration has an improv-ing importance in innovative strategies. Objective of this study is toevaluate the customer relation processes on in a variety of innova-tion strategies targeted by SMEs in food industry. Innovation factorsfor industry are investigated as a multi criteria decision problem anda framework is proposed by using ANP. The importance of the clus-tered criteria is ranked in five companies to determine the focus ofinnovative case strategies. This paper will be stimulating for theSMEs in designating the innovation strategies.

2 - Measuring technical efficiency of PhilippineGourmet Coffee Industry: An Envelopement andMultiplier DEA ModelBabeth Isa Fernando

The paper measures the technical efficiency scores of top 15Gourmet Coffee Companies in the Philippines between 2006 and2007 using a DEAfrontier model. Findings show there are 10 cof-fee shops that managed to utilize their three inputs efficiently whilethere are atleast 7 companies obtained the desired output level.

3 - A Methodology for Evaluating the Quality of State-sponsored Primary Health Care, João Neiva de Figueiredo, ,

A multi-stage DEA methodology for primary health care evaluationis developed. The methodology assesses municipal priorities at threelevels: basic health risk reduction and universal access; preemptiveand preventive community actions and individual treatment and re-covery; and different patient segments (e.g., by age group). An ap-plication in Brazil is presented.

� TA-16Tuesday, 8:00 AM - 9:30 AM - Sandton Sun: Jacaranda

Container Storage Planning

Stream: Container PortsInvited sessionChair: Akio Imai

1 - Classification of Container and Resources intoCenters

Due to current congestion crisis, ports take longer time to processcontainers with limited resources. Cost associated with waiting be-comes a major issue. Actually, different containers have differentcosts associated with waiting. Conversely, furniture, footwear, orplastic products can wait longer. It might be reasonable to establishdifferent berth centers (VIP centers and regular centers) to provideindividual services for classified containers. In our model, we simul-taneously classify containers and resources into berth centers. Ourgoal is to minimize the total waiting cost.

2 - Loading optimization of identical boxes into con-tainers via Tabu SearchVitoria Pureza, Guilherme Poli

The container loading problem consists in arranging orthogonallyand without overlapping, the maximum number of rectangular boxesinto containers. We present a tabu search approach for the casewhere the boxes are identical and available in large quantities. Aheuristic generates the initial loading pattern composed by blocks(boxes with the same orientation), followed by moves that increasethe size of selected blocks in the current pattern. As a result, otherblocks are decreased, eliminated or created, yielding solutions withhigh volume utilization. Computational results are reported.

3 - Container storage planning problem at the marineterminal which mega-containerships callEtsuko Nishimura, Akio Imai

Introducing the mega-containership with over 10,000 TEU capac-ity offers to the marine terminal operation more complicated, thismeans that more advanced operation technique is needed. In sucha terminal, a lot of containers arrive simultaneously. Thus in orderto control yard cranes and yard trailers effectively and smoothly, thedetermination of container arrangement on the yard space is very im-portant. Therefore, this study addresses the container storage plan-ning problem for transshipment with considering both ways betweenthe mega-containership and feeder ships through the yard space.

24


� TA-17Tuesday, 8:00 AM - 9:30 AM - Sandton Sun: Oleander

Software Demos I (GAMS/AMPL)

Stream: Software DemonstrationsContributed sessionChair: Robert Fourer

1 - Recent Enhancements in GAMSJan-Hendrik Jagla, Lutz Westermann

On the one hand the General Algebraic Modeling System (GAMS)protects user investments through its commitment to backward com-patibility and platform independence. On the other hand GAMSprovides continuous additions and improvements that allow users tobenefit from state of the art technologies. We will outline severalrecent enhancements of GAMS supporting efficient and productivedevelopment of optimization-based decision support applications.Among others we will present recently added solvers and enhancedinterfacing with other applications.

2 - Enhanced Solver Support in the AMPL ModelingLanguageRobert Fourer, David M. Gay

AMPL has set the standard for optimization modeling languages thathandle complex models naturally and large models efficiently. Thispresentation will provide a tour of AMPL’s current solver support,with emphasis on recent enhancements to AMPL’s facilities for op-timization problems of varied kinds. Topics will include CPLEX11’s new convex quadratic, solution pool, and parameter tuning fea-tures, KNITRO 5.2’s recognition of complementarity conditions,and LGO’s facilities for global optimization, as well as support foropen-source solvers from COIN-OR and other sources.

� TA-18Tuesday, 8:00 AM - 9:30 AM - Sandton Sun: Syringa

Telecommunications Trees and FlowsStream: TelecommunicationsInvited sessionChair: Michal Pioro

1 - Routing Optimization of Networks based on Multi-ple Spanning TreesDorabella Santos, Amaro de Sousa, Filipe Alvelos

This presentation focuses on routing optimization of telecommuni-cation networks. It addresses link load related optimization func-tions in the context of Metro Ethernet networks where demands arerouted based on multiple spanning trees. One compact mixed inte-ger programming model, and several models based on the Dantzig-Wolfe decomposition principle, which are solved by branch-and-price, are proposed and compared.

2 - Models for the Prize-collecting Steiner Tree Prob-lem with Node Degree Dependent CostsLuís Gouveia, Pedro Moura

The Prize-collecting Steiner Tree Problem (PCSTP) is a well-knownNP-hard problem with applications in the telecommunications area.We consider an extension including a concave piecewise non-linearpenalty function on every node, which depends on its degree. Weconsider several models based on discretization techniques and com-pare their linear programming bounds. Results taken from instanceswith up to 100 nodes are given.

3 - On Complexity of a Multicommodity Flow Problemwith Path DiversityArtur Tomaszewski, Michal Pioro

Multicommodity flows can be protected against link failures by as-signing flows to diverse paths. A related network design problemhas a non-compact linear formulation requiring in general difficultcolumn (path) generation. Therefore, the problem is "suspected" tobe NP-complete - a conjecture not demonstrated as yet. We provethe conjecture by providing a reduction from graph coloring, andexamine NP-complete variants of the problem.

� TA-19Tuesday, 8:00 AM - 9:30 AM - Sandton Sun: Acacia

Maintenance 2Stream: Maintenance and ReliabilityInvited sessionChair: Rose Baker

1 - Optimal Preventive Maintenance Policies with Im-perfect (or Perfect) RepairsErnie Love

Maintenance planning often requires determination of a cycle timebetween preventive replacements (PR). Random failures can occurand the system receives emergency imperfect repairs (ER) at thesetimes If an ER occurs between the start of a cycle and a second de-cision time (DT), then the planned PR would still be performed atthe end of the cycle. If the first ER occurs after DT, then the PR atthe end of the cycle is skipped over, and the next planned PR wouldtake place at the end of the subsequent cycle. Optimal policies withthese two decision variables are determined.

2 - Risk-averse maintenanceRose Baker

In risk-averse maintenance, we seek to maximise the expected utilityof a maintenance policy rather than minimising cost per unit time.This typically leads to more frequent maintenance, with fewer largeexpenditures. The methodology is exemplified, and the talk aims toshow how practitioners could use it to derive practical maintenancepolicies. The need to estimate model parameters from data modifiesthe optimum policy, such that it tends towards a minimax solutionas risk aversion increases. The talk concludes by discussing usefulareas for further work.

25

TB-01 IFORS 2008 - Sandton

Tuesday, 10:00 AM - 11:00AM

� TB-01Tuesday, 10:00 AM - 11:00 AM - Convention Centre: Ballroom1

Supply Chain Management

Stream: PlenariesPlenary sessionChair: Michel Gendreau

1 - The Role of Operations Research in Lean SupplyChainsH. Donald Ratliff

While the concept of "lean’ has its roots in manufacturing, there aresignificant efforts underway to extend these ideas across the broadersupply chain. The size and complexity of global supply chains makesome of the principles of lean manufacturing, such as continuousflow, extremely difficult to extend. However, other "lean’ principlessuch as variability management, waste elimination and continuousimprovement are critical to the design and execution of all supplychains. This session will illustrate why some of these lean princi-ples are so important, the role of technology in implementing theseprinciples, and the potential for Operations Research in closing cur-rent technology gaps associated with these principles.

Tuesday, 11:30 AM - 1:00 PM

� TC-01Tuesday, 11:30 AM - 1:00 PM - Convention Centre: Ballroom 1

Combinatorial Optimization: Data miningand clustering

Stream: Combinatorial OptimizationInvited sessionChair: Dorit Hochbaum

1 - A Data-Mining Linear Programming Model to Pre-dict Material Fatigue ParametersArmin Fügenschuh, Dorit Hochbaum, Chalid el Dsoki

Fatigue in material science is the structural damage of a materialsubjected to cyclic loading. Fatigue life is influenced by several fac-tors in complicated ways. We present a mathematical data-miningmodel using measured static material parameters and a database ofknown materials to predict unknown cyclic parameters describingthe material fatigue behaviour.

2 - Solutions to some open problems of the optimalstars clustering tree problemMichal Stern, Ephraim Korach

We solve some versions of the problem: given a complete graphG=(V,E) with weights on every edge and a collection S of subsets ofV, the aim is to find a spanning tree T with minimum weight wheresubsets in S induce a star in T. Possible applications are from the ar-eas of communication networks, database systems, key managementand combinatorial auctions.

3 - A ranking algorithm for academic journalsMachteld Strydom

Prompted by the success of the PageRank algorithm of Google weinvestigated a similar approach to rank academic journals using onlycitation data. The resulting algorithm determines an objective "Influ-ence factor" that correlates well with subjective rankings by experts.

4 - Efficient cut-based image segmentation tech-niquesDorit Hochbaum

Segmenting an image is to determine a partition to the salient fea-tures of the image. We describe a very efficient and best possiblepolynomial time algorithm for a convex variant of the problem. Thisalgorithm’s efficiency enables its use in an interactive set-up. It ismore efficient than most procedures based on spectral techniques,partitioning approaches or heuristic clustering. We then demonstratehow to apply the procedure for the purpose of de-blurring medicalimages and identifying structures hidden by noise.


Discrete Optimization

Stream: OptimizationContributed sessionChair: Zelda B. Zabinsky

1 - A new three index model and Branch&Cut algo-rithm for the Capacitated Vehicle Routing ProblemIrene Loiseau, Daniel Negrotto,

We present a new three index vehicle flow formulation for the Ca-pacitated Vehicle Routing Problem and a Branch&Cut algorithmthat we developed using this model. We obtained competitive com-putational results compared with the better ones found at the litera-ture. We were able to solve large number of instances to optimality.

26

IFORS 2008 - Sandton TC-05

2 - Additive bounds in Depth First Search algorithmsfor the ATSP, Jens Lysgaard

A commonly used search strategy in Branch and Bound is DepthFirst Search (DFS) because of its low memory requirements and rel-ative ease of implementation. We try to strengthen tolerance-basedDFS for the Asymmetric Traveling Salesman Problem by adaptingthe additive bounds from Fischetti et al. (1992)

3 - Continuous and Discrete Global Optimization us-ing Hit-and-RunZelda B. ZabinskyMany applications in engineering design and other fields require op-timization of non-convex functions with both continuous and dis-crete variables. We present optimization algorithms based on Hit-and-Run, with modified cooling schedules that are based on theoret-ical analyses.


Multicriteria Public SectorDecision-Making

Stream: OR and the Public SectorContributed sessionChair:

1 - Providing decision support through multi-criteriaclassification / sorting methodsRenee KoenMulti-criteria Decision Aid techniques are known to be useful inproviding decision support in various application areas. This paperhighlights practical aspects of the implementation of multi-criteriamethods for classification / sorting problems, as applied to the pro-vision of decision support in the public sector.

2 - Applications of Operational Research to Refine-ments of Administrative Inspections

In administrative inspections, enhancing accountabilities and im-proving transparencies would be crucial issues for auditing organi-zations at a time of budgetary restraint. In this study, we focusedon the contribution ratio of projects to a measure in the inspection,and formulated the quantifying scheme of the ratio in the inspectingprocess by applying the analytic hierarchy process. Then we showedthat the quantification of the ratio redounded the refinements of theadministrative inspection through actual cases of a local governmentin Japan.

3 - Preferential procurement in South Africa: howmany criteria?,

The South African Constitution provides for five criteria to be op-erative in public procurement: fairness, equitableness, transparency,competitiveness and cost-effectiveness. The application of these cri-teria in the Preferential Procurement Policy Framework Act environ-ment is a multi-criteria decision-making problem that is susceptibleto quantitative analysis. It is preferable to a legislative regime oper-ating exclusively on the basis of race.


Applications of OR in Forestry I

Stream: OR in Agriculture and Natural ResourcesInvited sessionChair:

1 - Yield Frontier Analysis of Forestry Standing Inven-toryStuart Mitchell

A Yield Frontier in Forestry is a description the standing inventorywithin a section of forest. Unlike other standing inventory descrip-tions that assume a unique outcome (volume by log-type) from har-vesting. A Yield Frontier describes the space of all possible out-comes. To determine a Yield Frontier an LP formulation of the buck-ing problem is used. The feasible region of this LP is then projectedinto the space of the volume by log-type variables. Numerical resultswill be shown for a 200 stem section of forest.

2 - Soft Systems Approach in Facilitating RegionalForest Policy ProcessesTeppo Hujala, Olli-Pekka Tikkanen, Annika Kangas, PekkaLeskinen, Leena Leskinen

In Finland, 13 Regional Forest Programmes (RFP) establish provin-cial objectives for e.g. loggings and preservation of biodiversity. TheRFPs are constructed through participatory planning, incorporatingkey stakeholder groups. This presentation introduces our project ondeveloping the legitimacy and effectiveness of RFP processes by us-ing soft systems methodology as a framework.

3 - A new method for robust planning in rolling horizonplanningMikael Rönnqvist, David Bredström

We describe a new method to solve LP problems with uncertain pa-rameters in rolling horizon planning. It is based on iteratively solvean upper level problem for the first period where the parameters areknown. The lower level problem uses the upper level solution tofind a worst case scenario for an anticipation period with uncertainparameters. Information in how the worst case scenario is affectedby the upper level decisions is given back as a series of valid in-equalities. An integrated production, transportation and inventoryplanning problem coupled to heating plants is used as a case.

4 - Sales and operations planning: issues for the for-est product industryYan Feng, , Robert Beauregard

This presentation addresses sales and operations planning in the for-est product industry. It discusses a series of mathematical programsused to plan sales, distribution, production, procurement. Thesemodels are integrated in different ways and embedded in a simu-lation framework. The performance varies according to the level ofintegration. A business case is used to illustrate the challenges ofsales and operations planning when market prices fluctuate, demandprocesses are contract based or spot based and availability of rawmaterial is constrained by seasonality.

� TC-05Tuesday, 11:30 AM - 1:00 PM - Convention Centre: Boardroom1

OR Education 2Stream: OR EducationInvited sessionChair: Fredrik Odegaard

1 - A Business Gaming Simulation in Innovation andTechnology ManagementElmar Kiesling, Markus Günther, Christian Stummer

Business games provide an excellent opportunity to gain valuablefirst-hand experiences in management fields such as innovation andtechnology management where (i) large amounts of resources areat stake and (ii) a considerable time lag lies between a manage-rial decision and its (perceivable) effects. Thus, we have developeda proper Internet-mediated business simulation for our Master stu-dents in Business Administration at the University of Vienna. In ourtalk, we will present key elements of the simulation and discuss ourexperiences in class.

27

TC-07 IFORS 2008 - Sandton

2 - Dynamic modelling of student throughputCJ Swanepoel

The throughput rate (defined as the number of students graduatingdivided by the number enrolled) at the University of South Africafalls significantly below the targets set by the government. A dy-namic model was developed to gain insight into the causes of lowsuccess and their interactions. We report our findings.

3 - Post-Graduate Students Selection Using AnalyticNetwork ProcessArmando Zeferino Milioni, Leila Nascimento, MischelCarmen N. Belderrain, Joana Ribeiro, Amanda Silva,

This paper use a Multi-criteria Decision Making method, Ana-lytic Network Process — ANP to support the student selection atMechanical Aeronautic Engineering post-graduation course of theAeronautics Technological Institute, located in São Paulo, Brazil.The criteria are rated and the alternatives are evaluated one by one,according with the rates determinate. This type of analysis modelis known as rating. The goal of this paper is to propose a vigorousselection model: applicable and helpful to professors during theircandidate evaluation process.

4 - Ruminations from a Rookie MS/OR ProfessorFredrik Odegaard

One aspect that separates OR/MS from other fields is the fact thatit is taught, and used, in Mathematics, Engineering, Social Science,and Business departments. The teaching style and learning objec-tive for these different groups are naturally different. In this talkI will share some of my rookie experiences and insights of teach-ing MS/OR to Business school students. The transformation froman Engineering trained OR student to a Business case MS Professorincluded many moments of doubt and frustration.


Healthcare policy analysis usingsimulationStream: OR and HealthcareInvited sessionChair: Boris Sobolev

1 - Surgeon resource allocation using logistic regres-sion and simulationMarie Persson, Jan Persson

One of the main management problems in a surgical clinic is to es-timate how the resources of surgeons should be allocated betweenpatient appointments and surgical cases. Patient appointments gen-erate elective surgical cases which makes them strongly intercon-nected. We use logistic regression to predict the surgical requirementand procedure time. Then we simulate the quantitative effects basedon these predictions with different approaches of management deci-sions of resource allocation. In the simulation model we incorporatean optimization model to represent the management decisions.

2 - A New Appointment Scheduling System in an Out-patient Clinic by Using the Queueing NetworkSeongmoon Kim, Sunju Park, Semin Sim

Patients visit some or all medical processes in a clinic in different or-ders. For each process, its arriving patients come from both outsidethe clinic system and the other medical processes. We use the Jack-son Network in queueing theory to model patient flow in the medicalprocesses and present a detailed analysis for patient waiting time.We conduct a case study from an orthopedic surgery department andshow that optimal scheduling of appointments will improve patientflow, decrease waiting time, and, eventually, increase customer sat-isfaction.


MCDA and Performance AssessmentStream: Decision AnalysisInvited sessionChair:

1 - Conduct DEA-Oriented Performance Analysis inthe Domain of MCDA,

In this paper, equivalence relationship between input-oriented DEAmodels and MOLP models is explored. It is proved that the input-oriented DEA dual models are identical to super-ideal point mini-max models in MOLP, both of which can be used for conventionalefficiency analysis. It is shown that all efficient solutions of a deci-sion making unit can be generated by solving a MOLP problem cor-responding to the super-ideal point minimax model. Both graphicaland analytical methods are investigated to generate data envelopesand efficient frontiers for integrated efficiency and trade-off analy-ses.

2 - Assessing the performances of decision makingunits in terms of their contributions to the wholesystem, Yingming Wang, Kwai-Sang Chin

In performance assessment practice, decision making units (DMUs)may sometimes form a system. For example, different departmentsmake up a university, whose performance will heavily rely on theperformances of different departments. As such, the performance ofa bank will depend upon the performances of its branches. In thispaper, we view DMUs as a system and evaluate their performancesfrom a systematic point of view. The DMUs will be assessed andranked in terms of their contributions to the whole system ratherthan their efficiencies as usual.

3 - A Study of the Decision Process Based on Multi-ple Types of Evaluation in Chinese Academy of Sci-encesXiaouxan Li, Guoliang YangIn this paper, the development process of the institute evaluationin Chinese Academy of Sciences (CAS) is reviewed and the prac-tice of the Comprehensive Quality Evaluation in 2005 is introduced,which includes several different types of quantitative and qualita-tive evaluations. In order to improve the decision process used bythe headquarters of CAS, some useful mathematical methods for thedecision process are introduced and the application of these mathe-matical methods is studied. A numerical simulation concerning theComprehensive Quality Evaluation in 2005 is reported.

4 - Prioritising Voices of Customer via Multiple CriteriaDecision Modelling and Evidential Reasoning, , Xinlian Xie, Anil Kumar Maddulapalli

This paper investigates a methodology to prioritise customer voicesby data mining, decision modelling and decision support throughcustomer satisfaction surveys. The methodology consists of anevidence-driven decision modelling framework to represent largedata sets and a reasoning-based decision support process to aggre-gate evidence. Rule-based functions are proposed for data transfor-mation and a generic method is investigated for ranking customervoices based on qualitative scales. A case study is examined to il-lustrate the application of the methodology for a leading car manu-facturer.


Urban Transportation

Stream: Urban TransportationInvited sessionChair:

28


Chair:

1 - Impact of Large Trucks on Signalized IntersectionOperationsScott Washburn, Carlos Cruz-Casas

A custom traffic simulation tool was developed to estimate the pas-senger car equivalency (PCE) value of three different sizes of com-mercial trucks at signalized intersections. Additionally, an equationwas developed to estimate the additional start-up lost time that oc-curs for large trucks at the front of the queue.

2 - Pareto-Improving Congestion Tolls for GeneralRoad Networks,

Although many have recognized it as an efficient method for mit-igating congestion, it is still a major obstacle to get the public toaccept congestion pricing. We propose a class of congestion pricingschemes called "Pareto-improving." When compared with the situ-ation without any pricing intervention, pricing schemes in this classensure that no user is worse off while reducing congestion, perhapsnot necessary to the minimum level possible. Because no one isworse off, it is our premise that Pareto-improving schemes are moreappealing.


Project Scheduling: Scheduling ofActivitiesStream: Operations ManagementContributed sessionChair: Boaz Golany

1 - A Heuristic Solution for Shorter Critical ChainProject DurationSheng-Hung Chang, Bo Liao

Critical Chain Project Management(CCPM)developed by Dr. Gol-dratt has recently emerged as one of the most successful approachesto project management. However, CCPM is still lack an effectivelyway to schedule project in order to obtaining shorter project dura-tion under resource-constrained. The aim of this paper is to pro-pose a heuristic solution called Two-Stage CC Scheduling(TSCCS),and compares with four CC scheduling methods: ProChain, PS8,Minimize Slack Time, and Critical Path First. The result showsthat TSCCS method is superior to the other scheduling methods inproject duration.

2 - Simulation and Sampling in Stochastic ActivityNetworksAbdelghani Elimam, Bajis Dodin

Monte Carlo Simulation (MCS) is used to analyze Stochastic Activ-ity Networks (SAN). Accuracy of the MCS results depends on thesample size. Sample size must consider the SAN characteristics in-cluding: number of activities, their distributions, number of paths,and the SAN structure. We investigate the impact of these character-istics on the sample size, and the simulated results accuracy.

3 - Decomposition Method for Project SchedulingProblems with Adjacent ResoursesJacob Jan Paulus, Johann Hurink

We present a decomposition method for a Project Scheduling Prob-lem with Adjacent Resources. The units of an adjacent resourceare ordered and tasks require a number of adjacent units (e.g. adry dock). The method first assigns the adjacent resources and thanschedules tasks. Test results demonstrate its applicability.

� TC-11Tuesday, 11:30 AM - 1:00 PM - Convention Centre: CommitteeRoom 2

Operations Management: Supply Chains

Stream: Operations ManagementInvited sessionChair: Mabel Chou

1 - Two-Sided Network Effects with Supply and Gen-eral Random Demand Considerations

This paper is a generalization of Parker and Van Alstyne (2005) tak-ing into considerations the supply aspect of a two-sided market, andwith random demand. With these considerations, we show that theresults in Parker and Van Alstyne (2005) can be generalized. Nu-merical results are given which provide insights into the theory de-veloped in the paper.

2 - Application of Conditional Value at Risk for Con-tract Costing ModelBerrak Dag, Paul Liston, Cathal Heavey

Conditional Value at Risk is one of the coherent risk measures in fi-nancial economics and it has started to be used in supply chain areato quantify risk. Our approach is to apply this measure in a contractcosting simulation model to find out the risk under different quantityflexibility contracts.

3 - A study on process flexibility in software supplychainMabel Chou,

We investigate the benefit of process flexibility in a software supplychain problem. We consider an n-product-n-plant model. Each planthas a limited capacity and the product demand is specially mod-eled to fit the special environment of a software supply chain. Ourstudy shows that the chain structure performs almost as well as thefull flexibility structure, which suggests that a chain structure is alsopreferred in the software supply chain problems. On the other hand,we find that the benefit of adding flexibility might not be as great asthat to a traditional supply chain.


Finance: Option pricing

Stream: FinanceInvited sessionChair: Antonio Cosma

1 - Pricing American Options Under Stochastic Volatil-ity and Stochastic Interest Rates, Marina Medvedeva

We introduce a new analytical approach to pricing American op-tions. We use an explicit proxy for the exercise rule, and derivetractable pricing formulas using a short-maturity asymptotic expan-sion. Numerical experiments show that the analytical approximationis fast and accurate. In the Black-Scholes case the approximation isas precise as a binomial tree with several hundred steps with thespeed of a 50-step tree. The main advantage of our approach lies inits application to a three-factor model with stochastic volatility andstochastic interest rates while keeping computational time low.

2 - Option Pricing under GARCH processes by PDEmethodsMichèle Breton, Javier de Frutos

29


We propose a numerical method for valuing American options un-der discrete time GARCH processes. The option value is expressedas a function of time and of three state variables, and is shown tosatisfy a partial differential equation with jump conditions. A fi-nite element method is proposed and numerical experiments are con-ducted. Results and convergence properties are compared with thoseobtained using lattice methods, dynamic programming and least-square Monte-Carlo.

3 - Fractal asset returns, arbitrage and option pricingPetrus H Potgieter

In the discrete-time fractional random walk model a market with onerisky asset affords an arbitrage opportunity as described by Cutlandet al. and Sottinen. We briefly discuss these results and compute anumerical example in a fractional binomial model as illustration andmention an option pricing model for assets the returns of which aredriven by a fractional Brownian motion.

4 - Option valuation by recursive wavelet liftingAntonio Cosma,

We develop a new simple and widely applicable method to priceBermudan options. Our method is based on the projection of theGreen function on a basis of wavelets. We allow for different spec-ifications for the process of the underlying asset, and in particular,for processes whose Green function is known only in the Fourier do-main, such as Lévy processes and stochastic volatility models. Weimplement the methodology to price Bermudan options, and showthrough simulations that it outperforms PDE-based approaches inall dimensions both in accuracy and speed.


Finance: Portfolio management

Stream: FinanceInvited sessionChair: Daniel Akume

1 - Selecting Portfolios for Mutual Funds Visually onDecision BallsLi-Ching Ma

The growth of investment in mutual funds raises the issue abouthow to assist investors with different preferences in selecting suit-able portfolios. This study proposes a visualization method to helpinvestors examine the ranks of and similarities among alternatives onthe surface of a sphere. Thus, investors can select portfolios visuallyand conveniently.

2 - Optimal VaR constrained hedging of fixed priceload following obligations in competitive electricitymarkets, Yumi Oum

Load serving entities providing electricity to regulated customershave an obligation to serve random load at fixed prices. Such doubleexposure to the wholesale spot price and volume results in a profitexposure that is quadratic in spot price and cannot be adequatelyhedged with forward contracts whose payoff is linear in price. Wedevelop a self-financed optimal VaR constrained profit hedging port-folio which is equivalent to an optimal mean-variance portfolio un-der particular distributional assumptions.

3 - Generating Interest Rate Scenarios for Fixed In-come Portfolio OptimizationHelgard Raubenheimer, Machiel Kruger

We estimate a model that fits the South African term structure ofinterest rates, using a Kalman filter approach. Our model includesfour latent factors and observable macro-economic variables. Theresulting model can be used to generate interest rate scenario treesthat are suitable for fixed income portfolio optimization.

4 - Risk-Constrained Dynamic Portfolio ManagementDaniel Akume

We consider a portfolio problem when a constraint in the form ofconditional value-at-Risk or expected loss is imposed. The financialmarket is composed of n risky assets driven by geometric Brown-ian motion plus one risk-free asset. The constraint is derived andimposed continuously. The method of Lagrange multipliers is com-bined with the Hamilton Jacibi Bellman equation to insert the con-straint into the resolution framework. A numerical method is appliedto obtain an approximate solution to the problem. We find that theintroduction of the constraint curbs investment in the risky assets.

� TC-14Tuesday, 11:30 AM - 1:00 PM - Sandton Sun: Maroela North


Stream: DEA and Performance ManagementInvited sessionChair: Rui Marques

1 - Multiplier Forms of VRS Radial DEA ModelsLifen Wu

We present non-convex linear fractional programming multiplierforms of the variable-returns-to-scale (VRS) DEA model. In thecase of constant-returns-to-scale, the level curve radiates from theoriginal point in both the input- and output-oriented models; whilein VRS case, the 0-level curve is not the horizontal axis (plane orsuper-plane) as in CRS models, or even not parallel to it.

2 - A multiplier bound approach to assess relative ef-ficiency in DEA without slacksInmaculada Sirvent, Nuria Ramón, José L. Ruiz

In most cases in practice, DEA models yield strictly positive val-ues for the optimal slacks when assessing the inefficient units. Inthis paper, we propose a new approach to deal with this problemthat uses weight restrictions. The two-step procedure we introduceguarantees strictly positive weights, and consequently, zero slacks.This approach does not need a priori information about substitutionsbetween inputs and outputs, and does not require the existence ofFDEFs on the frontier, as is the case of other existing approachesthat address this issue.

3 - Influential Observations in Frontier ModelsRui Marques, Kristof De Witte

This paper proposes a nonparametric model to measure the effi-ciency of utilities for which the production technology is unknown.It performs a specially tailored non-oriented model which accountsfor undesired outputs and heterogeneity in the sample. Althoughefficiency is estimated in a deterministic frontier approach, each po-tential outlier initially benefits from the doubt of not being an out-lier. We survey several outlier detection procedures. A simulated ex-ample demonstrates the usefulness of the approach developed. Themodel is applied to the Portuguese water utilities.

� TC-15Tuesday, 11:30 AM - 1:00 PM - Sandton Sun: Maroela South

Economic Development: StrategicPlanning and Management

Stream: OR and Economic DevelopmentContributed sessionChair:

1 - Relationship between the level of development anddimensionality of innovativeness

30


In the competitive environment, the firms try to innovate in all theiractivities, which creates new dimensions of innovativeness. We in-vestigated a dimensionality of innovativeness in one of the transitioncountries and compare it with the dimensionality of innovativenessin one of the innovative countries.

2 - Soft and Hard OR Support for Informal Shop Net-works (Spaza shops)Theodor Stewart, Jean-Marie SABWA

We use soft and hard O.R techniques in a systems analysis ofchains of small community shops (Spaza shops) in the Western Capeprovince of South Africa. SODA and SSM techniques provided dif-ferent perspectives. SODA helped to identify and structure multipleconflicting aspects of the business, while SSM provided views onwhat changes are culturally feasible and systematically desirable.Statistical analyses of surveys and product purchases are being usedfor computer simulations aimed at designing and testing stockingstrategies for Spaza shops.

3 - A Comparative Study of IT Diffusion Patterns in thePublic and Private Sectors of GhanaKweku Ewusi-Mensah

Information Technology(IT)is credited with significant increase inthe level of economic and human development in many countriesand regions of the world except, perhaps, sub-Saharan Africa. Thepaper examines the differences and similarities in the ways IT isadopted in the public and private sectors of Ghana’s economy anddetermines the bases for the differences. The problems associatedwith IT diffusion are also discussed together with some policy im-plications for policymakers in government and business.

4 - Identifying and Ranking Critical Success Factors ofthe Nine Challenges in Malaysian Vision 2020

By the year 2020, Malaysia aspires to become a fully developed na-tion. In order to realize this vision, Malaysia needs to address ninestrategic challenges identified by the former Malaysian Prime Min-ister Tun Dr. Mahathir bin Mohammad. The present paper intendsto identify and rank the critical success factors of those nine chal-lenges by applying Analytic Hierarchy Process. The findings areexpected to provide valuable guidelines to the Malaysian govern-ment in course of developing effective action plans to achieve vision2020.

� TC-16Tuesday, 11:30 AM - 1:00 PM - Sandton Sun: Jacaranda

Air Transportation

Stream: Air TransportationInvited sessionChair: Qing Ding

1 - Robust Crew Pairing Scheduling

Traditional crew pairing scheduling does not consider the possibil-ity of disruptions on the day of operations. We extend the tradi-tional model to allow creating schedules, which are more robust tosuch disruptions and changes of the environment. Furthermore wepresent an IP Column Generation method to solve the new model.

2 - An iterative approach to integrated aircraft routing,crew pairing and flight re-timing, David Ryan, Matthias Ehrgott

We consider the airline scheduling problems aircraft routing, crewpairing and flight re-timing. Aircraft and crew must be allocated toflights in a schedule in a cost minimal way while the departure timesof some flights are allowed to vary within a time window. We solvethe aircraft routing and crew pairing problem with an iterative ap-proach, alternately solving aircraft routing and crew pairing. Thisgenerates a series of low cost solutions that are also robust to disrup-tions. We additionally allow some flexibility for the departure timesand present results for real world data sets.

3 - Revenue Management in Aircraft Leasing BusinessQing Ding, Brian Rodrigues, Kwan Eng WEEIn this paper, we study a multi-period aircraft leasing problem. Ineach period, independent airline companies can lease aircraft froma leasing company to augment existing capacity, e.g., if needed forcontingencies such as demand peaks. Before the first period, theleasing company must decide on its lease capacity for the planninghorizon. For each period, the leasing company also decides rentalrates while each airline determines its lease quantity and fare pric-ing policy. Using a principal-agent model, we derive managerialinsights arising from the interactions of these decisions.

� TC-17Tuesday, 11:30 AM - 1:00 PM - Sandton Sun: Oleander

Software Demos II (AIMMS/MPL)

Stream: Software DemonstrationsContributed sessionChair: Bjarni KristjanssonChair: Robert Fourer

1 - Teaching OR and Solving Business Problems withAIMMSFrans de RooijAIMMS is the advanced modeling system for building optimization-based decision support applications.AIMMS can be used to teach OR very effectively: Students canfocus on problem analysis and try out various model formulationswithout having to learn a complicated language. The integrated vi-sualization enables interactive analysis and further model develop-ment.AIMMS can also be used to solve complex business problems, as wewill show with real-life cases. AIMMS models can be turned intofully-operational optimization applications, enabling business usersto experience the power of OR.

2 - Introducing New Release of MPL Modeling Systemfor Optimization with New and Enhanced FeaturesBjarni KristjanssonWe will be demonstrating the newest release of MPL, the fastest andmost scalable modeling language on the market today. The inno-vative OptiMax 2000 Component Library, which allows MPL mod-els to be easily embedded into end-user applications, has been aug-mented to include several new objects and methods. The speed andscalability of the model generation has been greatly enhanced, andwith new 64-bit machines capable of solving much larger modelsthan ever before. Several new solvers have been added and existingsolvers updated. Data access has been improved and now offers fullXML support.

� TC-18Tuesday, 11:30 AM - 1:00 PM - Sandton Sun: Syringa

Meta-heuristics for optimizingtelecommunications networksStream: TelecommunicationsInvited sessionChair: Mauricio Resende

1 - A simulated annealing approach for the replicaplacement problem in content delivery networksAndré Dahlmann, Rainer KolischA content delivery network represents a part of the internet and con-sists of servers (nodes) and data links (edges) between them. Weconsider the problem of allocating copies (replicas) of a single con-tent to servers such that the total costs of placement, update and de-livery are minimized subject to service level constraints (maximumlatency). A simulated annealing approach is suggested and experi-mentally tested.

31


2 - A GRASP for SS/TDMA with Contiguous AntennaBeamsFausto Pedro Garcia MarquezThe two means of connecting earth stations located in differentspot beams (SB): switching of traffic between different SB cov-erage areas on the satellite, called Traffic Scheduling in Satellite-Switched Time-Division-Multiple-Access,and; assuming a globalreceiver coverage on board the satellite, earth stations selectivelytransmitting on different TDMA carriers, on possibly different SB,to send traffic to other earth stations. A GRASP algorithm is pro-posed in this paper for solving the SS/TDMA problem. The resultsobtained have been compared with an adaptive station-grouping al-gorithm

3 - A memetic algorithm for optimizing routing in net-works using exponential flow splittingMauricio ResendeRouting protocols like OSPF work on a set of link weights set upto control network data flow. Finding link weights that minimizenetwork congestion for a given network topology and demand ma-trix is known as the NP-hard Weight Setting Problem (WSP). A newprotocol called Distributed Exponentially-weighted Flow Splitting,or DEFT, has been recently proposed by Xu, Chiang, and Rexford(2007). DEFT with real-valued weights was shown to outperformOSPF with integer weights and often obtains optimal congestion. Inthis talk, we propose a memetic algorithm for the integer WSP forDEFT.

� TC-19Tuesday, 11:30 AM - 1:00 PM - Sandton Sun: Acacia

Quality and Reliability

Stream: Quality and ReliabilityContributed sessionChair: Maximo Bosch

1 - Statistical Inference of a Reliability System withVarying Repair RateMwanga Alifas YekoReliability and availability estimation study has been made in thispaper. It is assumed that the repair time of a two unit cold standbysystem is assumed to be a generalised Erlangian. Besides obtainingexpressions for the system reliability, and steady state availability, anattempt is made in this paper to obtain a consistent asymptoticallynormal estimator and an asymptotic interval for these two measures.The Bayesian analysis is also studied to compare the results.

2 - Use of Wage Incentives in Quality ImprovementMethods as a Motivational Technique, Tolga Ezen

The purpose of this paper is to undertake and consider the impor-tance of wage incentives as a means of motivational factor in es-tablishing higher quality level for industrial employees. Consid-ering the importance of employees’ productivity and efficiency intoday’s economic system for any country and the obvious interrela-tions amongst the motivation, efficiency, productivity and inflationit seems necessary to investigate the nature of this interrelations forsome companies.

3 - Designing a comprehensive set of service qualityindicators for a large state bank.Maximo Bosch, Eduardo ContrerasTop management at a large state bank in Chile required a broad andcomprehensive set of service quality indicators. A model linkingproduction quality and perceived quality with customer behavior andbank results was hypothesized. Operational processes were modifiedso to register new information, and several qualitative and quantita-tive market research studies were conducted. This model was cali-brated using QFD matrices as well as structural equations and otherstatical models. The model made possible the definition of a com-prehensive, broad and non redundant set of quality indicators

Tuesday, 2:00 PM - 3:30 PM

� TD-01Tuesday, 2:00 PM - 3:30 PM - Convention Centre: Ballroom 1

Advanced Topics in CombinatorialOptimization

Stream: Combinatorial OptimizationInvited sessionChair: Paolo Toth

1 - Mathematical Models to Reconstruct PhylogeneticTreesJuan José Salazar González, Daniele Catanzaro, , RaffaelePesenti

A basic problem in molecular biology is to rebuild phylogenetictrees from a set of DNA or protein sequences. Among differentcriteria used for this purpose, the minimum evolution principle isan optimality based criterion aiming to rebuild phylogenetic treescharacterized by a minimal length. This problem is known to beNP-hard. This talk presents some mixed integer programming mod-els, and we also study possible cuts and lower bounds for the optimalvalue. So far, the number of sequences that can be involved in opti-mal phylogenetic reconstruction is still limited to 10.

2 - Graph transformations and pseudo-boolean opera-tionsDominique de Werra

Graph transformations were developed to decrease the stability num-ber of a graph or to reduce the number of vertices without chang-ing the stability number. They were inspired by a pseudobooleanformulation and justified afterwards with graph theoretical argu-ments.Connections with operations designed to merge vertices insome classes of perfect graphs without changing the chromatic num-ber have been discovered. We examine reductions applied to adja-cent vertices having complete links between their proper neighbors .Connections with procedures for even pairs of vertices are exhibited.

3 - Heuristic and Exact Algorithms for the Bin PackingProblem with ConflictsPaolo Toth, Albert E. Fernandes Muritiba, Manuel Iori,Enrico Malaguti

We address a variant of the Bin Packing Problem, in which somecouples of items are incompatible and cannot be assigned to thesame bin. The problem, known as the "Bin Packing Problem withConflicts’, is of practical and theoretical interest, because of itsmany real-world applications and because it generalizes both theBin Packing Problem and the Vertex Coloring Problem. We pro-pose new bounding procedures, a metaheuristic approach and an ex-act algorithm. Extensive computational experiments on benchmarkinstances from the literature are reported.


Optimization Heuristics andApplications

Stream: OptimizationContributed sessionChair: Machteld Fick

1 - Using Immune algorithms to solve Assembly Plan-ning Problem

32

IFORS 2008 - Sandton TD-04

Assembly planning (AP) needs to take constraint factors such as thegeometric characteristics into consideration to work out a specificassembly sequence. When the constraints are complicated in AP,previous approaches such as guided genetic algorithms (GGAs) andmemetic algorithms (MAs) have succeeded to overcome the prob-lem. In the study, Immune Algorithms (IAs) were recommended tosolve the AP problem. Practical examples have demonstrated thefeasibility. Compared with GGAs and MAs, IAs can generate thesame or better solutions in terms of quality and searching time thanthese two methods.

2 - The logistic districting problem: a comparison ofdistinct metaheuristics approachesArnaldo Vallim, Daniel Inhetvin,

A real-world large-scale logistic districting problem was solvedthrough four different metaheuristics. An experiment compared:Tabu Search (TS), Simulated Annealing (SA), GRASP and GeneticAlgorithm (GA), permitting to highlight the strong points of eachtechnique. TS demonstrated efficiency having a very short process-ing time, while SA and GA showed great ability to escape from lo-cal optimum, extending its search to unexplored areas. On the otherhand, GRASP allowed to explore local search in a deeply way. Thepaper presents a discussion of all these results.

3 - Analysis of Soft Computing Technique to Minimizethe Local Minima Problem in Pattern Recognitionfor Hand Written English AlphabetsManu Pratap Singh

The back-propagation algorithm suffers with the local minima errorsurface for a large set of problems. A local minimum is defined asa point such that all points in a neighborhood have an error valuegreater than or equal to the error value in that point. This paperdescribes the soft computing techniques for the performance evalu-ation of Back-propagation algorithm to recognize the hand writtenEnglish alphabets. The results of the experiments and result showsthat the conventional back-propagation algorithm does not suites tosolve the challenging problem most reliably and efficiently.

4 - A comparison of heuristic and structure-based ap-proaches to hyphenationMachteld Fick, CJ Swanepoel

Automatic hyphenation techniques for English text have alreadybeen developed to a satisfactory level. For languages in developingcountries this is however not the case. Heuristic and structure-basedhyphenation techniques will be discussed and an empirical compar-ison of two representative techniques as applied to an agglutinativelanguage will be given.


Complex Public Sector Social Problems

Stream: OR and the Public SectorContributed sessionChair: Jan Greben

1 - A quantitative method for the management of mu-nicipal service delivery in an environment of userdissatisfactionKris Adendorff

A stochastic model for the investigation of consumer complaints ispresented for auditing a municipal service system in a responsiblefashion. The data which relates to the daily flow and accumulation ofcomplaints is manipulated iteratively to create an unflawed constantcause system which results in the improvement and maximisation ofquality of service delivery.

2 - Urban Dynamics and the influence of CrimeJan Greben

One of the problems in using coupled differential equations in dy-namic systems modelling is the unrealistic growth and decline ofindividual components of the system. For example, safe communi-ties would grow without bound, while unsafe communities would dothe opposite. In order to avoid this situation we combine non-linearanalytic solutions, which take properly account of the environmentallimitations in each community, with the numerical treatment of thelinear crime-induced dynamics between communities.

3 - Urban Real Estate Appraisal for Municipal LandTax Administration: A Methodological Approach toMake a General Land Values MapJackson Savio Vasconcelos Silva

The present paper aims at contributing to the improvement of urbanplanning and management by developing a methodology for makinga general land values map in order to support the implementationof the municipal land tax administration. This methodology usesmultiple regression analysis, including transforms, and cluster anal-ysis. The cluster analysis makes use of geo-processing techniques,krigage and self-organizing Kohonen maps. Finally, some sugges-tions of how public managers can use a general land values mapsegmented for urban administration purposes are presented.


Wine Supply Chain

Stream: OR in Agriculture and Natural ResourcesInvited sessionChair: Esbeth van Dyk

1 - Scheduling and mathematical modelling in a com-plex wine supply network: a case study, Simon Dunstall

We describe a holistic approach to operations management for alarge wine producer. Interdependency between vineyard monitor-ing, harvesting, transport and wine production motivates new sys-tems integrating human decision-makers, statistics, scheduling andadvanced IT backbones. CSIRO Australia has worked with indus-try partners to build such a system. We give an overview of thework and then focus on scheduling models and algorithms that con-sider winemaking rules and the wine-quality implications of sched-ules alongside more familiar constraints on harvesting, transport andprocessing resources.

2 - Improving the Order Management Process of aWinery using a Simulation ModelSergio Maturana, Alejandro Sanchez

The wine industry faces a highly competitive and globalized mar-ket, where it is imperative to deliver orders to large customers asfast as possible. However, to do it without errors and on time, is noteasy. The wine must be bottled, labeled, put in boxes, and sent to theport before the ship arrives. Since a winery faces many uncertaintieswhen committing to a delivery date, a simulation model of the ordermanagement process of a large Chilean winery was developed, totry different strategies to help reduce the probability of not deliver aperfect order (right products and on time).

3 - Decision Support System for a Wine CellarRob van der Mei, Esbeth van Dyk, Jan van Vuuren

The authors report on the development of a decision support sys-tem to assist management at a wine cellar in three areas of expertise,namely assigning grapes from the different suppliers to the differ-ent tipping bins, rapid rescheduling thereof in case of unforeseencircumstances and evaluating the layout of the cellar.

33

TD-05 IFORS 2008 - Sandton

� TD-05Tuesday, 2:00 PM - 3:30 PM - Convention Centre: Boardroom 1

Bringing the real world of OR into theclassroom 1Stream: OR EducationPanel sessionChair: Peter BellChair: Graham RandChair: Fredrik Odegaard

1 - Workshop 1Peter Bell, Fredrik Odegaard, Graham Rand

This workshop will showcase a number of different teaching vehi-cles that can bring the real-world practice of OR into the classroom.We will demonstrate and discuss ways by which projects, cases,practice readings, and a "strategic OR’ perspective can introduce stu-dents to the practical side of the use of OR models. We will focuson how to provide students with the real-world experience of usingOR models to solve real-world problems using Excel and VBA.

(This workshop will continue in another, immediately subsequentsession.)


Efficient use of core hospital resources

Stream: OR and HealthcareInvited sessionChair: Erwin Hans

1 - Nurse reschedulingAlistair Clark

Nurse staffing schedules in hospitals need to ensure adequate cover.If rescheduling arises due to staff illness, etc, then changes shouldbe kept to a minimum to avoid disruption. This talk will presentmodels and methods for nurse rescheduling, along with initial com-putational tests results.

2 - Scheduling radiotherapy patients at a cancer cen-tre, Dobrila Petrovic, Colin R. Reeves

A radiotherapy patient scheduling problem at a cancer centre is ad-dressed. Patients wait for long times for their first radiation dose.They do not necessarily follow the same regimen. The objectivesare to minimise the mean flowtime, mean tardiness and number oftardy patients. The problem is divided into four scheduling subprob-lems for the treatment phases: planning, outlining, pretreatment andtreatment. Various dispatching rules are applied to the subproblemsseparately. A heuristic is proposed to combine and coordinate thesubschedules into one and tested using real life data.

3 - Improving Primary Care Access Using Optimiza-tion Methods

Primary care providers provide the majority of care patients receiveduring their lifetime. A physician’s panel consists of a set of pa-tients from one of many health-related categories. We consider theproblem of determining the size and composition of physician pan-els in primary care. Using real data collected at the Mayo Clinicin Rochester, Minnesota, we formulate a Stochastic Linear Programthat seeks to maximize patient visits to their own providers, reducewaiting times, and minimize overage. We compare these results tothose from a meta-heuristic simulation optimization model.


MCDA/SMAAStream: Decision AnalysisInvited sessionChair: Pekka Salminen

1 - Descriptive decision analysis using SMAAIan DurbachI evaluate the use of stochastic multicriteria acceptability analysis ina descriptive decision model using a series of simulation and real-world longitudinal panel studies. Some implications of the goodperformance of the SMAA model for descriptive decision researchare discussed.

2 - Risk-based classification system of nanomaterialsJosé Rui Figueira,Nanotechnology is a rapidly growing field that is expected to havea large impact on consumer products. Potential toxicity of nano-materials and resulting risks at different stages of product life cyclehas recently attracted significant attention of regulatory agencies andstakeholders. To guide nanomaterial research and application as wellas its safe use, we propose a decision support system for classifyingnanomaterials into different risk categories. The stochastic multicri-teria acceptability analysis (SMAA-TRI) sorting method is used asthe foundation for this task.

3 - The shape of the value function in multicriteria ac-ceptability analysisRisto Lahdelma, Pekka SalminenWe test how the assumed shape of the utility or value function affectsthe results of multicriteria decision analysis. We study a real-lifeproblem and a large number of randomly generated test problems ofdifferent size. We apply SMAA method, which allows representinguncertain and partially missing criteria and preference informationas probability distributions. The results indicate that in most casesslight risk aversion does not significantly affect the results. The pro-posed method can be used for detecting real-life decision problemthat are robust for the shape of the value function.


Heuristic Approaches to Routing

Stream: RoutingInvited sessionChair: Javier Faulin

1 - A Study on Improving Scatter Search Methods Ap-plied to the Problem of Vehicle Routing with Simul-taneous Delivery and PickupGladys Maquera, , Annibal Parracho Sant?AnnaThis work investigates methods of improving the Scatter Search (SS)metaheuristic in applications to the Vehicle Routing with Simultane-ous Delivery and Pickup (VRSDP) problem. SS is an evolutionarymethod based on combining solutions to generate improved solu-tions. Computational experiments with data available in the litera-ture of VRSDP show that SS is robust and competitive in terms ofquality of the solutions generated and of computational time.

2 - A Hybrid Solution Approach for Ready-Mixed Con-crete DeliveryVerena Schmid, Karl Doerner, , Martin SavelsberghCompanies in the concrete industry are facing the following problemon a daily basis: concrete produced at several plants has to be deliv-ered at customers’ construction sites using a heterogeneous fleet ofvehicles in a timely, but cost-effective manner. As the ordered quan-tity of concrete typically exceeds the capacity of a single vehicleseveral deliveries need to be scheduled in order to fulfill an order.Our solution approach effectively integrates optimization and meta-heuristic techniques. The integrated approach is far more effectivethan every component applied solely.

34

IFORS 2008 - Sandton TD-14

3 - Inventory Routing for Environments with Stochas-tic Product UsageVera Hemmelmayr, Karl Doerner, , Martin Savelsbergh

In this talk we consider a problem of periodic blood delivery, wherea central blood bank has to manage the supply of blood to severalhospitals so that they do not face product shortages. Uncertaintyin demand is taken into account. Two solution procedures were de-veloped, based on integer programming and variable neighbourhoodsearch.

4 - SimuRoute: A Simulation-Based Algorithm for theCapacitated Vehicle Routing ProblemJavier Faulin, Angel A. Juan, Ruben Ruiz, Miquel Gilibert

The Capacitated Vehicle Routing Problem (CVRP) has been deeplystudied during the last decades, and different metaheuristics —suchas genetic algorithms or taboo search— have been developed. Mostof these metaheuristics assume that the CVRP conditions are static.We present an algorithm, SimuRoute, which uses a simulation-basedapproach to provide alternative solutions to the ones obtained usingtypical metaheuristics. These alternative solutions can be very usefulfor managers because they allow them to consider multiple criteriawhen performing the final decision.

� TD-11Tuesday, 2:00 PM - 3:30 PM - Convention Centre: CommitteeRoom 2

Supply Chain Optimization

Stream: Supply chainsContributed sessionChair: Tibor Kis

1 - Optimization of Integrated Forward and ReverseSupply ChainMangesh Gharote

Worldwide, there is an increasing trend of producers accepting theused products back from the consumer, due to various reasons. Inliterature many forward and reverse supply chain models are sepa-rately modeled, focus on limited modes of reusing returned products,solution methodology is heuristics based. We propose an optimiza-tion model for production-distribution-inventory planning, for inte-grated forward and reverse supply chain based on linear program-ming. Using examples it is shown that optimal handling of returnedproducts significantly reduces total cost of operation.

2 - A minimal cost solution technique for a single-vendor single-buyer integrated inventory modelunder controllable lead time

This paper develops a deterministic single-vendor single-buyer inte-grated inventory model under controllable lead time. The productionflow is synchronized by transferring the lot in equal and/or unequalsized shipments. A number of properties that lead to the minimalcost solution to the model is presented and illustrated with numeri-cal examples.

3 - The application of supply chains in the electronicdata environmentPeter Schmitz

In the information age information in an electronic format (data) istraded as a commodity. To create these data sets, data is sourcedfrom various suppliers, turned into new data and sold to customers.This paper will discuss how the various supply chain componentssuch as logistics and materials apply.

4 - Application of Optimization Models in SupplyChain Decision MakingTibor Kis, Marija Cileg

The objective of this paper is the presentation of an application ofoptimization models. It comprises the possible implementation ofdeveloped optimization models and the generation of economic pa-rameters necessary for concrete realization of several partial tasks.Different supply-, production- and distribution programming mod-els will be analyzed of diverse linear and non linear types. In thiscontext we present modifications of the goal programming model,determination of intervals of validity of dual prices and point out thenecessary connectivity between different models.

� TD-13Tuesday, 2:00 PM - 3:30 PM - Convention Centre: CommitteeRoom 5

Finance: Risk analysis and management

Stream: FinanceInvited sessionChair: Jonathan Crook

1 - Two-Stage Financial Risk Tolerance AssessmentUsing DEAJoseph Paradi, Angela Tran

"Know Your Client’ rules require financial advisors to consider aclient’s personal circumstances, wealth, investment objectives andrisk tolerance. Financial services are offered over the Web; howto assess risk tolerance if investors do not interact with advisors asthey use an online service? Our novel tool measures relative risktolerance using Data Envelopment Analysis. Distinct risk elements:propensity, attitude, capacity, knowledge and time horizon — are as-sessed with a questionnaire. Overall risk tolerance score between 0(risk averse) and 1 (risk seeking) is then computed using DEA.

2 - Credit Scoring with Macroeconomic Variables Us-ing Survival AnalysisJonathan Crook, Anthony Bellotti

We report an application of survival analysis to model default on alarge data set of credit card accounts. We explore the hypothesisthat probability of default is affected by general conditions in theeconomy over time. We show that inclusion of these indicators im-proves model fit and affects probability of default yielding a modestimprovement in predictions of default on an independent test set.

� TD-14Tuesday, 2:00 PM - 3:30 PM - Sandton Sun: Maroela North


Stream: DEA and Performance ManagementInvited sessionChair: Jesús T. Pastor

1 - CHE method and NACHE as alternatives to DEAHeinz Ahn

The German Centre of Higher Education uses the so-called CHEmethod to assess, e.g., departments of business administration. Thehigh importance of this topic gave reason to analyse the CHE methodfrom the point of view of decision theory. This analysis reveals cru-cial methodical weaknesses. Against this background, the NACHEwill be discussed as a standardised-additive version of the CHEmethod. It can be shown that the NACHE is a methodical com-promise between the CHE method and DEA.

2 - Hedge Funds Performance Appraisal: A compari-son of different nonparametric approachesNicolas Nalpas, Anne Vanhems, Laurent Germain

35

TD-15 IFORS 2008 - Sandton

Hedge Funds selection process is becoming more and more chal-lenging due to their low correlation with major asset classes togetherwith their ability to provide significant positive returns. The aim ofthis article is to assess the reliability of several popular nonparamet-ric methods including the robust m-frontier method and recent DEAmodels in their hedge funds ranking. The empirical analysis is con-ducted using CISDM Hedge Funds database that contains more than4,000 hedge funds. We analyze their performance using monthlydata over the period January 1995 to December 2004.

3 - Efficiency in relational models: DEA approachmahta yekkalam

Abstract Traditional studies of, widely recognized technique, DEAconsider relational system as black box while measuring the ef-ficiency. Chiang Kao(European journal of operational research2007)builds a relational network DEA model consider the interre-lationship of processes within the system to measure the efficiencyof the system and those of the processes at the same time. The ef-ficiency of the system is decomposed in to the efficiencies of thestages in series and parallel. In this study we develop a new modelfor parallel systems which is more flexible.

� TD-15Tuesday, 2:00 PM - 3:30 PM - Sandton Sun: Maroela South

Economic Development: General Topics

Stream: OR and Economic DevelopmentContributed sessionChair: Dave Evans

1 - Hierarchical Ordering of Itaipu Dam ExtensometersReadingsMaria Teresinha Arns Steiner, Rosangela Villwock, PedroSteiner Neto, Andrea Sell Dyminski

An adequate instrumentation system capable of monitoring lifetimebehavior of the structures of a dam is very important to follow upthe dam safety. Data Mining techniques were used to present a de-tailed analysis of the instrumentation data of Itaipu Dam, one of thebiggest dams in the world.

2 - A systems approach to indicators of poverty reduc-tionIsabel Meyer, Mario Marais

Complex problems such as poverty occur as the result of interac-tions that increase the complexity of resolving the problem, but alsoallows measurement of the inherent ’fitness’ of the system. Thispaper explores approaches to the measurement of the inherent ca-pacity of a system, and not just of individual elements thereof, toperform. Focusing on poverty reduction, concepts are developedthat shift the focus from measuring outcome indicators to measuringelements that can sustainably impact the performance of a systemover time.

3 - Promoting OR in India: Challenges and Opportuni-tiesRavichandran Narasimhan

The Indian economy is growing at a GDP growth of 9% pa. Thishas created new opportunities for using formal methods in resolvingmanagerial problems. The use of these methods in the business andsocial context in India is somewhat limited but there are outstand-ing institutions and OR researchers in India who can assist. In thispaper, we review the state of OR in India and briefly describe someof the initiatives taken by the ORSI Ahmedabad Chapter to promoteOR in India. We conclude this paper with a broad set of approachesto create a vibrant community of OR practitioners in the country.

4 - The Role of Operations Research in the DevelopingWorldDave Evans

"Development’ is a term which is now well known in the global vo-cabulary, but the activity is only about fifty years old. The characterof interventions varies enormously, and what makes for successfuldevelopment in one context may not work elsewhere. It is still un-clear what constitutes a successful development model, and whethereach situation is so distinctive, that lessons from other superficiallysimilar projects may not be applicable. The paper will discuss theopportunities for OR, ranging from the classic ’hard’ approach tomore recent developments in soft systems areas.

� TD-16Tuesday, 2:00 PM - 3:30 PM - Sandton Sun: Jacaranda

Maritime logistics

Stream: Freight Transport & LogisticsContributed sessionChair: Ricardo Giesen

1 - An analysis of cabotage transport navigation inBrazilNEWTON NARCISO PEREIRAThis paper aims to identify the factors that stimulate technologicalevolution in maritime cabotage transport, besides analyzing the en-try of new ships in the market of cabotage in Brazil from the periodof 1970 to 1980, given the lack of more updated data. In this con-text, this paper has also a focus in the analyses more recent in thefactors which have stimulated the uses of cabotage in Brazil. Thereare big demands for this modal in several kind of cargoes in this as-pects also are analyzed to quantified your in influence in the increaseof demand in cabotage Brazilian transport.

2 - Optimisation of an empty container relocation net-workWillemiena Bean, Johan JoubertA major problem that is experienced with the transportation of goodsis the relocation of reusable packaging. Containers are sent to cus-tomers to meet the demand but the return of the empty containersis not efficient, this leads to an unbalanced system. To address thisproblem a stochastic mathematical model that can be used as a toolin empty container management is presented. A deterministic math-ematical model that can be used to deal with this problem is alsodeveloped and the two models compared against each other.

3 - An Efficient Solution Method for the Multi-Item Mar-itime Inventory Routing ProblemRicardo Giesen, Juan Carlos Muñoz, Victor ValdiviaWe consider a Multi-Item Maritime Inventory Routing Problem(MMIRP) in which a set of products must be repeatedly transportedfrom a set of production-ports to a set of consumption-terminals bya heterogeneous fleet of multi-compartment vessels. We present asolution method that allows solving real-size instances, which out-performs experienced personnel manual-plans.

� TD-19Tuesday, 2:00 PM - 3:30 PM - Sandton Sun: Acacia

Quality and Availability

Stream: Quality and ReliabilityContributed sessionChair: Philimon Nyamugure

1 - Weekly availability predictions for Catalyst Prepa-rationMelanie FourieThe fuels value chain has to plan throughput according to productionunit availability. The Catalyst Preparation unit is an integral part ofthis value chain and its availability is critical. Is the quick and dirtymethod of forecasting the week’s availability sufficient or is there abetter solution? Let’s find out.

36

IFORS 2008 - Sandton TE-03

2 - Holistic Application of Capability indices in reduc-ing waste at Mutare Board and Paper MillsPhilimon Nyamugure

A company has been experiencing a lot of production losses overthe past years, recording about 17% waste generation per every 1500tones of saleable newsprint every month. Process capability analy-sis was used to check if the process is capable of producing productsthat conform to the specifications. The holistic study revealed thatthe production process was not capable. This gave enough evidenceto conclude that waste generation was due to production of noncon-forming products.


� TE-01Tuesday, 4:00 PM - 5:30 PM - Convention Centre: Ballroom 1

Minimax Stochastic Optimization withApplications

Stream: OptimizationInvited sessionChair: Chung Piaw Teo

1 - Models for Minimax Stochastic Programs with RiskAversionKarthik NatarajanIn this talk, we develop models to solve minimax stochastic linearprograms with risk aversion and ambiguity in distributions. Re-sults are provided for two stage stochastic linear programs assum-ing mean-covariance information on the random objective and ran-dom right hand side. Semidefinite programming is used to solve thisproblem. Applications in facility location problems are discussed.This is based on joint work with Xuan Vinh Doan (MIT), DimitrisBertsimas (MIT) and Teo Chung-Piaw (NUS).

2 - Robust optimization applied to the aggregate pro-duction planning of a Brazilian sugarcane millingcompanyRafael Paiva, Reinaldo MorabitoThis paper presents a robust optimization counterpart of a mixed in-teger linear programming model that represents the aggregate pro-duction planning of a sugarcane milling company. Analyses aretaken in order to determine the trade-off between uncertainty pro-tection and optimality. Computational results using actual data andCPLEX software are reported.

3 - Product Line Selection with Interdependent Prod-uctsChung Piaw Teo, Karthik Natarajan,We propose a method to obtain choice probabilities for a discretechoice model, using only information on the means and covariancestructure of the utility functions. Our proposed model is tractableas a convex optimization model, and yet it does not have the inher-ent limitations nor the associated complexity in choice probabilityestimation for other models.Our choice model can lead to better design of the product line, andcan perform significantly better than the MNL model, especiallywhen the products share many common attributes.


Public Sector Transport andManagement

Stream: OR and the Public SectorContributed sessionChair: Rajluxmi Vaish Murthy

1 - Empowering lower income commuters by manag-ing service delivery through sophisticated tech-nologies.Alwyn MoolmanThis paper focus on the supply of organized public transport in de-veloping countries and how to overcome challenges with technol-ogy. Public transport is characterized by inflexibility, unpredictabil-ity, disconnectedness and safety, as well as competing with the au-tomobile. Low income people can not risk their jobs through un-predictable transport services and are willing to pay for more pre-dictable transport. We present an integrated automatic vehicle loca-tion and fare management system providing operators real time routeadherence information. It utilizes smartcards for safe transactions.

37

TE-04 IFORS 2008 - Sandton

2 - Evaluating the impact of government policies andregulations on M-Commerce in India: a system dy-namics modeling approachRajluxmi Vaish Murthy, Deepali sharma, Krishna SundarDiathaMobile Commerce (M-Commerce), a subset of e-commerce, in-cludes all e-commerce transactions carried out using mobile de-vice(s). Growth of M-Commerce is a function of interrelated stakeholders and appropriate regulatory and policy interventions. Sys-tem dynamics simulation models are developed to understand M-commerce growth facilitators in India, where M-Commerce is at itsinfancy. The models are used to analyze the impact of policy andregulatory interventions.


Agribusiness Supply Chain

Stream: OR in Agriculture and Natural ResourcesInvited sessionChair: Sergio Maturana

1 - A Business Intelligence approach to SCM Opti-mization in Biofuel ProductionMartin Marchetta, Fernanda Andrea Garcia, RaymundoForradellasBiofuel is an emergent alternative energy, very important to reducethe global worming and as a replacement for some products derivedfrom petroleum. Since raw materials, elaboration processes and lo-gistics depend on its type and production geographic region, it isnecessary to define the conditions for its efficient and economicallysustainable production. In this work we present an approach to thebiofuel SCM optimization, by applying Business Intelligence tech-nologies over a specially designed traceability information system.We also present the Mendoza (Argentina) case study.

2 - Modelling Animal Farming Systems as SupplyChainsFrancois GuerrinTraditionally considered as autonomous entities, farms are actuallyparts of complex supply chains. Addressing sustainability issuesneeds to account for this reality. My contribution will depict amodelling approach of the pig industry in the Reunion Island rep-resenting the pig production chain from the raw feedstuff suppliersto the pig products retailers, passing by the rearing enterprises. Thismodel aims at simulating both the material flows and the coordi-nation means among the stakeholders. It is based on a modellingontology where human action is explicitly represented.

3 - Contribution to the Competitivity Evaluation: AnApplication of the Spatial Equilibrium Model to theSoybean Production in the State of Tocantins inBrazil

In the production chain of soybeans in Brazil, a sizable part of thecost structure is due to transportation costs. This study proposes aspatial multimodal and temporal equilibrium model regarding theexports of the State of Tocantins in the Brazilian cerrado. The dataare treated by the GAMS/MINOS.

4 - A Production Planning Heuristic for an Export-oriented Wine ProducerSergio Maturana, Franz CarrilloProduction planning for wine producers is critical. They must beable to bottle, label, and ship wine orders to customers all over theworld, without delays or backorders. Bottling lines must be sched-uled to operate efficiently since bottling capacity is limited. How-ever production schedules are disrupted by late orders, which in-crease bottling line set-up times. We tested different heuristics todetermine which made the most efficient use of the bottling linesand met the clients’ orders in time. A simulation model of the pro-duction process was used to compare the different heuristics.

� TE-05Tuesday, 4:00 PM - 5:30 PM - Convention Centre: Boardroom 1

Bringing the real world of OR into theclassroom 2Stream: OR EducationPanel sessionChair: Peter BellChair: Graham RandChair: Fredrik Odegaard

1 - Workshop 2Peter Bell, Graham Rand, Fredrik Odegaard

(This is a continuation of a workshop from an immediately previoussession.)

This workshop will showcase a number of different teaching vehi-cles that can bring the real-world practice of OR into the classroom.We will demonstrate and discuss ways by which projects, cases,practice readings, and a "strategic OR’ perspective can introduce stu-dents to the practical side of the use of OR models. We will focuson how to provide students with the real-world experience of usingOR models to solve real-world problems using Excel and VBA.


Advanced Modeling Methods in HealthCareStream: OR and HealthcareInvited sessionChair: Kalyan Pasupathy

1 - Optimization in the Health System of the State ofParana: Flow of Patients and New Hierarchic Con-figurationsPedro Steiner Neto, Maria Teresinha Arns Steiner, CassiusTadeu Scarpin,

This paper presents a proposal for optimizing the health system inthe state of Paraná with respect to the flow of patients within thestate boundaries and also a new hierarchical configuration to dividethe state into regions. The results obtained were considered veryconsistent.

2 - The Home Care Routing ProblemJesper Larsen, Matias Sevel Rasmussen, Tor Justesen,Anders Dohn

In this presentation, we focus on a staff planning problem arisingin Denmark where elderly citizen receive home care from the lo-cal authorities. The objective is to plan schedules for care providersmaking the most out of existing personel. Each staff member hasskills and each citizen should preferably be visited by a small set ofcontact persons.

We formulate the problem using a set partitioning model, and de-scribe an implementation of an optimization-based system for de-termining the daily routes for the staff members based on columngeneration.

38


3 - Centralized vs. Distributed Pharmacy Configura-tion: A System Dynamics ApproachKalyan Pasupathy

Inpatient pharmacy is often the most profitable department in a hos-pital, responsible for pharmaceutical drugs that are utilized in thedelivery of care. Drug orders are placed with the pharmacy andthe drugs are required for medication at the point-of-care in nurs-ing units. One of the major strategic decisions is the dichotomy ofcentralized vs. distributed pharmacy centers. Distributed centersreduce transportation cost and delivery time of medication but at anincreased cost. This project proposes to identify the balance to attainoptimal operational efficiency using system dynamics.


Multiple Criteria Decision Analysis

Stream: Decision AnalysisInvited sessionChair: Jyrki Wallenius

1 - Decision Support by Interval SMART/SWINGAhti Salo, Jyri Mustajoki, Raimo P. HämäläinenA preference programming approach is used to accommodate pref-erential and informational imprecision by intervals in SMART andSWING weighting. Our generalization allows the reference attributein weighting to be any attribute and the decision maker can replywith intervals to the weight ratio questions. We describe the relatedWINPRE software.

2 - Bibliometric Analysis of Multiple Criteria DecisionMakingHannele Wallenius, Johanna Bragge, , Jyrki WalleniusWe have conducted a bibliometric study of MCDM/MAUT using theISI database covering years 1970 to 2007. We report how our fieldshave developed based on variations of the following key words: mul-tiple criteria, multiple attribute, multiple objective, goal program-ming, and vector optimization. To profile the knowledge domainswe use VantagePoint software. We complement our study with a co-citation analysis to identify the (invisible) scholarly communities ofresearch. Interesting results emerge.

3 - Visualization in the Multiple Objective Decision-Making FrameworkJyrki Wallenius,In this paper we describe various visualization techniques whichhave been used or which might be useful in the multiple objectivedecision-making framework. Several of the ideas originate fromstatistics, especially multivariate statistics. Some techniques aresimply for illustrating snapshots of a single solution or a set of so-lutions. Others are used as an essential part of the human-computerinterface.


Routing Applications

Stream: RoutingInvited sessionChair: Marcos Colebrook

1 - In-vehicle navigation using a time-dependent reli-able A* searchIoannis Kaparias, Michael BellA novel version of the A* algorithm for in-vehicle navigation sys-tems considering time-dependence and travel time reliability, is pre-sented. The algorithm’s objective is to minimise the amount of un-predictable delays along the route suggested to the driver. A tech-nique for calculating time-dependent travel time and reliability isdescribed.

2 - Topology-based Optimal Road Network Pruning: AMethod for Speeding up Shortest Path Computa-tionTuukka Puranen, Joni Brigatti, Pekka Hotokka, AnttiHallamäki

In this talk we present a new approach that can significantly speedup the solving of all-pairs shortest paths problem in road networks.The main idea is to reduce, once the set of target nodes is known,the number of road links that need to be considered. The suggestedmethod conducts an analysis of nodes in the graph based on its topo-logical characteristics. Computational results with real road net-works, using the Dijkstra and A* shortest path algorithms, demon-strate that speed up factors over 10 are possible, and that the methodcan be effectively combined with other approaches.

3 - Routing in protected areasJosé L. Pino, Ma Teresa Cáceres, Gerardo Valeiras, LuisaJurado

Scientific research, wilderness protection, preservation of speciesand genetic diversity (biodiversity), tourism and recreation, sustain-able use of resources, are some of objectives of protected areas.

Transportation infrastructure often has very significant impacts onprotected areas, even when its primary purpose is to allow accessby managers and researchers. Therefore, its design, routing andmanagement must be carefully planned. This paper shows the ap-plication of metaheuristics to develop a tool to help the managers ofDoñana National Park (Spain) when deciding a routing task.

4 - ERP Add-On Development for the Multi-item LotSizing Problem with Inventory CapacitiesMarcos Colebrook, Jose M Gutierrez, Beatriz Abdul-Jalbar,Joaquín Sicilia

Most companies schedule their operations for a finite planning hori-zon consisting of several weeks or months, which are referred to asperiods. In particular, the problem of determining the optimal pro-duction/replenishment quantities for multiple items at each period isaddressed. However, this problem becomes intractable when inven-tory capacities are introduced. Accordingly, we have developed aheuristic procedure which derives near optimal solutions in reason-able running times. This algorithm could be considered as an add-onmodule within an Enterprise Resource Planning (ERP) application.


Scheduling Applications 1

Stream: SchedulingInvited sessionChair: Thomas Sillekens

1 - On a mixed 0-1 separable nonlinear programmingapproach for water irrigation schedulingLaureano Fernando Escudero, Almiñana Marc, MercedesLandete, Juan Francisco Monge, Alejandro Rabasa, JoaquinSánchez-Soriano

We present a mixed 0–1 separable nonlinear approach for the opti-mization of the management of water resources for agricultural irri-gation usage on a daily basis. It considers the irrigation networktopography, water flow technical conditions and logistical opera-tion constraints. Computational experience on a large-scale real-lifeproblem is reported.

39

TE-13 IFORS 2008 - Sandton

2 - Batch Sequencing to Scheduling Problem in a Real-World Pipeline Network, ,

This paper addresses the problem of batches sequencing in a realpipeline network which involves 3 refineries,1 harbour and 5 dis-tribution centers,connected by 15 pipes.This scheduling problemis complex and demands much computational time.Thus it is de-composed in 3 blocks:assignment,sequencing and timing.This workpresents the sequencing block.The goal is to get a batch orderingthat implies in a reduction of time horizon, increases the pipeline userate,and attends the demand in the distribution centers.Seven meta-heuristics are tested and compared.Different objective functions areanalyzed.

3 - Aggregate Production Planning In The AutomotiveIndustry — A Practical ApproachThomas Sillekens, Achim Koberstein

Efficient usage of capacitated resources is crucial within the auto-motive industry. A mixed integer programming approach is intro-duced to solve an aggregate planning problem. It considers discretesteps in production capacities and workforce demands due to flexi-ble working hours organized in shiftgroups. Moreover worker fluc-tuation and regulatory requirements regarding changes in workforceand production capacity are considered.

� TE-13Tuesday, 4:00 PM - 5:30 PM - Convention Centre: CommitteeRoom 5

Finance: Markov models and Valuationmodelling

Stream: FinanceInvited sessionChair: Suleyman Ozekici

1 - Repeat-Sales modelling for predicting house pricesand constructing a housing indexL. Paul Fatti

The repeat-sales model lies at the heart of the Automated Valua-tion Modelling (AVM) system developed by Lightstone. This modelalso has the advantage that, in addition to the prediction of propertyprices, it can be used to construct a housing index.

2 - Portfolio Management in a Stochastic Market UsingHARA Utility FunctionsSuleyman Ozekici, Ethem Canakoglu

We consider the optimal portfolio selection problem where the in-vestor maximizes the expected utility of the terminal wealth in astochastic market. The utility function belongs to the HARA classand the market states change according to a Markov chain. Theproblem is solved using dynamic programming to obtain an explicitcharacterization of the optimal policy. We also discuss the stochasticstructure of the wealth process under the optimal policy. The return-risk frontiers of the terminal wealth are shown to have linear forms.Special cases are discussed with numerical illustrations.

� TE-14Tuesday, 4:00 PM - 5:30 PM - Sandton Sun: Maroela North

DEA Theory: Recent developments innonparametric frontier estimation

Stream: DEA and Performance ManagementInvited sessionChair: Abdelaati Daouia

1 - Frontier Estimation and Extreme Values TheoryJean-Pierre Florens, Abdelaati Daouia, Léopold SimarThe production frontier is the locus of the optimal combinations ofinputs and outputs. From a statistical point of view, it can be viewedas the upper surface of the support of a random vector under shapeconstraints. In this paper we investigate the problem of nonparamet-ric frontier estimation from an extreme-values theory perspective.This allows to revisit the popular FDH estimator, to derive asymp-totically Gaussian estimators and to provide useful asymptotic con-fidence bands of the production function.

2 - A moving window approach for nonparametric es-timation of extreme level curvesLaurent Gardes, Stephane Girard,We address the estimation of an extreme quantile from a Pareto-typedistribution when covariate information is available. In such a case,the extreme quantile is function of the covariate. A nonparametricestimator of this extreme level curve of the conditional distributionis introduced and studied both numerically and theoretically.

3 - Functional Convergence of Quantile-type Frontierswith Application to Parametric ApproximationsAbdelaati Daouia, Jean-Pierre Florens, Léopold SimarFlorens and Simar (2005) suggest a two-stage approach which triesto capture the shape of the cloud of points near its frontier by provid-ing parametric approximations of a nonparametric production fron-tier. We propose an alternative method using the nonparametricquantile-based frontiers introduced in Aragon et al(2005). Thesequantile-type frontiers have the superiority of being more robust toextremes. Our main result concerns the functional convergence ofthe nonparametric frontier process. Then we provide asymptoticnormality of the resulting estimators of the parametric approxima-tion.

� TE-16Tuesday, 4:00 PM - 5:30 PM - Sandton Sun: Jacaranda

Information technology in logistics

Stream: Freight Transport & LogisticsContributed sessionChair: Antti Hallamäki

1 - The Potential of GPS to Replace Self-Report Sur-veysPeter Stopher,GPS measurement of travel is growing as a method of data collec-tion, both for validating conventional travel surveys and also for theevaluation of travel behaviour change policies. This paper exploreshow GPS measurement of travel has been used, the necessary speci-fications of a GPS travel survey, and the potential of GPS to replaceconventional travel diaries. Data from recent GPS surveys are usedto explore the information provided, and compare it with diary sur-vey data. Issues of potential bias are examined together with theimplications of multi-day GPS data are explored.

2 - Tracking cellular telephones at a large event to pro-vide advisory services for spectatorsAntony Cooper, Peter Schmitz, ,We conducted an experiment to track the cellular telephones of asmall group of people as they moved to and from a large event, toassess the viability of using such tracking to provide the participantswith useful traffic information through their telephones. We reporton the results here.

3 - Feeder Transit Services: Choosing Between Fixedor Demand Responsive PolicyLuca QuadrifoglioWe develop analytical and simulation models to assist transit plan-ners when having to decide between a demand responsive andfixed-route operating policy as a feeder service. We derive thecritical customer demand densities, in the range of [0.3¸1.7] cus-tomers/mile2/hr, which represent switching points between the com-peting service policies.

40


4 - The benefits of vehicle routing sofware in real lifetransportation planningAntti Hallamäki, , Pekka Hotokka

In this talk we present several different case studies and a compar-ison of different vehicle routing software using real-life data fromFinnish companies. The results indicate that there exists huge po-tential for savings in transportation industry and municipal sector.

� TE-18Tuesday, 4:00 PM - 5:30 PM - Sandton Sun: Syringa

Wireless Telecommunications NetworksStream: TelecommunicationsContributed sessionChair: Long Tran-Thanh

1 - Frequency Management as a Factor in EconomicGrowth and Development - an Empirical Study, Giel Hattingh, Albert Helberg

The lack of fixed line infrastructure and access in developing coun-tries has reduced their opportunities. Mobile phones and cellularnetworks may alleviate some problems. This paper explores pos-sible links between frequency management strategies, transactioncosts and the deployment of telecommunications technology on eco-nomics welfare and growth.

2 - Cost-effective Designs of Fault-tolerant Radio Ac-cess Networks in the GSM SystemBo Chen, Xujin Chen

We are concerned with cost-effective design of fault-tolerant ra-dio access network in the global system for mobile communication(GSM): To deploy network links between base transceiver stations(BTSs) and base station controllers (BSCs), and to determine man-agerial capacities of the BSCs, so as to minimize the link and man-agement costs while ensuring error tolerance ability. Three Max-SNP-hard problems are studied to address the scalable and modu-lar feature of BSC capacities. Constant-factor approximation algo-rithms are proposed to solve these problems in cubic time.

3 - A multi-commodity model for power-preservingrouting in ad hoc mobile networksYskandar HAMAM

A mobile ad hoc network is a group of mobile nodes that dynami-cally exchange data without a fixed infrastructure. In this work theauthor presents a method for power conserving routing in such net-works. The routing problem is modelled as multi-commodity trans-shipment problem where the cost is the power consumption at themobile nodes. The routing is based on the power availability atnodes which are represented by links with limited capacity. Theproblem is then solved using linear programming. Tests are thenmade on randomly generated networks and the numerical results arepresented.

4 - A novel reliability based routing protocol for en-ergy aware communication in Wireless Sensor Net-worksLong Tran-Thanh,

In this paper new reliable packet forwarding schemes are proposedfor Wireless Sensor Networks (WSNs). The proposed schemes areoptimized with respect to minimal energy consumption and to en-sure long lifespan. Reliability is guaranteed by ensuring the proba-bility of correct packet reception on the Base Station (BS) will ex-ceed a predefined threshold. This cast reliable and energy awarepacket forwarding as a constrained optimization problem. This opti-mization is reduced to looking for shortest path in the correspondinggraph and using the Bellman Ford algorithm.

41

RA-02 IFORS 2008 - Sandton

Thursday, 8:00 AM - 9:30 AM

� RA-02Thursday, 8:00 AM - 9:30 AM - Convention Centre: Ballroom 2

Optimization Methods

Stream: OptimizationContributed sessionChair: Vlasta Kanková

1 - Robust Linear Models: A Comparison between aMILP and LMS ApproachGiel Hattingh, Hennie Kruger,

The area of robust modelling has attracted much research effort. Oneapproach is the simultaneous selection of regressors and detectionof outliers in linear regression. This paper considers a mathematicalprogramming approach and compares it empirically with the well-known LMS technique for robust regression.

2 - A multi-objective optimisation framework for opti-misation algorithm parameter derivationAntoine Dymond,

A multi-objective optimisation framework is proposed for optimi-sation algorithm parameter derivation. The method is applicable toall computational intelligence algorithms. The quality of a parame-ter set is judged against two objectives : function value and numberof function evaluations, over a given problem set. The method isillustrated using the particle swarm optimisation algorithm.

3 - Methodological Approaches to MultiobjectiveStochastic Linear ProgrammingSegun Adeyefa

The paper considers core ideas that makes up the burgeoning bodyof Multiobjective Stochastic Linear Programming (MOSLP) empha-sising the methodological view. The key idea is that a MultiobjectiveStochastic Linear Programming Problem can be framed into a deter-ministic context using some appropriate transformations. The paperweaves ideas from optimization, multicriteria analysis and probabil-ity theory in a synergetic way. It can be used as an introduction tothe interesting field of MOSLP, as a learning tool as well as a roadmap to further research directions.

4 - Multistage Stochastic Programs via Autoregres-sive Sequences, Stability and Empirical EstimatesVlasta Kanková

Multistage stochastic programs correspond to applications that canbe considered with respect to a finite "discrete" time interval andsimultaneously they can be decomposed to a system of parametric(one-stage) problems. Consequently to this approach, we can em-ploy the results achieved for one-stage case to obtain stability, em-pirical estimates properties, and approximate solutions schemes formultistage problems. To this end, we restrict to autoregressive "un-derlying" random sequences.


Transportation Systems: TrafficAssignment Modelling

Stream: Intelligent Transportation SystemsInvited sessionChair: Janny Leung

1 - A dynamic traffic assignment model of car and bustravel under road pricingC.O. Tong

A road-pricing scheme with the objective of diverting travelers fromthe private car mode to the bus mode is proposed. The plan is to setup some tolling points on roads used by buses and cars. Compared tobus only lanes it is more flexible because it allows the road capacitythat are allocated to buses to be more finely adjusted.

A case study highway network with six at-grade signalized junc-tions and four OD pairs is used. A cyclic bus line is operated. Asimulation-based dynamic traffic assignment model is developed todetermine the modal choice and route choice of the travelers.

2 - Dynamic Stochastic Transit Assignment Model Un-der Advanced Public Transportation InformationSystemAgachai Sumalee, William Lam

This paper proposes a stochastic dynamic transit assignment modelfor assessing the effects of advanced public transportation informa-tion system with the information on transit vehicle arrivals and seatcapacity availability. The model is applicable to a general networkand formulated as an equivalent fixed-point problem. A seat alloca-tion model is proposed to estimate the probabilities of getting seatsfor passengers already on-board or waiting to board. The explicitseating model allows a better differentiation of the in-vehicle con-gestion experienced by the sitting and standing passengers.

3 - Route Choice among Cross-boundary ContainerTruck DriversBecky P.Y. Loo

Ever since the Open Policy in 1979, there has been increasing in-tegration between Hong Kong and China. The subsequent rapidexport-oriented industrialization in Zhujiang Delta has given rise tomuch cross-boundary container truck traffic (XBCT). The volume ofXBCT rose from 1.50 to 4.68 million vehicles from 1992 to 2003.Hence, a new customs check-point, the Shenzhen Western Corri-dor (SWC), was built. With the opening of SWC, how would XBCTdrivers change their choice of customs check-point? What were theirmajor route choice considerations? The current paper seeks to ad-dress these questions.


Applications of OR in Forestry II

Stream: OR in Agriculture and Natural ResourcesInvited sessionChair: David Martell

1 - Parameter estimation in compound Weibull distri-butionsClara Simon de Blas, francisco josé cano-sevilla

We present a new method to estimate the parameter distributions incompound Gumbel and Weibull distributions. The model is appliedon a set of simulated data on fire occurrence for the Amador-El Do-rado Unit (AEU) of the California Department of Forestry & Fireprotection (CDF). The results obtained verify empirically the the-oretical assumptions on suitable methods to determine fire seasonswhen characterizing fire occurrence.

2 - The recursiveness of the age-class forest modeland simulations of climate policy impactJani Laturi, Jussi Uusivuori

We present an age-class forest model to study behaviour of privatelandowners with amenity values. The recursiveness properties of themodel are analysed. We apply the model to study policy tools aimedat compensating landowners for climate services. Two alternativepolicies are compared: one where landowner receive a rental com-pensation, and another one where an auction of providing climateservice is arranged. The simulations are carried out using Finnishregional level data.

42

IFORS 2008 - Sandton RA-09

3 - Locating Fuel Treatments on Flammable ForestLandscapesDavid Martell, Jason Myers, Juan José Troncoso, AndrésWeintraub

Fire and forest managers establish fuelbreaks and carry out otherfuel treatments to reduce the flammability of forest landscapes. Theplacement of fuelbreaks and fuel treatments reduces the risk thatfire poses to timber in industrial forests and to people and propertyin and near wildland urban interface areas. We formulate the fuel-break location problem as stochastic integer programming problemand apply it to a forest management unit in the boreal forest regionof Canada. We derive exact solutions for small test problems andheuristic solutions for our study area.

� RA-07Thursday, 8:00 AM - 9:30 AM - Convention Centre: Boardroom3

Health Care for the Community

Stream: OR and HealthcareInvited sessionChair: Giuditta Callea

1 - A case study on model selection using Akaike andBayesian Information criterion

Model selection is an important part of any statistical analysis, andis central to the pursuit of science in general. Many authors haveexamined it, from both frequentist and Bayesian perspectives. Inthis paper model selection by AIC and BIC are two of the tools usedin model selection. The problem of model selection is concernedwith which functional form should be used to model the probabilityof a malignant tumour as a function of age.The procedures will beapplied to a data set of all the tumours diagnosed in children andadolescents from 1987 to 2002.

2 - Changes in hospital efficiency following the Lom-bard health care reform: An application of theMalmquist TFP indexGiuditta Callea

This paper investigates the effects of the 1997 regional healthcare re-form on the Lombard hospital efficiency. Using the DEA techniquewe measure the hospitals’ TFP change between 1999 and 2006. Sec-ondly, we take a regression analysis to the Malmquist TFP index toanalyse the productivity change determinants.


Visual Aspects of MCDA

Stream: Decision AnalysisInvited sessionChair: Bertrand Mareschal

1 - Visual MCDA using new GAIA type methodsQuantin Hayez, Bertrand Mareschal

Extensions of the GAIA method are proposed to achieve a better ex-ploratory analysis of multicriteria problems (conflict detection andcompromise analysis) and to improve the link between the descrip-tive method GAIA and its prescriptive counterpart PROMETHEE.

2 - Integration of outranking methods to geographicalinformation systemsKarim Lidouh, Yves De SmetFor the past few years a number of authors have considered theintegration of multi-criteria decision analysis (MCA) tools to ge-ographical information systems (GIS). In this talk, we will ana-lyze the potential application of outranking methods, such as thePROMETHEE methodology, to this recent field Finally, we will il-lustrate an example of integration applied to a study of the Brussels’region.

3 - Graphical representation of the preferences of adecision makerYves De Smet, Bertrand Mareschal, Philippe NemeryIn this paper we propose an interactive method to determine the pref-erences of a decision maker when he compares pairwise actions ina ranking, sorting or choice problem. On each criterion a prefer-ence index will be obtained and aggregated to a global preferenceindex. Special attention is devoted to the representation of theseuni-criterion and multicriteria preferences.

4 - Using the GAIA method for multicriteria sorting andclusteringBertrand Mareschal, Philippe Nemery, Quantin HayezMulticriteria sorting and clustering techniques are usually based onexisting ranking methods. We propose to adapt the GAIA descrip-tive outranking method for sorting and clustering. The visual dimen-sion of GAIA provides a way to better distinguish classes of actionsand to understand their specific characteristics. In the context ofclustering, it becomes possible to detect clusters of incomparableactions as well.


Game Theory and Applications 1

Stream: Game TheoryInvited sessionChair: Fouad El Ouardighi

1 - Patent licensing in a Cournot duopolyFlávio Ferreira, Fernanda A. FerreiraPatent licensing is a source of profit for the innovator (also calledpatentee) who earns rent from the licensee by transferring a newtechnology. In this work, we study optimal licensing arrangementswhen a new technology is transferred from a firm which is relativelycost-inefficient in the pre-innovation stage compared to the recipientfirm. We consider three possible licensing policies offered by thepatentee: (per unit) royalty; (lump-sum) fixed fee; and a two parttariff, i.e. a fixed fee plus royalty.

2 - Cooperation in Service SystemsShoshana Anily,We study a cost allocation problem, where a group of servers, eachcharacterized by its capacity and stream of customers, may cooper-ate and generate a "pooled" server system with an arrival/service ratebeing the sum of the respective individual rates. The cost of a coali-tion is the equilibrium mean number of customers in the system. Wedefine the problem as a transferable utility cooperative game, andshow that the grand coalition is beneficial. In one of our main re-sults, we fully characterize the non-negative core of the game, albeitthe fact that the game is neither monotone nor concave.

3 - Cooperation, attachment and forgivenessFouad El OuardighiThe impact of players’ psychological profiles on the evolution of analliance is decisive, even more so than the social and historical con-texts (El Ouardighi, 2007). In other words, determining potentialpartners’ psychological profiles from the outset is one of the mostefficient ways to differentiate beforehand between opportunistic andnon-opportunistic behaviors. This paper seeks to provide an ex-haustive characterization of the outcomes resulting from asymmetricpartnerships, i.e., with different player profiles.

43

RA-10 IFORS 2008 - Sandton


Creating and Building IFORS

Stream: IFORS 50th AnniversaryInvited sessionChair: Heiner Müller-Merbach

1 - IFORS and BeforeK. Brian Haley

Memories of the youngest delegate to the Oxford Conference andthe events leading to the foundation of IFORS immediately beforethe ’Second’ Conference in Aix-en-Provence.

2 - Not what I did for IFORS, but what IFORS did for meGraham Rand

My close involvement with IFORS through 25 years, 1980-2005,has provided some of the most satisfying aspects of my professionalcareer. In this presentation, I will discuss issues of leadership, con-ferences and publications.

3 - Growth and Development of IFORSHeiner Müller-Merbach

After its start in 1958, IFORS grew quite rapidly to a size of some 40member societies, accompanied by necessary alterations of its orga-nizational structure. The content, the purpose, the activities, as wellas the style of the activities changed accordingly. The field as such,i.e. Operational Research, has a different (internal as well as exter-nal) image today compared to the 1950s and 1960s. In the paper,some observations of change will be reported, recalling Heracleitus(544-483 BC): "Nothing endures but change."

� RA-11Thursday, 8:00 AM - 9:30 AM - Convention Centre: CommitteeRoom 2

Supply Chain Risk, Pricing, andCoordinationStream: Supply chainsContributed sessionChair: Johan Joubert

1 - Entrepreneurs and newsvendors: do small busi-nesses follow the newsvendor logic when makinginventory decisions?Charles Corbett, Jan C. Fransoo

We empirically assess the degree to which inventory decisionsmade by small businesses are informed by the logic underlying thenewsvendor model and are influenced by the decision-maker’s riskprofile. We find that entrepreneurs do follow the newsvendor logic,but more so for high-margin than for best-selling products. We findthat entrepreneurs’ risk profiles are consistent with prospect theory.Furthermore, we find that risk aversion for profits is associated withhigher safety stocks, in contradiction to existing theory, and discussseveral possible explanations for this finding.

2 - Dynamic Pricing and its Effects on Supply Chains,

Enhancing the applicability of dynamic pricing beyond service in-dustries, seasonal sales or perishable items is the main focus of thisresearch by combining cognitive decision making, inventory man-agement, dynamic pricing and supply chain management to improvesystem performance. Current research focuses on determining opti-mal pricing policies without considering what effects these policiesmight have on the application domain. The model proposed in thispaper studies the effect of dynamic pricing on supply chains to betterunderstand the applicability of such policies.

3 - Safety stock placement in a multi-echelon supplychain with probabilistic demandJohan Joubert, Marlize Van Zyl

Supply Chain Management has been conceptualized to integrate andmanage key business process across the supply chain, making thepositioning of inventory in a supply chain network an essential ac-tivity. In this paper a particular trade-off is made between customersatisfaction and inventory holding costs using a mathematical op-timization model. The model takes probabilistic demand into ac-count; and transport costs between facilities in the supply chain arepiecewise linear functions. The reliability of the stochastic model isdetermined.

4 - Combinatorial auctions in supply networksPetr Fiala

Supply chain management is more and more affected by networkand dynamic business environment. Networking transforms staticand isolated resources in a pool of dynamic and connected resources.Auctions are important market mechanisms for the allocation ofgoods and services. Combinatorial auctions are those auctions inwhich bidders can place bids on combinations of items. The paperis devoted to using of combinatorial auctions in supply networks.Distributed units’ formation, allocation and coordination problemsin supply networks can be solved through combinatorial auctions.

5 - Price Discrimination with Multi-Part TariffsHemant Bhargava

A relatively small menu of three-part tariffs (3PTs) can be moreprofitable and, sometimes, socially more desirable than a largermenu (more items) of two-part tariffs (2PTs). Often, a single three-part tariff can beat a sorting menu comprising multiple two-part tar-iffs. The 3PT menu can be designed with less information aboutconsumer preferences, relative to the menu of two-part tariffs. Thisadvantage holds even when buyers are uncertain about future valua-tions.

� RA-12Thursday, 8:00 AM - 9:30 AM - Convention Centre: CommitteeRoom 4

Finance: Finance and banking

Stream: FinanceInvited sessionChair: Günter Schmidt

1 - Risk Management Cultures and Practices at LargeFinancial InstitutionsChaiho Kim

The deteriorating subprime mortgages and other debt led to write-downs in $billions by large investment banks and financial conglom-erates in 2007. A significant write-down by a financial institutionwill likely depress its shareholder value and raise the question of thefirm as a going concern. This paper will examine the risk manage-ment cultures and practices of large financial institutions by studyingpublicly available documents including the company reports as 10K.We will try to relate their risk management practices to the risk man-agement frameworks such as Basle II and COSO ERM.

2 - Quantifying Operational Risk of Financial Institu-tionsStanislav Bozhkov, Andrew Scurr, Antoaneta Serguieva,Keming Yu,

Introducing capital charge for operational risk with Basel II has in-creased interest in quantification approaches. Current research fo-cuses on bottom-up techniques using loss databases. Alternatively,we infer the operational component of extreme risk using a panelof share prices and proxies capturing market risks and idiosyncraticcharacteristics. We enhance model specification and estimation pro-cedures, and use data on large European financial institutions, thusimproving the calibration of bottom-up models. The proposed workhas benefited from the collaboration with an international bank.

44

IFORS 2008 - Sandton RA-19

3 - Evaluation of Algorithmic Trading RulesGünter Schmidt, Esther Mohr

There are two elementary types of trading strategies related to as-sets, one is market timing and the other is buy and hold. The ques-tion, if one of these strategies is superior to the other, is heavilydiscussed in the literature. We concentrate on answering this ques-tion using average case performance measures generated from anempirical analysis. We show that there is a market timing strategythat outperforms the buy and hold strategy. The evaluation is basedon historical data from Xetra DAX; volatile, increasing, decreasingmarket movements are simulated using artificial stock prices.

� RA-14Thursday, 8:00 AM - 9:30 AM - Sandton Sun: Maroela North

DEA: Applications in education

Stream: DEA and Performance ManagementInvited sessionChair: aziz kutlar

1 - Efficiency Evaluation of Islamic Azad University ofIran, Firouzkoh Branch, (IAUFB) during 1995-2005:Using DEA MethodMohammad Mehdi Movahedi, , MohammadBamenimoghadam

Efficiency of IAUFB, was calculated and evaluated using Data En-velopment Analysis. For this purpose, activities and performanceof IAUFB during 1995-2005 were collected and efficiency of eachyear as a DMU, with respected the other years was compared andevaluated. Since the data of initial years establishment of the uni-versity was poor, efficiency of each vice-presidency was separatelyevaluated. Many suggestions were made based on the results of theresearch.

� RA-16Thursday, 8:00 AM - 9:30 AM - Sandton Sun: Jacaranda

Location and Layout

Stream: Location and LayoutInvited sessionChair: Edward Gimadi

1 - The Reliable Maxmin-maxsum Location Problemon a TreeEmanuel Melachrinoudis, José Santiváñez

A facility is to be located on a tree network that serves demand re-quests originating at the nodes. The edges of the network fail withcertain fixed probabilities. Two simultaneous criteria are consideredregarding responses to demand requests: maximizing the minimumexpected number of successful responses over all nodes and maxi-mizing the expected total number of successful responses. Proper-ties of the efficient and nondominated sets are developed based onwhich an efficient algorithm for their construction is provided.

2 - A new method for solving a layout problem usinggroup technology approach, Halim Mahdi, Lionel Amodeo, Farouk Yalaoui

This article presents the solution of a facility layout problem. Forsolving it, we used group technology approach mixed to a quadraticassignment problem. In the GT approach,we have generated productfamilies belonging to groups of machines using a specific genetic al-gorithm. The objective function consists of maximizing the amountof quantities inside the machine’s groups.The QAP is resolved by anhybrid ant colony optimisation mixed to a global local search algo-rithm. The method has been tested on a large number of problems;also it has been applied on an industrial case.

3 - A Heuristic Approach to Solving Block LayoutProblems with Shape ConstraintsR. Aykut Arapoglu

We consider the block layout design problem with shape constraints.Typically departments are rectangular in shape. In this case, how-ever, we allow some departments to assume an L-shaped configu-ration. Departments are subject to an aspect ratio constraint whoseviolation results in a penalty in the objective function. In this talk,a genetic algorithm based heuristic search procedure will be pre-sented. The goal is to generate feasible block layouts having mini-mized distance based material handling costs.

� RA-19Thursday, 8:00 AM - 9:30 AM - Sandton Sun: Acacia

Queues 1Stream: QueuesInvited sessionChair: Jeff Griffiths

1 - Fair charge heuristics for dynamic resource alloca-tion: a queueing exampleDavid Hodge, Kevin Glazebrook, Christopher Kirkbride

We develop a notion of a fair charge for the allocation of additionalresource to a reward earning object (project) whose stochastic evo-lution depends upon how much resource it receives. Under a criti-cal condition, heuristics based on such fair charges are effective forproblems concerning the dynamic allocation of a divisible resource(money, manpower) among a set of such projects. We illustrate ideasby reference to a queueing example in which a set of servers is to bedynamically allocated to a collection of service stations.

2 - Service Level Trade-Offs in a Reservation-BasedPriority SystemKi-Seok Choi

We consider a service system where the customers who make areservation in advance have higher priority. The capacitated ser-vice provider may guarantee different service levels for the high-and low-priority customers. We show the trade-offs between the to-tal service time and other system characteristics such as the propor-tion of high-priority customers, when the service level is high. Wealso suggest how to differentiate service using the asymptotic ratioof the service levels for both customer types.

3 - Approximation to the Transient Solution of theM/Ek/1 QueueJeff Griffiths

This paper considers a single-server Erlang queueing system, wherecustomers arrive at random and the service times have a general Er-lang distribution. It is difficult to obtain a transient solution in ex-plicit form to the queue equations because of their complex struc-ture. We propose a simple method of computing the mean waitingtime of a customer arriving in the queue at time t, based on an expo-nential function approximation.

45

RB-01 IFORS 2008 - Sandton

Thursday, 10:00 AM - 11:00AM

� RB-01Thursday, 10:00 AM - 11:00 AM - Convention Centre: Ballroom1

Humanitarian Logistics

Stream: PlenariesPlenary sessionChair: M. Grazia Speranza

1 - Supply Chain Management in the Context of Hu-manitarian DisastersLuk Van Wassenhove

Recent events have made it perfectly clear that both the frequencyand intensity of humanitarian disasters have increased. The year2005 has been a particularly dark one in terms of number of peoplekilled or otherwise affected (tsunami, Katrina, Pakistan, Darfur). InIndonesia, one-third of the relief supplies were still blocked at theairport three months after the tsunami, while in Pakistan one millionpeople were still waiting for winter tents more than a month after theearthquake. At the core of these are fundamental supply chain man-agement problems (bottlenecks, procurement issues, tracking andtracing problems).

The world of humanitarian disaster logistics is very different fromthe familiar context of business logistics. Humanitarians operatein very dynamic environments and need to preserve a humanitar-ian space (respecting the principles of impartiality, neutrality andhumanity). This space is increasingly challenged by political andmilitary forces, to the point that security of humanitarian personnelhas become a key concern.

Since traditional funding sources are declining, and natural and man-made disasters become more frequent, humanitarian organizationsrealize that they will have to increasingly rely on private organiza-tions for help. Simultaneously, companies in search of a coherentcorporate social responsibility agenda, increasingly turn to the hu-manitarian world. However, it is far from obvious how private com-panies and humanitarian organizations can effectively work togetherin disaster relief.

This presentation will introduce supply chain management in the hu-manitarian context, and outline the need for better preparedness andresponse to disasters. We also discuss opportunities for more ef-fective collaboration among humanitarian organizations, as well asbetween humanitarians and private organizations. The presentationuses a number of real cases to highlight opportunities to do interest-ing research with a high potential impact.

Thursday, 11:30 AM - 1:00PM

� RC-02Thursday, 11:30 AM - 1:00 PM - Convention Centre: Ballroom2

Dynamic Programming

Stream: OptimizationTutorial sessionChair: Augustine ESOGBUE

1 - Dynamic Programming for the Real WorldAugustine ESOGBUEDynamic programming has appeal to assorted real world problems.We present it from its multifacets to modern day developments anddiscuss approaches focal to ORMS. We review novel models includ-ing nonserial, heuristic, fuzzy and neuro-dp and present an overviewof applications.


Transportation Systems: Supply andDemandStream: Intelligent Transportation SystemsInvited sessionChair: Janny Leung

1 - Optimizing the Number of Operators in anOligopolistic Transit MarketWilliam Lam, Zhi-chun Li, S.C. WongWe propose a model to optimize the number of operators in anoligopolistic transit market involving three types of players: tran-sit authority, operators and transit passengers. The transit authorityaims to maximize total social welfare by setting the number of op-erators permitted to enter the market. All transit operators seek tomaximize their own profits by determining frequencies and fares,while accounting for transit passengers’ responses. The model isformulated as a multi-level hierarchical problem and solved by ameta-heuristic algorithm. A numerical example is used for illustra-tion.

2 - Impact of Transport Network Management on Ac-tivity Location Cost Over TimeHong K. Lo, , Xiaosu MaBoth transportation supply and demand management are importantfor transportation management, which will affect activity locationchoices, as they change the travel costs between zones. This studywill analyze the interactions between these two aspects. The resultswill have important implications on developing land use and trans-portation improvement strategies.


Optimization for sustainable production1Stream: OptimizationInvited sessionChair: Jiri KlemesChair: Ferenc Friedler

46

IFORS 2008 - Sandton RC-08

1 - Process Integration as an Optimisation Tool inClean Coal TechnologyRhulani Shingwenyana, Thokozani Majozi

The integrated gasification combined cycle (IGCC) represents oneof the promising technologies for clean coal power generation. Thecurrent study seeks to use pinch technology in the optimisation ofIGCC units. This is done by graphical approach to set practicalperformance targets for the achievable energy performance of thesystem.

2 - A Mathematical Optimisation Technique for the Ef-fective Scheduling of Zero-Effluent Batch PlantsJacques Gouws, Thokozani Majozi

Wastewater minimisation in industry is becoming ever more impor-tant as environmental legislation becomes more stringent on efflu-ent discharge. Wastewater minimisation through proper schedulingis an economic means of meeting discharge targets. Presented is acase study in which wastewater is minimised through reuse in prod-uct, thereby producing almost zero-effluent. The model used in thecase study determines the optimal schedule to achieve the maximumwater reuse in product.

3 - Aroma Guided Energy Integration in Sludge Treat-mentThokozani Majozi, Jaime Rojas-Hernandes, Toshko Zhelev

The paper focuses on the problems of energy efficiency improve-ment of wastewater treatment processes. The case study concernsbio-energy recovery in autothermal aerobic digestion (ATAD). Thepractice shows that the process supports generation of ammonia,which speaks for insufficient agitation and aeration. It involves anabsorption tower (Mona shell absorber) where the ammonia con-tained in the gases drawn from the sludge storage tank and two re-actors is absorbed. The paper solves the problem of heat recoverywhere the reactor model accounts for the ammonia generation andtemperature.

4 - CO2 Emissions Reduction via a Graphical AnalysisMethod for Renewable Energy Supply ChainsJiri Klemes, Hon Loong Lam, Petar Varbanov

A novel renewable energy supply chain method is addressed inresponse to carbon-constrained energy generation. A graphicalmethod is proposed to reduce the energy waste and improve the ef-ficiency of energy supply chain systems. It illustrates how supplychain with integrated renewables and waste-to-energy would com-ply with environmental policies.

� RC-07Thursday, 11:30 AM - 1:00 PM - Convention Centre: Boardroom3


Stream: Humanitarian LogisticsInvited sessionChair:

1 - A Multi-objective and Stochastic Warehouse Loca-tion Routing Problem for Disaster ReliefStefan Rath

We present a multi-objective warehouse location routing problemwith stochastic demands and stochastic accessibility of routes, whichis faced by non-governmental organizations after disasters such asearthquakes, floods or tsunamis. A bi-objective integer linear pro-gram and metaheuristics were chosen as solution methods and un-certainty was tackled by sampling.

2 - A Decision Support System for the Distribution ofHumanitarian AidBegoña Vitoriano, M. Teresa Ortuño

In recent years, natural and man-made disasters have been affectingincreasing numbers of people throughout the world. Organisationsfor emergency and humanitarian aid have experienced an importantgrowth, and efficiency in management becomes crucial. There is alack of specific tools devoted to logistics of this special kind of inter-ventions in developing countries, demanded by the organisations. Adecision support system currently in development, is presented fo-cusing on the transport problem to distribute humanitarian aid to theaffected population of a disaster in a developing country.

3 - A Stochastic Model for Hurricane Logistics, Pre-paredness, Response, and Humanitarian Aid, Richard Larson

With an approaching hurricane, emergency managers face a set oflogistical action timing decisions: supply, pre-positioning, mobi-lization, evacuation, and humanitarian aid. The location, time andintensity of the hurricane at landfall are random variables governedby nature. We developed a stochastic dynamic program incorporat-ing economic, social, and logistical impacts of alternative sequentialdecisions. Results are given for a set of global case studies.


Decision support systems for economicor environmental issuesStream: Management Information SystemsInvited sessionChair: Lidija Zadnik Stirn

1 - Assessment of software investmentsEmil Numminen, Birger Rapp, Anders Hederstierna, HenrikSällberg

It is difficult to assess the value of software investments. The mainreason is the investment’s characteristics, i.e. the high degree ofintangible consequences. We propose a solution based on the netpresent value approach with a stochastic cash flow function incorpo-rating the intan-gibles from the expected software usage.

2 - Decision@Work: a framework for supportingdecision-intensive business processesKathrin Kirchner,

The purpose of Decision@Work is to generalize and integrate com-mon aspects in DSS development in a framework, which is applica-tion domain independent. The framework helps especially SMEs tobuild customized DSS for different industries. A method base offersreusable components to support users in decision-intensive work-flows. The system will be designed to be used by technical engineersas well as non-technical clients. In addition to a technical modellingour research focuses on a customer friendly modelling for clients,which enables cooperation in the DSS development process.

3 - AHP, ANP and SWOT analysis for optimal manage-ment of ecosystemsLidija Zadnik Stirn

When controlling an ecological system we have to create an equi-librium of several conflicting objectives. Here we present a multi-criteria model using the SWOT, AHP and ANP analysis. The AHPand ANP are integrated with SWOT to yield analytically determinedpriorities for the SWOT factors. Pairwise comparisons betweenSWOT factors and within each SWOT group are carried out. Fi-nally, the composite value of the management alternative is calcu-lated. Application of the presented model deals with environmentalprojects in the Primorska region, Slovenia.

47

RC-09 IFORS 2008 - Sandton



Stream: Game TheoryInvited sessionChair: Georges Zaccour

1 - Modeling and Learning in Settings with Misspeci-fied ModelsAnton Kleywegt

The operations research approach involves the formulation of mod-els that approximately describe a situation of interest, and the studyof the models to obtain results for decision making. Typically, mod-els are simplified, and thus are misspecified. The analyst may collectdata and update parameter estimates while using the model. An im-portant question is what properties the resulting dynamical processhas, for example, whether the sequence of estimates converges toa good value. The talk will present some research that addressesmodeling and learning in settings with misspecified models.

2 - Technology, Operations and Marketing Strategy ina Dynamic DuopolyFouad El Ouardighi, Bowon Kim, Sangsun Park

To be successful in launching a new product, the firm must take intoaccount internal operations in addition to external factors like com-petition, technology and market attributes. Suppose that the firmseverely underestimated sales and failed to prepare enough capacityto meet the overwhelming demand. Then, even if the new productsells well in the market, the company cannot get profitability. It isessential for the firm to optimally plan for technology, market, andoperations simultaneously. In this paper, we develop a differentialgame model of duopoly that analyzes the dynamic problem.

3 - Merger in a Stackelberg oligopolyFernanda A. Ferreira

In this work, considering a Stackelberg oligopoly market, we showthat the profitability of merger depends on the market structure andon the involved firms’ strategic power. Furthermore, the decisionto merge depends neither on the intercept nor on the slope parame-ters from the demand curve. However, when the incentive to mergeexists, it increases with the first parameter and decreases with thesecond one.

4 - Advertising Strategies in Fashion LicensingGeorges Zaccour, Alessandra Buratto

We characterize cooperative and noncooperative advertising strate-gies of the licensor and the licensee involved in a licensing contractin the fashion business. Licensing is a contractual agreement be-tween the owner of the property (licensor), and the renter of therights (licensee). Licensing is a win-win strategy where the twopartners can achieve their objectives. We show that if the licensor(leader), uses an incentive strategy which depends on the licenseeadvertising, then it can reach in a decentralized manner the jointlyoptimal solution.


The Present State of IFORSStream: IFORS 50th AnniversaryPanel sessionChair:

1 - The Present State of IFORS

Current activities of IFORS revolve around its objectives of devel-oping OR as a unified science and to advance it world-wide. It isorganized to: sponsor international meetings; facilitate exchange ofinformation; encourage nations to establish societies; support stan-dards of competence; promote OR education as well as the growthof existing and new fields of OR.

Groups headed by members of the Administrative Committee per-form these roles. In addition, IFORS has taken into account devel-oping country concerns although it has yet to put in place programsto support standards of competence.

This panel discussion will provide an overview of all of the activitiesof IFORS, with presentations by Administrative Committee mem-bers and Committee Chairs.

� RC-11Thursday, 11:30 AM - 1:00 PM - Convention Centre: CommitteeRoom 2

Supply Chain Applications

Stream: Supply chainsContributed sessionChair: Ozias Ncube

1 - Rail Capacity Expansion in the Coal Supply Chain:A Simulation StudyLeorey Marquez, David Sier, Andreas Ernst

The Australian Government has joined the coal industry in exploringshort and long term measures to address capacity and infrastructureissues and to improve throughput in the coal supply chains. This pa-per describes a simulation study conducted by CSIRO for the Dal-rymple Bay Coal Terminal aimed at evaluating various operationaland logistical options for expanding rail capacity and increasing effi-ciency in DBCT’s rail transport network. The study builds on previ-ous work showing significant increase in throughput from improve-ments in train speed and loading rates.

2 - Supply Chain management in the FMCG industry:Perspectives from South AfricaJenny Fletcher, Ozias Ncube

This paper investigates supply chain management practices in theforward deployment of stock in the FMCG industry, exploring theinfluence of information and communication technology and exam-ining their influence on operational performance. Data collectedfrom 27 organisations supplying FMCG in South Africa is testedusing Pearson’s Correlation Coefficient.

3 - Strategic Challenges for Supply Chain Integration:Perspectives from Southern AfricaOzias Ncube, Venkata Yadavalli

There has been steady growing interest on improving supply chainperformance. One area identified by many researchers has been sup-ply chain integration. This paper gives a chronology of strategicchallenges common to major supply chains in South Africa on sup-ply chain integration and performance, and offers a framework thatcan be used to facilitate this integration. A case study is presented


Finance: Financial modelling

Stream: FinanceInvited sessionChair: Marc Schauten

48

IFORS 2008 - Sandton RC-15

1 - Leasing and options for entrepreneursAlejandra Gomez Padilla, Tsutomu MishinaEffective use of lease by applying a Put Option nature in a contracthelps avoid any financial difficulties, especially for newly estab-lished firms. This paper demonstrates applications of the conceptand presents results obtained using simulation. In this option basedevaluation, we expect to increase managerial effectiveness for bothparties.

2 - A Framework for Capital Structure DecisionsMarc Schauten, Jaap SpronkWe show that capital structure decisions can be framed as multiplecriteria decision problems which can then benefit from multiple cri-teria decision support tools that are widely available. We analyze fi-nancing solutions as proposed by practicing specialists for a mergerand acquisition problem. We find that i) the solutions between thespecialists differ, ii) the solutions and criteria applied by the special-ists depend on the stakeholder the solution is made for, and iii) someeconomic criteria appear not to be as relevant as suggested by theory.


Scheduling Applications 2

Stream: SchedulingInvited sessionChair:

1 - A heuristic based on partial ordering for solving theminimization of tool switches problemHoracio Yanasse, Rita de Cássia Meneses Rodrigues, EdsonSenneIn the minimization of tool switches problem we seek to determine asequence to process a set of jobs so that the number of tool switchesrequired is minimized. In this work we propose a heuristic thattries to determine good sequences expanding promising partial se-quences. Computational test results are presented.

2 - Optimizing the Slab Yard Crane Scheduling Prob-lem Using a Two-Stage Planning/Scheduling Ap-proachAnders Dohn, , Jesper LarsenWe present the Slab Yard Crane Scheduling Problem. We need togenerate a schedule for two gantry cranes sharing tracks. The ob-jective is to minimize the tardiness of the schedule. We introduce atwo-stage planning/scheduling solution approach to the problem andpresent the results achieved with this method.

3 - Scheduling in Open Pit Mining: Clever Disaggrega-tion and Stochastic ModelsGary Froyland, ,We present two important extensions to the Open Pit Mine Produc-tion Scheduling Problem (OPMPSP). Firstly, due to the typicallymassive size of the OPMPSP, a standard approach to reduce theproblem size is to aggregate scheduling units. We introduce an LP-based disaggregation approach that produces very accurate solutionsvery quickly. Secondly, as the geological data are only incompletelyknown, it is current practice to produce many stochastic simulationsof the geology. We describe a new MIP formulation of the OPMPSPthat uses this geological information in a non-anticipative way.

� RC-14Thursday, 11:30 AM - 1:00 PM - Sandton Sun: Maroela North


Stream: DEA and Performance ManagementInvited sessionChair: Mohammad-Reza Alirezaee

1 - A new integrated DEA model for finding most BCC-efficient DMUMehdi Toloo, Soroosh Nalchigar

In many applications of DEA finding the most efficient DMU is de-sirable. Using basic DEA models, decision maker isnâ EUR TMtable to identify most efficient DMU. Amin & Toloo (2007) [Com-puter & Industrial Engineering, 52, 71-77] introduced an integratedDEA model for finding most CCR-efficient DMU. In this paper, wepropose a new integrated model for determining most BCC-efficientDMU by solving one LP. This model is useful for situations in whichreturn to scale is variable. The applicability of the proposed modelis shown, using a real data set of 19 facility layout alternatives.

2 - A Comparison of the position of Turkey with OECDCountries in the Area of Science and TechnologyUsing Data Envolopment AnalysisErhan Berk, Muhammet UNAL

Nowadays, mastering the science which generates technology is theonly way to gain a prosperous, sustainable economy as well as to besuccessful in international competition. In this study, developmentlevel of Turkey among OECD countries is determined in terms of in-vestment in science and technology and its rate of return using DataEnvelopment Analysis. Efficient virtual units for the inefficient deci-sion units are found, and factors values are determined with respectto reference units.

3 - An extended numeration method for solvingFDH/DEA models of standard economies of scaleMohammad-Reza Alirezaee, , Behzad Babazadeh

Free Disposal Hall Model which introduce by Deprins et. al. in 1984is one of the most applicable methods of evaluating DMU’s, becauseof the ability of finding a single reference. This paper a numerationmethod will be extended which can be apply for solving all fourstandard economies of scale (constant, variable, increasing, and de-creasing) for FDH technologies. The extended models will be testedby using a real data set of 25 Iranian Gas Distribution companies.

� RC-15Thursday, 11:30 AM - 1:00 PM - Sandton Sun: Maroela South

OR for Sustainable Development 1

Stream: OR for Sustainable DevelopmentContributed sessionChair: Tatjana Tambovceva

1 - Raw Material Logistics for BiorefineriesMagnus Fröhling, Otto Rentz

Biorefineries aim to utilize renewable resources completely to re-duce the dependency on fossil resources and the emission of green-house gases. In comparison to fossil resource-based value chains theraw material logistics for biorefineries differ. Renewable resources,like wood or straw, are e.g. spatially distributed in greater areasand their energy density is comparably low, making long distancetransports unfavorable. The presentation characterizes the specialrequirements of raw material logistics for biorefineries and the con-sequences for logistical planning tasks and methods.

2 - Optimal Jobs-Housing Assignment Problem Con-sidering Density and Growth Limitations, Tsutomu Suzuki

The purpose is to find out how much the travel time and energy con-sumption could be saved by considering densities of jobs and hous-ing uses within growth limitations. Mathematical programming isused to solve the optimal allocation pattern of jobs and housing.Adopting the model, we evaluate that only public transport infras-tructure is invested and only controlling densities without invest-ment, and also both are considered. Moreover, several cases willbe evaluated by controlling densities for jobs and housing uses tosuggest the optimal urban structure.

49

RC-16 IFORS 2008 - Sandton

3 - Implementation of environmental managementsystems (ISO 14001): a case study at the construc-tion enterprises in LatviaTatjana Tambovceva, Ineta Geipele

The purpose of this study is to explore the experiences of the Lat-vian construction enterprises working with environmental manage-ment systems (EMS). The aim of an EMS is to improve the environ-mental performance of the organization. Our results show that theenterprises primarily expect an external recognition of the EMS ac-tivities. ISO14001 fulfils many of the expectations and can be seenas an instrument serving a wide range of purposes. The authors con-clude that it is necessary to improve work with the EMS and to letall line personnel to participate in this work as early as possible.

� RC-16Thursday, 11:30 AM - 1:00 PM - Sandton Sun: Jacaranda

Locational Planning

Stream: Locational PlanningContributed sessionChair:

1 - Reactive GRASP based Heuristics for the Capaci-tated Multi-source Weber ProblemGábor Nagy, , Said Salhi

This paper examines the continuous capacitated location-allocationproblem. Reactive GRASP based heuristics are put forward. Therestricted candidate list is constructed and updated at each iterationby applying a self adjusted threshold parameter. The experimentshows that the proposed methods provide encouraging results whencompared to recently published papers.

2 - Optimal site selection for self-storage enterprisesRichard Lackes

The site selection problem is a long-term strategic problem that aimsfor the determination of warehouses sites and the point in time whenthey should be put on stream. The decision criterion consists in the(expected) net present value. The paper introduces a decision modelBased on binary decision variables representing the sites and thepoints in time of investment with respect to a given budget and theconstraints of the market situation and volume. The business volumedepends on the population of the trading area, its spending powerand the competing warehouses.

3 - Site Location of a Manufacturing Plant.

A Location Model of a concrete manufacturing plant which accom-modates a multifunctional facility is presented. The model dealswith a range of criteria which are of critical, objective or subjectivenature. The criteria are combined in a quantitative fashion whichfacilitates the optimal choice of plant location and affords consider-ation of solution sensitivity.

� RC-18Thursday, 11:30 AM - 1:00 PM - Sandton Sun: Syringa

Cutting and Packing 1

Stream: Cutting and PackingInvited sessionChair: José Fernando Oliveira

1 - A One-Dimensional Cutting Stock Problem from theWood-Processing Industry: Usable Leftovers andAdditional Handling Operations

The author will present a case study from the wood-processing in-dustry where a cutting stock problem with usable leftovers is com-bined with a preceding handling process. Both activities — the han-dling process and the cutting process — are strongly interdependent,and thus, cannot be planned independently. A model formulation ofthe problem will be given and it will be shown how it has been in-tegrated into a decision support system. Finally, some experiencefrom the application of this system is reported.

2 - A Cutting and Lot-Sizing Coupled Problem in theFurniture IndustryMarcos Arenales,

An integer programming model is proposed to combine cutting andlot-sizing decisions for small furniture industries whose objective isto minimize costs of production, storage, setup and material waste.Also, a Lagrangian based heuristic for a relaxed problem is proposedand preliminary computational experiments are presented.

3 - Research support tools in cutting and packing: asurveyJosé Fernando Oliveira, A. Miguel Gomes

Cutting and Packing has developed, as a specific combinatorial op-timization problem, in the last years, as the number of publicationsin journals show. Several tools to support research in cutting andpacking have also been developed but they seem not to be widelyknown and used. In this talk a survey on research tools on cuttingand packing will be presented, including web sites, benchmark prob-lem instances repositories, problem generators, organized literature,available software code and problem typology. The aim is to supportboth researchers and journal reviewers in their work.

� RC-19Thursday, 11:30 AM - 1:00 PM - Sandton Sun: Acacia

Queues 2Stream: QueuesContributed sessionChair: Christopher Kirkbride

1 - A continuous review perishable inventory systemwith multi-server facility and negative customersVenkata Yadavalli, Balasubramanian Sivakumar,Gunaseelan Arivarignan

This paper introduces a multiserver facility in the inventory systems.We assume that the arrivals of customers are according to a Marko-vian arrival process (MAP) and that service time has exponentialdistribution. In addition to the regular customers, a second flow ofnegative customers following an independent MAP is also consid-ered which will remove one of the customers in the orbit. Variousmeasures of stationary system performance are computed and thetotal expected cost per unit time is calculated. The results are illus-trated numerically.

2 - Analysis of the C2/G/1 Queueing Model with Excep-tional Services

We have considered a single server queueing system in which the ar-rival process is assumed to be Coxian with two stages and customersare served on the FCFS basis. The service time distribution dependsupon the number of customers served since the beginning of cur-rent busy period. The service time distribution becomes stable afterserving N customers in the current busy period. We have obtainedrecursive formulas to find the generating function of the stationaryqueue length distribution. This type of queueing models find niceapplication in computer systems, telecommunication systems.

3 - Building a Simulation Model to Control ComplexProduction Processes using Queuing ApplicationsSaad Hasson

50

IFORS 2008 - Sandton RD-03

A simulation was used to model a problem of controlling threestages production system using queuing theory applications. Thereare many parallel production machines in each stage. The productsproduced by the first stage machines are used as input to the sec-ond stage machines and the 2nd stage output used as input to thethird stage machines. The spaces between the production stages areassumed to be limited. Each machine in each stage will be startedwhen it was required and stopped when it was not required in a con-trol automatic manner

4 - Dynamic pricing of queuesChristopher Kirkbride

We consider a problem in which a system manager has to determinethe appropriate charge to offer arriving customers to maximise rev-enue net of holding costs. We derive dynamic pricing heuristics viathe application of a single dynamic programming policy improve-ment step to a strongly performing static policy.

Thursday, 2:00 PM - 3:30 PM

� RD-02Thursday, 2:00 PM - 3:30 PM - Convention Centre: Ballroom 2

Convex Optimization

Stream: OptimizationContributed sessionChair: Alexander Kruger

1 - Approximation of Stochastic Differential Equationsby Additive Models Using Splines and Conic Pro-grammingPakize Taylan, Gerhard-Wilhelm Weber, Devin Sezer

Many phenomena in nature are modeled by means of determinis-tic differential equations. But, this type of modeling omits stochas-tic fluctuations. For this reason, we consider stochastic differentialequations,since we know that,in reality, many processes are normalyaffected with noise. We propose an approximation by additive mod-els based on splines and we construct a penalized residual sum ofsquare and a Tikhonov regularization problem for this model.wetreat it using continuous optimization techniques. In particular,weapply the elegant framework of conic quadratic programming.

2 - Feasibility of Integer Knapsacks and the FrobeniusnumberIskander Aliev

The Integer Knapsack Problem is a classical combinatorial optimi-sation problem with many applications. Its feasibility is known to berelated to the famous (linear diophantine) Frobenius Problem: givena set S of relatively prime positive integers, the problem asks forthe largest natural number, called the Frobenius number, that has norepresentation as a non-negative integer combination of the elementsfrom S. In the talk we will discuss recents results on the behaviour ofthe Frobenius number in the light of the Integer Knapsack Problem.

3 - About Extremality, Stationarity, and Regularity ofCollections of SetsAlexander Kruger

Extremality, stationarity, and regularity properties of collections ofnonconvex sets in a normed linear space are characterized in termsof certain primal and dual constants. The extended extremal princi-ple is formulated. The relationship between regularity of collectionsof sets and metric regularity of multifunctions is explained.


Arc Routing

Stream: RoutingContributed sessionChair: Elena Fernandez

1 - Probabilistic Arc RoutingRichard Wong, Si Chen, Bruce Golden, Hongsheng Zhong

A vehicle follows a predetermined route for covering a given set ofstreets. On any day, a subset of the streets requires service and otherscan be skipped. The probabilistic arc routing problem determines apredetermined route that minimizes the expected daily routing cost.We discuss modeling and solution approaches.

2 - Capacitated Arc Routing Problem: School BusRoutingJohnson De-Graft Ankamah,

51

RD-04 IFORS 2008 - Sandton

One major problem persistently facing operators of school buses indeveloping countries is how to route the buses. The problem of rout-ing school buses deals with the important question of how to trans-port students to and from schools in the safest, most economical andmost convenient manner. A constructive heuristic model for a singleload school bus routing problem as seen in developing countries hasbeen presented. The model seeks to aside minimizing operationalcost and time, gives different priority level to different category ofstudents (physically challenged, six years and below etc).

3 - Heuristics for the Capacitated Rural Postman Prob-lemFábio Usberti, Paulo Morelato França, Vinícius Garcia,André Luiz Morelato

In this talk the Capacitated Rural Postman Problem (CRPP) withopen routes is focused. Differently from the classical CRPP, in thiscase the routes are not constrained to form cycles. Two local search-based heuristics for the CRPP with open routes are presented. Thefirst one proposes improvements on a "route first, cluster second"procedure and the other strengthens the results obtained by a "clus-ter first, route second" heuristic. Computational tests show that thenew heuristics attained better results in all but one comparison crite-ria, namely execution time.

4 - The exact solution to the Clustered Price-collectingArc routing ProblemElena Fernandez, , Carles Franquesa

In Prize-collecting Arc Routing Problems, in addition to the costfunction, there is a profit function on the edges that is taken intoaccount only the first time an edge is traversed. In the ClusteredPrize-collecting Arc Routing Problem there are clusters of arcs and,for each cluster, we require to serve either all its links or none ofthem. We study properties and dominance conditions that allowstrengthening the formulation. We also propose an exact algorithmand present the numerical results from a series of computational ex-periments that assess the good behavior of the algorithm.


Optimization for sustainable production2Stream: OptimizationInvited sessionChair: Ferenc FriedlerChair: Jiri Klemes

1 - Integration of Fuel Cells into Combined Power Cy-clesFerenc Friedler, Petar Varbanov, Jiri Klemes

Research on integrating high-temperature fuel cells into Fuel CellCombined Cycles (FCCC) indicates achieving high efficiencies andeconomic viability is readily possible. Various FCCCs and the po-tential for utilising renewables are analysed considering generationcapacity and application context. Further research and developmentare outlined, giving priority to CO2 emissions reduction.

2 - Optimization of Industrial Plants and Production:P-graph and S-graph MethodologiesFerenc Friedler, L. A Fan

Industrial optimization problems become highly complex partly be-cause of the requirements related to sustainability. This presentationintroduces methodologies exploiting the specific structures of indus-trial production systems for their effective optimization; P-graph isfor the design and S-graph is for the scheduling of them. They canhighly contribute to sustainable development.

3 - S-graph Based Continuous-Time Approach forThroughput Maximization in Multipurpose BatchPlantsThokozani Majozi, Ferenc Friedler, Tibor Holczinger, MateHegyhati

This work presents a recently developed scheduling technique formultipurpose batch plants that is based on a graph-theoretic frame-work, the S-graph. The main advantage of this technique is that itdoes not rely on discretization of the time horizon of interest whichis almost intrinsic in all mathematical programming approaches.

� RD-05Thursday, 2:00 PM - 3:30 PM - Convention Centre: Boardroom1

OR Applications to Education

Stream: OR EducationContributed sessionChair: Londiwe Masinga

1 - Decision Support System For Managing Educa-tional Capacity Utilization For a New Private Col-lege in Sub-Saharan AfricaKweku-Muata Osei-Bryson, Muhammadou M.O. Kah,Jainaba M.L. Kah

Decision making for new private universities in developing coun-tries with aspirations of world class programs are confronted withvarious challenges including: uncertainties of enrollment, provisionof the requisite infrastructure and its cost tradeoffs, being able toprovide competitive remuneration packages and wealth fare. In thispaper, we propose a methodology for assessing educational capacityutilization and facilitating resource allocation.

2 - Allocation of funds among university departments- a DEA studyJosef Jablonsky

The main aim of the paper is to verify alternative approaches for al-location of resources among university departments at the UEP. Thecurrent practice is based on multiple criteria decision making tech-niques. The paper introduces an approach based on standard DEAmodels. The model offers how to change inputs of efficient and inef-ficient units under the assumption that the outputs remain unchangedin order to move all the originally inefficient departments to the effi-cient frontier. Given results are compared to current practice and toresults of several multicriteria decision making methods.

3 - A Mathematical Approach to Enrolment Planningand Management with Targets Over a Finite PeriodLondiwe Masinga

We use a multi-stage Mathematical Programming model to investi-gate how a South African Higher Education institution can optimallymanage enrolment inputs in order to achieve set target output levelsat the end of a planning period, and derive optimal state funding.


Data Mining 1

Stream: Data Mining & Knowledge ManagementContributed sessionChair: MF Immelman

1 - A Multi-Dimensional Measure of PovertySathie Naidoo

This paper presents and discusses the multivariate analysis and mea-surement of poverty. The analysis advances the French social exclu-sion theory and Sen’s analysis of functioning and capabilities, mak-ing them operational by applying the fuzzy set theory and the NeuralNetwork Self Organizing Maps. The outcome of this research is ap-plied to RSA census 2001 data.

52


2 - Machine Learning Techniques for CryptanalysisB Chandra, Pallath Varghese, Pramod Saxsena, ShrikantShrikant

Machine learning techniques like Self Organizing Maps has been ex-plored for the first time for efficient key clustering in cryptanalysis.In the first phase, wavelets have been employed for efficient featureextraction, in the second phase, additional transformations have beenemployed to find the dominance in the features and finally SOM hasbeen employed for clustering.

3 - Support vector machines for credit scoring anddiscovery of significant featuresAnthony Bellotti,

Support vector machines are applied to determine risk of default attime of application for a large credit card database provided by a Eu-ropean bank. We use them for feature selection and find them com-petitive for prediction of default compared with traditional creditscoring techniques such as logistic regression.

4 - Visualising relationships in qualitative dataMF Immelman

Exploratory analysis techniques for quantitative data are well devel-oped, widely published and readily included in numerous statisticalpackages. In contrast to that exploratory analysis techniques for datameasured on a qualitative scale is rarely mentioned or included insoftware packages. Visual methods to quantify bivariate categoricalrelationships are described.



Stream: Humanitarian LogisticsContributed sessionChair: Kelvin T Chirenje

1 - Securing extramural health care for people withspecial needs in times of crisis and disasterPatrick Hirsch

Through a wide range of mobile care services it is possible for thefrail and elderly people with special needs to stay at home as long aspossible. We study the consequences of complex disaster situations,like floods and pandemics, on the formal care and support structuresof these people. This work is based on a research project togetherwith the Austrian Red Cross. It deals with the logistics of the careservices, evaluates possible disruptive factors, and should provide anextensive concept of improvements. The used methods will includesimulation techniques and optimization heuristics.

2 - An Analysis of the Inventory Levels and Distribu-tion of Anti-retroviral drugs by Non-governmentalOrganisations in ZimbabweBRIAN CHIRIPANHURA

This paper is an analysis of the inventory levels and distribution ofanti-retro viral drugs by non-governmental organizations in Zim-babwe to come up with an optimal and cost effective inventorymodel and distribution channels.

3 - Location decisions in a blood serviceJeanne le Roux, Jan van Vuuren

Recent centralisation of blood transfusion services in South Africanecessitates reviewing of inventory management policies and opera-tional and logistical procedures.

The objective is more efficient blood service delivery.

A binary programming formulation of a facility location decisionproblem within the blood demand-supply chain is presented.

4 - An investigation into the Humanitarian Logisticsemanating from the distribution of the HIV virus inSub Saharan AfricaKelvin T Chirenje

This is a quantitative and qualitative analysis into the HumanitarianLogistics behind the socio-economic effects of the HIV virus and itsimpact on emerging markets: The Zimbabwe Case



Stream: Game TheoryInvited sessionChair: Guiomar Martin-Herran

1 - A family of values for cooperative games in gener-alized characteristic function formDaniel Gomez Gonzalez, Enrique González-Arangüena,Conrado M. Manuel

In this communication we focus on cooperative games in general-ized characteristic function form. For these games the worth of acoalition depends on the order in which its members add to the coali-tion. We introduce a family of values for this type of games whichis characterized in terms of some appealing properties. Moreoverit contains as extreme cases the two well-known values for thesegames: the Nowak-Radzik and the Sánchez-Bergantiños values.

2 - Consistency and graphs in cooperative gamesJosé Manuel Zarzuelo, Josune Albizuri

We use graph theory to study situations in which there is a tradeoff between cooperation and noncooperation. One of the proper-ties used to find solution concepts in cooperative game theory hasbeen consistency. Roughly, this property requires that whenever asubgroup of players are paid and leave the others in a reduced prob-lem, then the payoffs of the remaining agents do not change. Weanalyze the consistency property in situations where the cooperationbetween players is modelled by graphs. In particular, we character-ize the Myerson value (1977) by means of consistency.

3 - Strategic Interactions Among Franchisees: AreFranchisors Always Better Off?Guiomar Martin-Herran, Simon-Pierre Sigue, GeorgesZaccour

We examine the issue of cooperation in a franchise system, a fran-chisor dealing with two adjacent franchisees. Both vertical and hor-izontal interactions are considered. A Stackelberg differential gamewith the franchisor as leader is formulated. Cooperative and non-cooperative scenarios are compared in terms of optimal decisionsand outcomes.


The Future of IFORSStream: IFORS 50th AnniversaryPanel sessionChair: Michael Trick

53

RD-13 IFORS 2008 - Sandton

1 - The Future of IFORSMichael Trick, , , Alexis Tsoukiàs, Andrés Weintraub

What is the future of the International Federation of Operational Re-search Societies (IFORS)? How do social and technological trendsaffect the direction of international professional societies? A "blueribbon panel" of top experts from within and outside of IFORS willdiscuss the future of operational research, national societies, inter-national societies, and IFORS itself.

� RD-13Thursday, 2:00 PM - 3:30 PM - Convention Centre: CommitteeRoom 5

Multicriteria Scheduling

Stream: SchedulingInvited sessionChair: Jean-Charles Billaut

1 - Personnel vs Machine Performance Costs in Open-Shop Scheduling Problems with Time-Windowswith Minimal MakespanDavid Alcaide Lopez de Pablo, Antonio Sedeño-Noda,Carlos González-Martín

This paper deals with several tricriteria open—shop schedulingproblems with pre—emptable jobs. Job time—windows must bestrictly respected. Criteria are personnel and machine performancecosts and makespan. Network flow approaches are sequentially usedin bicriteria open—shop problems with bounded makespan to solvethe considered tricriteria problems.

2 - Bi-Objective Due Date Setting in Make-to-OrderManufacturing by Integer ProgrammingTadeusz Sawik

New bi-objective due date setting problem and integer programmingformulations are presented. The problem objective is to select max-imal subset of orders that can be completed by customer requesteddates and to quote delayed due dates for the remaining orders tominimize total or maximum tardiness of the orders.

3 - A very fast algorithm for finding non dominated so-lutions for the bicriteria flexible jobshop schedul-ing problemJean-Charles Billaut,

We consider a bicriteria flexible job shop scheduling problem(makespan and maximum lateness objectives) for which we deter-mine a set of non-dominated solutions. We adapt the Pareto LocalSearch (Paquete 2004) to our problem and compare these solutionsto other metaheuristic algorithms. The new algorithm returns inter-esting results both in terms of computation time and quality.

� RD-14Thursday, 2:00 PM - 3:30 PM - Sandton Sun: Maroela North

DEA: Applications in farms andagricultural

Stream: DEA and Performance ManagementInvited sessionChair: Antonio G.N. Novaes

1 - Modeling Milk Cost in Estonian Farms: A Stochas-tic Frontier Analysis ApproachReet Poldaru, Jyri Roots

The purpose of this paper is to analyse stochastic frontier modelsfor milk cost in Estonian farms. The data is an unbalanced panelof 149 Estonian milk producers drawn from FADN observed duringthe period 1999 to 2005. In this paper we consider SFA as an al-ternative method for econometric model parameters estimation andas an instrument to predict technical efficiencies of farms. Two dis-tinct models of milk cost are compared and discussed: a) almost allessential independent variables are included in the model; b) someessential independent variables are not included in the model.

2 - Farming system performance and water use effi-ciency in Tunisian semi-arid regionFraj Chemak

The objective of this investigation is to analyze the performance ofirrigated farms in the region of Sidi Bouzid. Farmâ EUR TMs sur-vey was carried out in both public and private irrigated areas. Theapplication of Data Envelopment Analysis (DEA) approach allowedthe measurement of technical and allocative efficiency. The com-parison between both production systems revealed more importantresources wasting in private farms than in public ones. On the otherhand, the allocative efficiency of private farms appears to be muchmore interesting than that of public ones.

3 - Applying research on the management of fishingcapacity for Chinese inshore fleets by DEA methodyi zheng

The fishing capacity and capacity utilization for Chinese inshorefleets in the latest 13 years were measured by DEA method. Theresults reflected: the resource declining and over fishing in BohaiSea were the worst in China. Then the relevant models between ca-pacity output, capacity utilization and income were set up, and thefunction of collecting tax for controlling fishing capacity was quan-titative simulated. It pointed out: tax system would be effective forcurtailing fishing capacity and improving the efficiency of the wholefishing industry in China, if only the tax rate isnâ EUR TMt too low.

� RD-15Thursday, 2:00 PM - 3:30 PM - Sandton Sun: Maroela South

OR for Sustainable Development 2

Stream: OR for Sustainable DevelopmentContributed sessionChair: Marc Despontin

1 - Introducing systems dynamics in multi-criteriaanalysis: An application to the evaluation of inno-vative road safety measures.Klaas De Brucker, Cathy Macharis, Alain Verbeke

This paper introduces concepts from systems dynamics into the fieldof MCA, with a focus on improving the generation of alternatives.We demonstrate the significance of this approach through an applica-tion in the area of road safety measures. Here, we adopt a multi-actorMCA to allow various stakeholders with an interest in improving thepresent state of forgiving and self-explaining roads to understandwhich future states lead to the highest expected value added for thecommunity of stakeholders in its entirety, thereby uncovering infor-mation on the probability of successful implementation.

2 - Use of models in natural resource managementdecision-makingOliver Chikumbo

Decision-making for natural resource management has evolved overthe years to a "green-centric’ sustainable development effort thatmust simultaneously achieve profitability and feasibility, social, andenvironmental values. What is resolved is the ’what-when-where’problem for land uses and the ’timing-frequency-intensity’ prob-lem for management alternatives. A case study of a 1492ha farmin the Rotorua basin, New Zealand, highlights the application ofa multi-objective optimisation and simulation modelling based onsound professional judgment (in situations where data were scarce).

54


3 - Multicriteria Measures in Well-beingBrent Bleys, Marc Despontin

In economic policy-making, G.D.P. is often used as a normativebenchmark for the economic and even social performance of a coun-try. Yet,it has many flaws as a measure of well-being.

Many well-being conceptualizations have been provided. These in-clude the index of sustainable economic welfare (ISEW), the humandevelopment index (HDI) and the happy life expectancy (HLE).

In aggregating indicators into one single index, issues of item selec-tion and weighting emerge. Multicriteria analysis offers possibilitiesin this regard. This paper looks to using the Promethee methods.

� RD-16Thursday, 2:00 PM - 3:30 PM - Sandton Sun: Jacaranda

Pickup and Delivery Routing

Stream: RoutingInvited sessionChair: Juan José Salazar González

1 - Multi commodity split pickup and split deliveryproblemsBjørn Nygreen, Frank Hennig

We discuss alternative formulations of the multi commodity splitpickup and split delivery problem without predefined pairing of loca-tions. Some commodities have only one pickup or delivery location,while other commodities can have several locations both for pickupand delivery. Only commodities with only one pickup point and onedelivery location are paired. We assume that we have available aheterogeneous fleet of vehicles (with emphasis on ships) which cancarry several commodities simultaneously. We plan to report com-putational results for small test cases solved by different approaches.

2 - Performance Analysis of Procedures for the dy-namic Multi-Load Pickup and Delivery Problemwith Time WindowsSteffen Schorpp, Bernhard Fleischmann

In a dynamic environment Logistics Service Providers have toquickly adapt the current schedule to new information (orders, traf-fic situation), so as to maximize profit and customer service. Thepaper presents two fast optimization approaches for this problem(assignment and insertion/improvement based) and reports on com-putational tests with realistic data.

3 - Open location management in automated stor-age/retrieval systemsYugang YU, René de Koster

This paper studies the management of opening locations in Au-tomated Storage/Retrieval Systems (AS/RS) by minimizing themakespan of a block of storage and retrieval jobs. We propose an ef-fective storage area (ESA) policy that defines a part of the openingsas effective, while the others as ineffective. Based on NN (NearestNeighbor) heuristic, a model is proposed and solved to determinethe optimal number of effective openings and the boundary of theESA. Our results show that the makespan can be reduced by morethan 10% averagely by using the ESA policy.

4 - Pickup-and-Delivery TSP to serve customers withstochastic demandsJuan José Salazar González,

This talk shows how to set the vehicle capacity for Travelling Sales-man Problems where some of the customer demands are stochastic.The analyses are done for the one-commodity Pickup-and-DeliveryTSP, as this problem also includes the setting of the initial load. Thepaper considers feasibility issues. This includes finding the smallestvehicle capacity and some initial load such that a given tour is feasi-ble for all scenarios. Different variants are considered as a functionof the time when information becomes available. Computational re-sults are presented.

� RD-17Thursday, 2:00 PM - 3:30 PM - Sandton Sun: Oleander

Energy: General

Stream: OR Applications in the Energy SectorInvited sessionChair: Charles Mbohwa

1 - Sizing PVP systemsVasco ChikwashaIn designing photovoltaic power systems, the coordination of the re-newable energy source, the power generating source, energy storageand the load is very complicated. A sizing method using geneticalgorithms is proposed here.

2 - The impact of the forward market on producers’market power

Increasing of efficiency for homogeneous good markets (electricitymarkets) is an important problem. One way to reduce the marketpower is to introduce the forward market. Bushnell,2005, studies amodel of the two-stage market and estimates reduction of the devi-ation from the competitive equilibrium due to such reconstruction.However, this paper proceeds from the bounding assumption aboutequal prices at the forward and spot markets. We consider severalvariants of the forward market organization and estimate the devia-tion of the Nash equilibrium outcome from the competitive equilib-rium.

3 - The Application of Simulation Modeling to SupportMulti Billion Rand Infrastructure Investment Deci-sions in Developing EconomiesAnton du Plooy, Cecil Lourens, Louis van der WaltA new multi billion Rand products petroleum pipeline system will beconstructed from the port of Durban to the South African economicheartland, Gauteng. The system comprises a 24’ pipeline with accu-mulator terminals on either side of the pipeline. The key objectivesof the simulation modeling were to validate the design parametersand determine the operational feasibility of the new pipeline systemwithin the complexities of the overall supply infrastructure.

4 - Life Cycle Assessment of a Coal-fired ThermalPower Plant in ZimbabweCharles Mbohwa, NKOSINATHI NCUBEThis paper discusses a life cycle analysis of a thermal power plantto quantify environmental impacts of thermal power production atMunyati Power Station in Zimbabwe. Thermal power stations emitcarbon dioxide, sulphur dioxide, nitrous oxides and solid waste andthis study analyses these for thermal power systems.

� RD-18Thursday, 2:00 PM - 3:30 PM - Sandton Sun: Syringa


Stream: Cutting and PackingInvited sessionChair: Marcos Arenales

1 - Four new algorithms for the two-dimensional two-phase variable-sized bin packing methodFrank Ortmann, , Jan van VuurenEfficient heuristics have been developed in order to find approximatesolutions to the variable-sized bin packing problem. These includemethods that pack items into large bins first, then repacking intosmaller bins. Some two phase algorithms have been developed forthe fixed-sized bin packing problem. Here we present a two-phaseapproach to rectangular, oriented, two-dimensional variable-sizedbin packing, comparing four new strip packing algorithms for thefirst phase with four previously published strip packing algorithms.

55

RE-02 IFORS 2008 - Sandton

2 - A waste reduction algorithm based on alternativeunderestimates for the modified Wang method, Giel Hattingh, Tjaart Steyn

In this presentation different heuristic functions are investigated andapplied to the modified Wang method. This involves the sharpeningof underestimates used in the method’s heuristic function. Resultsobtained on problem instances reported in the literature as well assome real life problems from the glass industry are given.

3 - The two-dimensional cutting stock problem withusable leftoversAdriana Cherri, Andréa Vianna, Marcos Arenales

Consider two-dimensional cutting stock problems that leftover mate-rials in cutting patterns can be used in future periods, if large enough.To solve this problem, characteristics of good solutions, taking intoaccount mainly low waste and small leftovers, are defined and modi-fications an AND/OR-graph approach are proposed. Computationaltests are given.


� RE-02Thursday, 4:00 PM - 5:30 PM - Convention Centre: Ballroom 2

Knapsack Problems and Extensions

Stream: OptimizationContributed sessionChair: Stephan Visagie

1 - Solving the Extended Tree Knapsack problem(ETKP) with fixed cost flow expansion functions, Giel Hattingh

In Local Access Telecommunication network design Tree Knapsackproblems play a major part. Large numbers of potential nodes makethese problems almost intractable. Algorithms for exact solutionsare presented and the performance investigated by empirical exper-iments. Fixed cost models for flow expansion at nodes are consid-ered.

2 - A Genetic Algorithm for the Generalized QuadraticKnapsack ProblemTugba Saraç, Aydin Sipahioglu

The quadratic multiple knapsack problem (QMKP) is an extensionof the well known quadratic knapsack problem. In the QMKP, thereare k knapsacks with (possibly) different capacities instead of a sin-gle knapsack. In this study, the QMKP is generalized by addingnew constraints that they occur in real life problems. Addition-ally, a new genetic algorithm is proposed for solving the generalizedQMKP. The results of the proposed algorithm are compared withthe achieved best solutions of randomly generated test instances byusing GAMS/DICOPT solver. The effectiveness of our approach isalso shown.

3 - Branch-and-bound solution techniques for a spe-cial class of nonlinear knapsack problemsStephan Visagie

Convex enveloping with branch-and-bound techniques is presentedto solve a special class of nonlinear, separable, nonconvex knapsackproblems. Three algorithms are presented which improve the solu-tion times of the standard branch-and-bound technique considerablywhen determining an optimal solution. These algorithms and theirimprovements on solution times are also compared.


Military Simulation and StochasticModelsStream: Military Applications of ORInvited sessionChair:

1 - Modeling Civil Defense Operations for TrainingAmnon Gonen,

The main objective of this paper is to describe the modeling ap-proach of Civil defense and to create the most realistic environmentfor the training decision makers in civil defense missions. The paperpresents part of a broad effort of modeling the OOTW missions in asimulation environment built to train both military and civilians. Thetrainees include commanders of rescue teams, disaster management,military units, fire fighters, police, and other leaders of the civilianauthorities

56

IFORS 2008 - Sandton RE-06

2 - Optimal Location of Canadian Forces OperationalSupport HubsAhmed GhanmiThe Canadian Forces (CF) are looking to establish international sup-port hubs to improve their ability to sustain missions. The hubswill be used to preposition non-perishable supplies to improve thelogistics responsiveness. In this paper, an optimization model wasformulated to determine the optimal hub locations under various op-erational constraints. Performance metrics were developed to assessthe logistics distribution effectiveness using the hub-based supportconcept. This study addresses a complex military hub-and-spokeproblem and contributes to the CF operational support development.

3 - Aggregated Force Attrition with a Dynamic Compo-sition of Models

We look at the problem of aggregated force attrition in ground com-bat. Traditionally, this problem is solved using Lanchester-stylemodels. We propose a new methodology which uses high-resolutionMonte-Carlo models and synthesizes their results to create aggre-gated attrition results. The aggregated model relies on parametersestimating the fire opportunities for low-level entities throughout theprogression of ground combat. We present the rationale of the modeland its application to wargaming.


Job Shops and Assembly Lines

Stream: ManufacturingContributed sessionChair:

1 - Multiobjective Particle Swarm Optimization forcomplex job shop schedulingJacomine Grobler, Andries P. Engelbrecht, VenkataYadavalli,This paper investigates multiobjective Particle Swarm Optimiza-tion strategies for addressing a complex variation of the job shopscheduling problem. Production downtime, sequence-dependentset-up times, release dates and multiple predecessors per job were in-cluded and benchmarking against currently used algorithms on realcustomer data demonstrated significant performance improvement.

2 - Efficient Mixed-Model Assembly Lines for HighProduct Variety - A practical ApproachSimon Altemeier,Known methods for balancing assembly-lines mostly neglect the ef-fects of high product variety resulting in an increasing variance inthe workload at different assembly stations. Poor assembly quality,high costs for additional workers or even line stops are the result. Alocal search approach based on a reassignment of tasks to stationsgaining efficiency in production and reducing the described effectsis introduced.

3 - Variance Reduction in Assembly LinesJan Erik Gans,A major problem in final assembly lines for mass customizationproducts is the variance of the work load of the assembly stations.One approach to handle this problem is the splitting of the line. Amethod is presented that finds variance minimal assignments of vari-ants to assembly line segments.

� RE-05Thursday, 4:00 PM - 5:30 PM - Convention Centre: Boardroom1

Human Resources Management

Stream: Human Resources ManagementContributed sessionChair: Marthi Harmse

1 - Analysis of the Total Productive Maintenance Fac-tors for a Productivity and Quality Increase in Ko-reaOH Yon -Woo

The TPM program, which is a methodology for improving manage-ment results through the management of the innovative activities ofa company, has been widely introduced in the field of the serviceindustry as well as the manufacturing industry. The education andtraining for employes is most important for the improvement of themanagement results through the TPM program, and an individual’simprovement activity is also important.

2 - Remuneration for management complexity: anMCDA application,

A number of academic managers were to be compensated accordingto the complexity of their tasks. Using structured inputs from thegroup concerned, the complexity of different subtasks were scaledto a common basis. From this the relative complexity of the task ofeach manager was found and the experience curve applied norma-tively to allocate the given budget.

3 - Cultural transformation of petrochemical industryMarthi Harmse

A cultural transformation of an organisation requires amongst oth-ers the alignment of their values and beliefs with their actions andbehaviours as codified in the collective procedures, processes andstructures of governance, according to Barrett (2006). This paperdiscusses the preparation phase for such an alignment in a petro-chemical industry.


Data Mining 2

Stream: Data Mining & Knowledge ManagementContributed sessionChair: Anatoly Zhigljavsky

1 - The Study of TRADEOFF in Conjoint Analysis, Hiroaki Ishii, Shogo Shiode

Conjoint analysis is a scaling method used in many of the socialsciences. It’s used for measuring each factor’s contribution to thewhole evaluation of products consisting of some factors. TRADE-OFF is a method of conjoint analysis, using when the evaluation isnon-metrical data. The part worth values obtained by TRADEOFFgive an approximate comparison of each factor’s contribution to thetotal evaluation, but it’s impossible to utilize them for statistical use.Moreover, their solution isn’t unique. So, we show the problems ofTRADEOFF and then propose their solution.

2 - Selecting Predictive Models for Inclusion in an En-sembleKweku-Muata Osei-Bryson, Muhammadou M.O. Kah,Jainaba M.L. Kah

For sometime it has been well known that for some datasets anensemble of individually trained predictive models can outperformeach individual model. Many ensemble generation techniques typi-cally involve using some form of weighted or unweighted voting tocombine the given set of candidate classifiers. In this paper we areconcerned with ensembles that may include multiple classifier types,and in which not every candidate classifier is automatically includedin the ensemble.

57


3 - Mixed Poisson models with random arrivals, Anatoly Zhigljavsky

Assume that we have a large population. In this population, eachindividual may enter a panel at a random time. The individual staysin this panel for a random period of time and then leaves the panel.During the period of time the individual stays in the panel, he hasevents that occur according to a Poisson process with random inten-sity. We assume that the arrival process is stationary and that thedurations of stay in the panel for different individuals are indepen-dent and have the same distribution. The main aim of study is theestimation of this distribution.


Improving policy in healthcare

Stream: OR and HealthcareInvited sessionChair: Hazel Pilgrim

1 - A novel case based reasoning approach to radio-therapy planningNishikant Mishra, Sanja Petrovic, Santhanam Sundar

Radiotherapy planning is a complex problem which needs both ex-perience and expertise. A novel case based reasoning is proposed,which suggests the dose and parameters of radiations for new patientusing the memorized cases of the previous patients. The proposedmethodology is validated on real data collected from a hospital.

2 - Cost-effectiveness of spinal cord stimulation forchronic pain of neuropathic origin in the UKAlejandra Duenas, Emma Simpson, Jim Chilcott, MichaelHolmes

Neuropathic pain is produced by damage to, or pathological changesin the peripheral or central nervous systems. We analysed the cost-effectiveness of spinal cord stimulation in the treatment of neuro-pathic pain in the UK considering failed back surgery syndrome andcomplex regional pain. Sensitivity analyses were performed varyingthe key parameters.

3 - Cost-effectiveness of antenatal anti-D for pregnantwomen in England and WalesHazel Pilgrim, Myfanwy Lloyd-Jones, Jim Chilcott, JohnStevens

Antenatal anti-D is routinely provided to RhD-negative pregnantwomen in England and Wales to prevent haemolytic disease of thenewborn which could lead to a small proportion of babies experienc-ing developmental problems or being stillborn. We review whetherroutine antenatal anti-D remains to be cost-effective within Englandand Wales. Valuation of quality of life is considered and bias withinthe efficacy evidence base associated with different regimens of anti-D is assessed using elicitation of experts.


Decision Analysis: Methods andProcessesStream: Decision AnalysisContributed sessionChair: Moshe Sniedovich

1 - The errors underpinning the St Petersburg paradoxRobert Vivian

The most fundamental decision rule is the expected value rule, for-mulated by Pascal and de Femat in the 1600s. In 1713 NicholasBernoulli cast doubts as the correctness of this rule when he for-mulated a number of problems the most famous leading to the StPetersburg paradox. However if the expected value of the game iscorrectly calculated there is no paradox. This paper will examinethe errors which underpin the traditional calculation of the expectedvalue of the St Petersburg game. The same errors are encounterd inmany solutions to games of chance.

2 - Adaptive decision analysis in planning of naturalresourcesPekka Leskinen, Olli-Pekka Tikkanen, Teppo Hujala, ,Annika Kangas, Mikko Kurttila

For effective planning of natural resources, we developed an adap-tive decision analysis framework by which the decision analysis pro-cess can be adapted with respect to the characteristics of the decisionmaker and the decision problem. The framework involves a) a prioriselection of the general decision support framework and b) adaptiveprocessor, which evaluates the acceptability of the overall solutionof the decision problem. The approach involves both qualitative andquantitative decision analysis tools in a unified framework.

3 - Evaluation of decision making practicesK. Nadia Papamichail, Stephen Brewis,

Nowadays, managers have to take increasingly complex decisionsin the face of uncertainty and conflicting information. Poor decisionmaking may have a significant adverse impact on the viability oftheir organisation. This work presents an evaluation framework andan intelligent decision support system for assessing decision mak-ing capabilities within decision lifecycles. The setting is a projectinvolving a large investment in a large telecom company. Examplesof good practice in decision making are presented.

� RE-11Thursday, 4:00 PM - 5:30 PM - Convention Centre: CommitteeRoom 2

Production Planning & Control

Stream: Production and Inventory SystemsInvited sessionChair: Leorey Marquez

1 - A GRASP procedure for scheduling orders of mul-tiple products on parallel machines with setupsManuel Mateo, Imma Ribas, Ramon Companys

We deal a scheduling problem where some orders are composed bya set of batches of products (classified in families). Each order isformed by several products of different families, with their respec-tive amounts, and a due date. There are some not related parallelmachines. For each kind of product, a machine has a processingtime and a setup time, which appears when two consecutive prod-ucts are from different families. The problem consists in assigningeach batch from the orders to one machine and scheduling the ma-chines, to minimize the mean tardiness. The proposed procedure isa GRASP.

2 - Petri Nets of the Materials Flow at the Steel PlantPiotr Lebkowski

A Petri time network applied to the modeling of the material flowat the production plant in steelworks is a effective research tool.This paper presents a Petri net expanded with the attribute vectorsof places and transitions, as well as the logical rules of transitionfiring and the procedures that update the attribute values. Owing tosuch an expansion, we can observe the properties of the streams thatflow through the system, e.g. the costs and the project completiontime, simulate the effects of the changes introduced into the systemand analyse the results of unexpected disturbances.

58

IFORS 2008 - Sandton RE-15

3 - Arena in the Petrochemical Operations Environ-mentLorraine Malherbe

The use of discreet event simulation within Sasol, an internationalpetrochemical company, is fast becoming an important tool for sup-porting strategic decision making in the continuous operations envi-ronment. This study examines the use of Arena to identify capacityconstraints to investigate the feasibility of new undertakings for ex-pansions at Secunda.

4 - Optimal Stockyard SchedulingLeorey Marquez, David Sier, Andreas Ernst, ,

Stockyard scheduling involves the movement of material through aset of stockpiles. The objective is to maximise throughput subject toconstraints such as capacity limits, and allowed machine and stock-pile interactions.

This paper describes a system designed to prepare optimal stockyardmanagement plans for large scale mining operations.

� RE-13Thursday, 4:00 PM - 5:30 PM - Convention Centre: CommitteeRoom 5

Logistic Scheduling

Stream: SchedulingInvited sessionChair: Jens Peter Kempkes

1 - Integrated Project and Vehicle Routing Planning forConstruction ProjectsFrank Schultmann, Nicole Sunke

Logistics for construction projects is subject to numerous con-straints. An integrated model for resource-constrained projectscheduling and vehicle routing planning is introduced that can beapplied for the coordination of construction activities and related lo-gistics tasks.

2 - A Two-stage Scheduling Model for the In-flightCatering ServiceSan Nah Sze, Daniel Oron, Ada Suk Fung Ng

This paper studies the in-flight catering scheduling problem. Agroup of loading teams have to transport food from the kitchen andload it into aircrafts, which land and depart within tight time win-dows. The number of aircraft that can be served in a trip is restrictedby the capacity of the loading truck and a food exposure time limit.A two-stage scheduling model is proposed to assign each loadingteam to aircrafts and decide the shift hour of each team in order tominimize the number of loading team needed.

3 - Supply Network Optimization in OperationsJens Peter Kempkes

We will gain insight into detailed scheduling of logistic operationsin inbound logistics, which consider a wide range of degrees of free-dom (transport modes, trays, stock, and transport paths). Even ad-vanced MIPs can not handle those complex models in reasonabletime.

A new heuristic approach to achieve appropriate solutions in a prac-tical period of time is presented.

� RE-14Thursday, 4:00 PM - 5:30 PM - Sandton Sun: Maroela North


Stream: DEA and Performance ManagementInvited sessionChair: Wenbin Liu

1 - Fairly Sharing a New Total Fixed Input Variable withParametric DEAArmando Zeferino Milioni, Elisabeth Freitas Rodrigues,

In this paper we address the problem of fairly assigning shares of anew total fixed input to a group of Decision Making Units (DMUs)using Data Envelopment Analysis (DEA), by assuming the existenceof a predefined locus of points that characterizes the DEA frontier(parametric DEA). We study the case in which the new input vari-able assigned to each DMU affects the values of the output variablesof that DMU.

2 - Framework for problem classification in DEADieudonne Kabongo Kantu, Theodor Stewart

We develop a typology of problems addressed by Data EnvelopmentAnalysis (DEA), and review reported applications of each. Differ-ent models for DEA are cross-linked with the problem typology toidentify needs for future development particularly for obtaining ro-bust solutions i.e. not sensitive to extreme weights.

3 - DEA Performance Indexes Based on Distances toEfficient and Anti-efficient FrontiersWenbin Liu

In this work we developed some DEA performance indexes usingthe distances to both the efficient frontier and the anti-efficient fron-tier to enhance discrimination power of DEA. These performanceindexes can rank those decision making units on the efficient fron-tier. The standard DEA models and the models with undesirablevariables are used to identify the efficient and anti-efficient frontiersrespectively. Empirical study showed these performance indexes in-deed have much more discrimination power. We also carried out asimulation study to test their performances in ranking efficient units

� RE-15Thursday, 4:00 PM - 5:30 PM - Sandton Sun: Maroela South

Systems Approaches to Public SectorProblemsStream: OR and the Public SectorContributed sessionChair: Rui Marques

1 - Developing a Systems Analysis Framework for theNational Poverty Alleviation SystemMarita Turpin, Mario Marais, Leanne Scott

The CSIR and UCT researched the application of various SystemsThinking methods to understand and evaluate poverty alleviationefforts in South Africa. The National Poverty Alleviation System(NPAS) is defined by means of its various role players (government,NGOs, CBOs, etc). We developed a framework to describe and eval-uate a system with technical, social and economical components, us-ing the Multiple Perspectives Approach as a basis. The discussionwill focus on our first case study where the framework was appliedto the Community Based Public Works Programme.

2 - Measuring the influence of operational environ-ment in the Portuguese solid waste servicesRui Marques, Pedro Simões, Kristof De Witte

The aim of this research was to compute the influence of operationalenvironment in the efficiency on Portuguese solid waste services.We use a sample of 29 solid waste utilities encompassing all thecountry. Several methodologies are used. Particularly, we used thedouble bootstrap to estimate the effect of various explanatory factorson DEA results obtained in the DEA efficiency. We found the rele-vant influence of Gross Domestic Product (GDP), efficiency as wellthe distance to the sanitary landfills and the population density.

59


� RE-16Thursday, 4:00 PM - 5:30 PM - Sandton Sun: Jacaranda

Applications in logistics

Stream: Freight Transport & LogisticsContributed sessionChair: Anne Lange

1 - A New Approach to Planning Tours for a LogisticsService ProviderAydin Sipahioglu, Ümit Ali Özkazanç

Problems faced by a logistics service provider are not only a vehiclerouting or a bin packing problem. It consists of clustering/groupingsuppliers, routing vehicles depend on time windows and loadingtrucks depend on stackability of the boxes. In this study an Inte-grated approach -focusing on all these three problems- is proposedfor the planning of tours and shipments within a Decision SupportSystem being able of serving alternative solutions. Proposed ap-proach is tested on some literature test problems and also imple-mented on real instances via the past planned trips of a logistics firm.

2 - Modeling and Measurement of Ethanol Transportby Hidrovia Tietê-ParanáRui Carlos Botter

The model makes possible the evaluation of a fluvial convoy fleetand its characteristics, including terminals and tankage necessary ineach store terminal allocated along the Tietê-Paraná Waterway. Itrepresents all restrictions to navigation such dams, narrow bridgesand channels, atmospheric conditions and operation fleet regime.The model enables evaluation of economical viability, includingchoosing the most viable scenery among more than 200 operationaloptions that attend product demand associated to pre-established ser-vice levels.

3 - Optimal use of current logistics infrastructure andplanning for expansionLeilani Morison

The South African economy is seeing substantial growth. How-ever, the current logistics infrastructure is constrained. Especiallythe Durban port is very congested. A liquid chemicals companywith considerable export volume growth thus has an outbound lo-gistics challenge. To develop a logistics strategy, use is made ofvarious OR techniques. This presentation focuses on a model devel-oped to optimise the current bulk-liquid storage and further modelexpansion to provide an optimised strategy for addition of storageinfrastructure

4 - The impact of prioritization on a LTL networkAnne Lange

General cargo networks (less-than-truckload, LTL) often provide theidentical service level to all customers. This paper examines the im-pact of transporting consignments with varying priorities in a LTLnetwork, as done for letter mail. A large scale simulation inspired byreal-life offers valuable insight in the prioritized network’s behavior.It allows identifying trade-offs between incurred cost and providedservice. It thus provides a basis for offering multiple service classesto customers and supports pricing strategies. Further research willanalytically probe the simulated results.

� RE-18Thursday, 4:00 PM - 5:30 PM - Sandton Sun: Syringa


Stream: Cutting and PackingInvited sessionChair: A. Miguel Gomes

1 - Optimal One-Dimensional Assortment Cutting fromFixed Stock LengthsFrancois van Staden, Giel Hattingh, Tjaart Steyn, PeterForbrig,

Exact algorithms similar to the Wang-type algorithm for the two-dimensional guillotine type cutting stock problem are discussed.This paper considers the problem of cutting one-dimensional ma-terial for a given order using fixed stock lengths. Algorithmic per-formance is presented based on empirical experiments.

2 - Variable Neighborhood Search for the Three-Dimensional Bin Packing ProblemVinícius Armentano, Marcelo Tozo de Araujo

For a set of rectangular-shaped items, we address the problem oforthogonally packing all items into the minimum number of three-dimensional rectangular bins. A variable neighborhood search ap-proach is proposed for solving such a problem. Computational testson instances from the literature show that the proposed method out-performs other approaches.

60

IFORS 2008 - Sandton FA-03

Friday, 8:00 AM - 9:30 AM

� FA-01Friday, 8:00 AM - 9:30 AM - Convention Centre: Ballroom 1

Stochastic ModelsStream: Stochastic ModelsContributed sessionChair: Mardi Jankowitz

1 - Combining decomposition and dynamic program-mingPierre Girardeau, Kengy Barty

We consider a large scale multi-stage stochastic optimization prob-lem with at least a coupling constraint at each step-time. We ob-tain subproblems formulation by the use of a Lagrangian relaxationtechnique, afterward we perform a primal-dual algorithm to approx-imate iteratively the optimal solution. In order to use dynamic pro-gramming algorithm for subproblems resolution, randomness is ap-proximated by mean of a Markov model. We provide a numericalexperiment related to electric power generation.

2 - New developments in nonlinear smoothingMardi Jankowitz

Smoothing of time series plays an important role in various disci-plines, such as the financial and mediacal fileds. Nonlinear smooth-ing became more popular during the last few decades with numer-ous new developments. A new class of Lower-Upper-Lower-Upper(LULU) smoothers have been studied and compared to the mostused nonlinear median smoother.


Nonlinear Optimization

Stream: OptimizationContributed sessionChair: Paulo Ferreira

1 - Robust Non-Linear Regression Models with Mini-mal Assumptions, Giel Hattingh, Hennie Kruger

Managerial decisions are often based on the modelling, evaluationand interpretation of data. In this paper, mathematical programmingtechniques are investigated that may contribute to the developmentof non-linear models. The feasibility of robust regression for suchnon-linear models, with minimal assumptions, is investigated.

2 - Motion Planning of Robot Manipulators Based onVariational CalculusAlberto Olivares, Ernesto Staffetti

We study the motion planning problem in robotics. Given an articu-lated object together with a set of constraints and an optimality crite-rion we find a general strategy which provides the trajectory and theactuator forces that satisfy the given set of constraints, and globallyoptimizes the given criterion. We solve this optimal control prob-lem using variational calculus. Our method consists of an uncon-strained reformulation, in which the dynamic equation is regardedas an additional constraint, a discretization, and of the resolution ofa non-linear system of difference equations.

3 - A Global Optimization Algorithm for the NonlinearSum of Ratios ProblemPaulo Ferreira, Rúbia Oliveira

An outcome space approach for minimizing a nonlinear sum of ra-tios over a convex set is proposed. Ratios of convex and concavepositive functions are considered. Convex and multiobjective pro-gramming concepts and methods are used to representing the orig-inal nonconvex problem in the outcome space as a linearly con-strained indefinite quadratic optimization problem with special char-acteristics. Numerical experiences based on test problems from theliterature demonstrate the efficiency of the global optimization ap-proach proposed.


Military Threat Evaluation / DSS

Stream: Military Applications of ORContributed sessionChair: Jan van Vuuren

1 - Rational-analytic decision making versus natural-istic (or intuitive) decision-makingWinnie Pelser

Analysts and decision-support practitioners favour the rational-analytic style of decision making (formal methods). In reality manyreal-world high-level officers tend to use intuitive approaches basedon their experiences.

It is widely believed that the rational-analytic style is obviously rightand that high level decision makers should adhere to it. The onlyproblem is that the intuitive style of decision making is often veryeffective.

Is it possible to reconcile the two approaches? In this presentationwe will look at this situation for military applications.

2 - Design of a threat evaluation decision support sys-temJaco Roux, Jan van Vuuren

The design of a real-time threat evaluation decision support sys-tem for a ground based air defence environment is discussed. Morespecifically, the threatening behaviour of fixed wing aircraft towardsa number of defended assets is assessed or quantified, aiding theair defence operator in an environment which requires rapid opera-tional planning and decision making under severe stress conditions.The aim of this talk is to suggest a threat evaluation model hierar-chy consisting of flagging (qualitative), deterministic and probabilitybased (quantitative) models.

3 - Probability of attack by a fixed wing aircraft in aground based air defence environmentjacques du toit, Jaco Roux, Jan van Vuuren

In a Ground Based Air Defence (GBADS) environment, the threatposed by fixed wing enemy aircraft relative to protected assets isassessed by a decision support system known as a Threat Evalua-tion (TE) system. This paper is concerned with a sub-component ofa probabilistic TE model which aims to estimate, in real-time, theprobability that an enemy aircraft, flying a particular profile, mayattack a specific asset within a given time window. This approachnecessitates the consideration of aircraft kinematic quantities to es-tablish a collection of feasible future flight paths.

4 - Evaluation of a threat evaluation and weapon as-signment systemBasie Kok, Jan van Vuuren, Jaco Roux

In this paper the aim is to test a conceptual Threat Evaluation andWeapon Assignment (TEWA) system by evaluating various TEWAsub-components for different asset deployments. This process istwo-fold in the sense that TE sub-components monitor and assign avalue of threat to all hostile airborne targets with respect to protectedassets and then the WA sub-components utilise available weaponsystems with the aim of defending the most important assets. Evalu-ation of the system then ensues by means of performance measuressuch as asset preservation and resource utilisation efficiency.

61

FA-05 IFORS 2008 - Sandton

� FA-05Friday, 8:00 AM - 9:30 AM - Convention Centre: Boardroom 1

OR and EthicsStream: OR and EthicsInvited sessionChair: Leroy White

1 - Ethical issues in tracking cell phones at an eventHans W. Ittmann, Antony Cooper, Theodoulos Stylianides,Peter Schmitz, Chris Elphinstone

This paper describes an experiment to track the cellular telephonesof a small group of people as they moved to and from an event, withthe intention of assessing the viability of using such tracking to pro-vide the participants with useful traffic information. This projectraised a number of ethical issues, including the ethics of modellingdata, which will be addressed in the paper. We also provide anoverview of the experiment and the results obtained.

2 - A Meta-Model for Values and EthicsCathal Brugha

We present a meta-model which shows that ethical and values-baseddecisions require balance between one’s subjective position/ objec-tive realities, and between one’s inner world/ the outer world. We usethe model to show how values adduce (draw out) one’s behaviour,and ethical behaviour decision-making follows a conflict resolutionadapting process.

3 - Role of the Dialogue and Arguments in Decison -MakingBernard Reber, , RAZAFIMAHATOLOTRA Dawidson,Caroline Guibetlafaye

Researchers try to integrate the analysis of dialogue and argumentinto the theory of institutional design. We shall argue that this couldhave an impact on OR-MS research because authority and decision-making capacities in organisations turn out to be dependent on argu-mentative procedures, their ethical properties and their results.

4 - OR, ethics and research ethics committeesLeroy White

This article is concerned with Operational Research (OR) practicein light of growing concerns about ethical conduct. It asks whetherOR, in the context of increasing regulation through research ethicscommittees, should consider whether there are certain ethical issuesthat are affected by the specific context of OR. The article sets outsome of the central concerns about research ethics committees andthe nature of OR.


Data Mining and KnowledgeManagement

Stream: Data Mining & Knowledge ManagementContributed sessionChair:

1 - A Decision Rule-based Method for Feature Selec-tion in Predictive Data MiningPatricia Lutu, Andries P. Engelbrecht

Feature selection is an essential step in any predictive data miningtask. In the area of classification, commonly used approaches tofeature selection involve feature ranking based on correlation withthe class, followed by forward selection search which uses a mea-sure of merit to determine the next best feature to add to the subsetof selected features. It is argued that the use of merit measures isnot always the best approach for feature subset selection. The pro-posed method employs decision rules based on domain knowledge,to guide the search for the best subset of features.

2 - Pattern Languages — a novel facilitator for dataminingKathrin Kirchner, , Johannes Ruhland

The potential of data mining can only be unleashed through coor-dinated choice of both core algorithms and pre- and postprocessingsteps. To holistically guide a non- expert user in mapping needs anddata idiosyncrasies to a process, we propose a building block ap-proach that is guided by a usage of Patterns which in its scale goesfar beyond the well known use in software engineering. Proven as apowerful design principle in architecture, lack of concise and com-plete pattern description languages has hampered widespread use inother fields. We will introduce a prototype for clustering problems.

3 - Building bridges through partnership

Developing a closer working relationship between two existing part-ner organisations with their diverse strengths, expectations and ap-proaches requires an open approach to problem solving. This pa-per highlights the approaches, challenges, opportunities and lessonslearned from the development of these partnerships. Throughvarious initiatives, "pockets’ of trust emerge, an important partof enhancing information sharing and enabling a deeper cross-organisational perspective of knowledge management to develop, toinform future market centred initiatives.


Fuzzy Computing and Its Applications

Stream: Fuzzy Sets and SystemsContributed sessionChair: Ta-Chung Chu

1 - A graph coloring approach for image segmentationDaniel Gomez Gonzalez, Javier Montero, Susana Muñoz

In this work we develop a segmentation scheme for digital imagesbased upon an iterative binary coloring technique that takes into ac-count changing behavior of adjacent pixels. The output is a hier-archical structure of images which allows a better understanding ofcomplex images. In particular, we propose an algorithm that shouldbe considered as image preprocessing technique.

2 - Premium Allocation — Fuzzy Approach in Insur-ance BusinessSanjeev Kumar

With lot of players in insurance market the challenge for holding themarket share and maintain a consistent growth rate has been verytypical. Actuaries have always been on toes in formulating newrules to provide extra leverage over other counterparts in market.Since the factor to which the risk value is attached is always giveto new ideas towards insurance, this approach provides a third di-mension to the existing method of calculating premiums, which isdynamic and may be further refined to address to every individualneeds. Therefore a fuzzy logic approach is used to for the premium.

3 - Evaluating Suppliers through a Mean of Removalsbased Fuzzy MCDM ApproachTa-Chung Chu, Yichen Lin

Suppliers’ selection is a very important part of the supply chain man-agement of a company. A suitable supplier can increase the effec-tiveness of the supply chain of a company and thus improve its com-petitiveness. This work suggests a fuzzy MCDM model to evaluatesuppliers, where the ratings of alternatives versus criteria and theimportance weights of all criteria are assessed in linguistic valuesrepresented by fuzzy numbers. The mean of removals is applied torank all the final fuzzy evaluation values for final decision making.An example demonstrates the feasibility of the proposed model.

62



Decision Support Systems 1

Stream: Decision Support SystemsContributed sessionChair: Arifusalam Shaikh

1 - Diffusion of rumours among different clusters un-der non identical set upAsis Kumar Chattopadhyay

Epidemic models are very useful to explain the spread of rumourbecause of the similarity in underlying processes. The present paperattempts to study diffusion of rumours through a simple epidemicmodel, taking into account two possible sources of rumour viz pri-mary and secondary among several clusters of persons involved. Theobjective of the study is to find out the distribution of time requiredto reach the number of hearers at n (some fixed value) by using theproposed model.

2 - A Decision Support System for Multi-Objective Pro-gramming ProblemsMoeti Ramokgadi

Many concrete problems may be cast into a multi objective opti-mization framework. The redundancy of existing methods, suscep-tible to inconsistencies coupled to the necessity of knowing inherentassumptions before using a given method, make it hard to a nonspe-cialist to choose a method that well fits the situation at hand. More-over, blindly using a method as suggested by the Hammer principle(when you only have a hammer, you want everything in your handto be a nail) is an awkward approach at best and a caricatural one atworst.

3 - Integrating GIS and Genetic Algorithm for a Classof Competitive Discrete Facility Location ProblemArifusalam Shaikh, Said Salhi, Malick M. Ndiaye

GIS systems still lack a sophisticated modeling capability to addressmany of the needs of Location Analysts. A decision support systemis proposed that utilizes the data generating and visualization powerof GIS with a GA based heuristic that can be integrated to obtain thenear optimal solutions. A competitive discrete location problem isconsidered where p facilities are to be located to capture maximumdemand in the presence of competition from other existing facilities.


Group decision making and voting

Stream: Decision AnalysisContributed sessionChair:

1 - Strategic Power in the EU Council in Consultationand Codecision Procedures

We evaluate the distribution of power within the Council of Min-isters of the EU. We don’t make the assessment in isolation of theother EU institutions like is most earlier studies. Specifically, weuse a procedural inter-institutional non-cooperative framework ofEU decision-making and our method of assessment is strategic mea-sure of power of Napel and Widgren (2004) that uses the equilibriumoutcome of the non-cooperative game as the base. We also extendthe previous analyses by considering weighted voting in the Councilin spatial inter-institutional set-up.

2 - Dispersion of Group Judgments: The GeometricExpected Value OperatorLuis Vargas

To achieve a decision with which the group is satisfied, the groupmembers must accept the judgments, and ultimately the priorities.This requires that (a) the judgments be homogeneous, and (b) thepriorities of the individual group members be compatible with thegroup priorities. There are three levels in which the homogeneity ofgroup preference needs to be considered: (1) a single comparison(monogeneity), (2) an entire matrix of comparisons (multigeneity),and (3) a hierarchy or network (omnigeneity). In this paper we studymonogeneity and the impact it has on group priorities.


Regulatory Networks, Inhibitors andDrug Design

Stream: Computational Biology and InformaticsInvited sessionChair: Metin Turkay

1 - Modeling of Regulatory NetworkMetin Turkay, Ugur KAPLAN

Sequencing of whole genome makes huge amount of informationabout living organisms reachable. Abundance of this dependabledata requires development of new system based approach which es-tablishes the emergent relationships between genome or cell compo-nents. Mathematical models about feedback relation within cellularparts and between cell and environment is developed in this study topredict the behavior of living organism on a given conditions.

2 - Mixed-Integer Nonlinear Programming in Structure-Based Drug DesignMetin Turkay,

An important consideration in the structure-based drug design is thegeneration and testing of promising drug molecules. The effective-ness of the drug candidates is highly correlated with their geometriccomplementarity with the active site of the target proteins. A struc-turally efficient drug molecule should be generated for maximumgeometric complementarity. This problem is formulated as a mixed-integer nonlinear programming problem and illustrated on severalcase studies.

� FA-11Friday, 8:00 AM - 9:30 AM - Convention Centre: CommitteeRoom 2

Production Lot Sizing

Stream: Production and Inventory SystemsInvited sessionChair: Silvanus Enns

1 - Solution approaches for the soft drink integratedproduction lot sizing and scheduling problemDeisemara Ferreira, Reinaldo Morabito, Socorro Rangel

In this work we present an integrated lot sizing and scheduling modelfor beverage plants with sequence-dependent setup costs and times.The model considers that the production bottleneck may alternatebetween two stages (syrup preparation and bottling) and synchro-nisation of the production between these stages is required. A re-laxation approach and relax-and-fix heuristics are proposed to solveinstances generated with real data from a Brazilian soft drink plant.

63

FA-12 IFORS 2008 - Sandton

2 - Lot-sizing Problems with Time Windows and Short-agesNabil Absi, , Safia Kedad-Sidhoum

In this work we address new lot-sizing problems with time windows,early deliveries, lost sales and backlogging. We propose differentmathematical formulations of the multi-item versions. Moreover,we derive some complexity results and a dynamic programming al-gorithm for some single-item cases. Some Computational resultswill be reported.


Finance: Emissions trading and powermarketsStream: FinanceInvited sessionChair: Max Fehr

1 - Risk Management in Power Markets: The HedgingValue of Production FlexibilityHans-Jakob Lüthi, Max Fehr

Power Markets are not fundamentally different from other markets.However, electricity trading differs from the usual commodity trad-ing since depending on the maturity of a supply contract differentmarket player enter the transactions: Intra-day traders, day-aheadtraders (physical delivery) and on the long-term scale (financial)contract traders. We will discuss the implications of this hetero-geneity for quantitative Risk Management in a business environmentof a utility. In particular we will address different quantitative ap-proaches for valuating production flexibility of power plants.

2 - Modelling the European electricity and emissionmarket — the impact of different allocation rulesAnke Esser, Dominik Möst, Massimo Genoese, Otto Rentz

The introduction of emissions trading and the design of national al-location plans have far-reaching implications on the development ofpower plant mix, power generation, CO2-emissions and power plantinvestments. The presented PERSEUS-NAP model is a technology-based modeling approach employing lin-ear programming algo-rithms, where the real energy supply system in Europe is detailedrepre-sented. Results show that emission allocation schemes havea significant influence on power plant investments, the associatedevolution of CO2-emissions and on electricity prices.

3 - Design of Cap and Trade Schemes for EmissionMarketsMax Fehr, Hans-Jakob Lüthi

Recent price development of carbon allowances in the EU ETS andit’s impact on European electricity prices exhibits the importanceof a clear understanding of such Trading Systems. We propose astochastic equilibrium model for the price formation of allowancesand products, whose production causes pollution. It turns out thatfor any cap and trade scheme, designed in the spirit of the EU ETS,the consumers’ burden exceeds by far the overall reduction costs,giving rise for huge windfall profits. Following this insight we showhow to adapt regulations to reduce windfall profits.


Just-in-Time Scheduling

Stream: SchedulingInvited sessionChair: Rui Alves

1 - Heuristic algorithms for minimizing earliness-tardiness penalties in one-machine schedulingaround a common due dateRamon Alvarez-Valdes, Enric Crespo, Jose Tamarit,

In Just-In-Time scheduling not only tardiness but also earliness arepenalized. In this paper we consider the one machine schedulingproblem in which the objective is minimizing the total weighted ear-liness and tardiness around a common due date. We first proposea new constructive algorithm and then a set of moves which can beused to improve a given solution. These moves are studied sepa-rately to assess their relative efficiency for unrestrictive and restric-tive problems and are then combined in GRASP and VNS schemes.The computational study shows the efficiency of the proposed algo-rithms.

2 - Hybrid Genetic Algorithms for the Single MachineScheduling Problem with Quadratic Earliness andTardiness CostsRui Alves, Jorge Valente, M. Rosario Moreira

We consider the single machine scheduling problem with quadraticearliness and tardiness costs. A genetic algorithm based on randomkeys is proposed. Several versions of the proposed genetic approachare presented. These versions differ on the use of local search andon the generation of the initial population and some subsequent so-lutions. The genetic procedures are tested on a set of randomly gen-erated instances. The results given by the genetic algorithms arecompared with those provided by existing heuristic procedures, aswell as with optimal solutions for smaller size instances.

� FA-14Friday, 8:00 AM - 9:30 AM - Sandton Sun: Maroela North

DEA: Applications in finance andbanking

Stream: DEA and Performance ManagementInvited sessionChair: Don Galagedera

1 - Technical Efficiency of Megaworld CorporationSubsidiaries Measured by a slack-based Data En-velopment AnalysisNoli Hernandez, Babeth Isa Fernando

The research measures technical efficiency scores of 7 SubsidiaryCompanies of Megaworld Realty Corporation for years 2005-2006.Three input-one output slack-based model DEA is employed to cal-culate overall performance. Findings show that around forty percentof the companies used inputs efficiently while fifty seven percent gotthe desired output level.

2 - Quantitative Modeling Extreme Financial RiskZoran Ciric,

Quantitative modeling for a special class of applications for the mea-surement of more extreme risk levels is necessary. This also re-quires the use of advanced techniques, usually extreme value theory(EVT). Quantative risk modeling technology has evolved at a rapidpace. Risk models need to take into account the three stylized factsregarding financial returns: fat tails, volatility clusters, and nonlin-ear dependence. The actual implementation of EVT is relativelystraightforward, and it delivers good probability-quantile estimateswhere the EVT theory holds.

3 - National and Foreign Bank Branches After Nafta inMexicoFrancisco Vargas, Luis Javier Preciado, ,

The aim of the paper is to compare the performance of national andforeign bank branches in Mexico after Nafta has been approved. Thebranches are consolidated by state regions. The Malmquist Index isobtained and compared under the null hypothesis of no significa-tive differences of productivity between national and foreign bankbranches.

64


4 - Relative performance of equity markets: an as-sessment in the conventional and downside frame-worksDon Galagedera, Carla Bainbridge

This paper appraises the performance of equity markets in the cross-section using DEA. Generally, developed markets are closer to thefrontier of their best performance than in the case of emerging mar-kets. Efficient emerging markets are stable across both frameworks.More developed markets are deemed efficient under the conventionalframework than in the downside framework. Total risk tends toincrease overall performance of equity markets and downside co-skewness may not be a contributory factor of performance. Inclusionof co-skewness increases the number of efficient developed markets.

� FA-16Friday, 8:00 AM - 9:30 AM - Sandton Sun: Jacaranda

Urban transportation: planning andmodeling

Stream: Urban TransportationContributed sessionChair: Hong K. Lo

1 - Making compatible two distinct zoning systems intransportation network modelsLídia Montero, Esteve Codina

Transportation planning network equilibrium models require de-tailed information from origin zone to destination zone trip flows.Changing the zone definitions brings the problem of rendering use-less trip tables estimated at previous studies. This paper presentssome transportation network equilibrium models which combinedistinct zoning systems. The structure of the problems is that ofa multicommodity network very similar to the classical combinedtrip distribution assignment models. An algorithm based on the par-tial linearization method is developed and computational results areshown.

2 - Urban Traffic Scenario and Transportation Plan-ning Process for BangaloreGuruprasad Nagaraj, Kumaraswamy Y S

Transportation occupies a high place in modern life. Transport plan-ning is a science that seeks to study the problems that arise in provid-ing transportation facilities in an urban, regional or national settingand to prepare a systematic basis for planning such facilities. Banga-lore has been acclaimed as one of the fastest growing cities in Asiancontinent. In our paper Urban transportation issues are presentedlooking into present traffic scenario in Bangalore with particular em-phasis on traffic congestion and congestion alleviation strategies.

3 - Spatial interaction in South Africa: A Quantitativeanalysis of travel dataHannelie Nel

ational census data contain information on place of stay and placeof work. It is possible to combine this information and create flows,using the balance-of-flow rules. The process of establishing theseflows will be presented. Interesting applications, such as the de-marcation of regions will be discussed and demonstrated using GISmaps.

4 - Reliability based trip scheduling under uncertaintravel timeHong K. Lo,

This paper formulates the optimal departure schedule withcommuter-specific travel time reliability requirement. Traffic con-gestion is both time- and flow-dependent while travel time uncer-tainty arises from random network capacity degradation. Com-muters seek to minimize their travel costs associated with travel timeand schedule delay penalty. Interactions of these cost-minimizingindividuals are captured by a bottleneck model under equilibrium.

� FA-17Friday, 8:00 AM - 9:30 AM - Sandton Sun: Oleander

Energy: Policy and Planning

Stream: OR Applications in the Energy SectorInvited sessionChair: Bernard Lamond

1 - Scenario Analysis in the Energy Sector withBayesian Causal MapsDidem Cinar, Gulgun Kayakutlu

This paper provides a general overview of the current and plannedenergy policy of Turkey by using Bayesian Causal Maps (BCM). ABCM is both a causal map and a Bayesian network which is basedon probability theory. Since Turkey is an energy importer country,producing policies considering domestic resources under differentscenarios is inevitable for sustainability. BCM is a useful tool to an-alyze the complex structures, so it is possible to see the structure ofthe energy in Turkey and to understand the basic consequences ofany strategic change that may occur in the system.

2 - An agent-based Business Simulation for the longterm analysis of electricity marketsDaniel Ziegler, Christoph Weber

We present an agent-based model for the long term analysis of in-terrelated wholesale electricity markets. A future market translatesexpectations and spot market outputs into noisy prize signals by ap-plying approaches from Behavoural Finance with Traders followingdifferent strategies with hedging and speculative purposes. Thesesignals form the input for capacity planning. The study focuses onexamining the intensity and sensitivity of investment cycles and im-balances in a setup with distorted investment signals and externalshocks like changing perception of uncertain future influences.

3 - The Cofiring Problem of a Power Plant under PolicyRegulationsHanna-Liisa Kangas, Jussi Lintunen, Jussi Uusivuori

Cofiring of fossil and renewable fuels can contribute to reachingtightening climate and renewable energy goals. First, we present andsolve an electricity producer’s profit-maximization problem with de-tailed technical description of cofiring. We then study the effective-ness of policy instruments (e.g. feed-in laws) on biomass utilizationin cofiring. The study offers a novel approach to explore the cofiringproblem.

4 - Some monotonicity results for stochastic hy-dropower optimization models with concave re-wardsBernard Lamond

We will review and discuss a number of interesting monotonocityresults that apply to stochastic optimization models with concaverewards. We will examine various situations such as single versusmultiple reservoir models, neglecting or taking into account turbinehead variation, same or different routings for spilled and turbinedflows, and time sequence of decisions, inflows, turbined flows andspillages. Finally, we will briefly look at some computational ap-proaches to exploit such properties.

� FA-18Friday, 8:00 AM - 9:30 AM - Sandton Sun: Syringa


Stream: Cutting and PackingInvited sessionChair: José Fernando Gonçalves

65

FB-01 IFORS 2008 - Sandton

1 - Statistical Matching as a Packing ProblemUlrich Bartling

Usually, databases to be matched contain similar units. In our case,records in the recipient database can be looked at as aggregationsof records from the donor database. We solved the resulting pack-ing problem with up to 500,000 loosely coupled knapsack problemsusing a Breeder Genetic Algorithm.

2 - A survey on floorplan representations and algo-rithms in ASIC designA. Miguel Gomes, Marisa Oliveira, Ma Eduarda PintoFerreira

The design of an ASIC is the process of completely designing anintegrated circuit to perform a specific function. Floorplan designis an important step in the physical design of ASIC circuits. It isthe problem of placing a set of flexible blocks inside a chip, withoutoverlapping and minimizing the chip size. Additionally, blocks withinterconnections between them should be kept closer and areas withinterconnections congestions should be avoided. This is a hard com-binatorial problem with multiple objectives. We will present surveyabout floorplanning representations and algorithms.

3 - Using graph path search algorithms to optimize ap-parel manufacturer cutting order programsJoão Neiva de Figueiredo, , Sérgio Fernando Mayerle,Rafael Machado Casali, Hugo Nascimento

A methodology for optimal cutting order programs in apparel man-ufacturing with the objective of meeting demand for all SKUs atminimum cost is presented. The difficult multi-variable combina-torial problem of optimal stretching and cutting is solved using astate-space approach and path search algorithms in a directed graph.

4 - A Random Key Genetic Algorithm for a Facility Lay-out ProblemJosé Fernando Gonçalves

In this paper a hybrid genetic algorithm for the unequal area facilitylayout problem is presented. The problem consists on a collectionof several small rectangles to be packed into a large rectangular re-gion. The objective is to determine a packing that minimizes thesum of the weighted distances between the centroids of all pairs ofrectangles. A random keys based genetic algorithm and a heuris-tic are used to determine the position of each rectangle. To validatethe effectiveness of the proposed approach the algorithm is tested onbenchmark problems and compared with existing procedures.

Friday, 10:00 AM - 11:30 AM

� FB-01Friday, 10:00 AM - 11:30 AM - Convention Centre: Ballroom 1

Markov Models and Brownian MotionStream: Stochastic ModelsContributed sessionChair:

1 - Fractal geometry of the zerosets of a Brownian mo-tion.Willem Fouche

We present some recent results on the zerosets of a generic Brownianmotions. We indicate how the mincut-maxflow theorem for count-able trees can be used in this context. We conclude with some openproblems.

2 - Local times of Brownian motionS Mukeru

We shall discuss Levy’s original construction of local times of Brow-nian motion. Next we will discuss how this approach leads to an un-derstanding of the fractal geometry of zero sets of Brownian motion.

3 - Markov processes on the unity disk of a plane

Markov processes are useful tools for modelling time-dependentrandom phenomena. The future development of a Markov chain de-pends on its present state and not on the sequence of previous states.It is usually rather complicated to check wether a particular stochas-tic process has the Markov property. In this paper we present a wayfor constructing Markov processes on the unity disk of the plane.To this end, we highlight and explore interplays between Markovprocesses, semigroups and partial differential equations.


Optimization and Local Search

Stream: OptimizationContributed sessionChair: Kenneth Sörensen

1 - A New Heuristic for the Capacitated Network De-sign ProblemInmaculada Rodríguez Martín, Juan José Salazar González

This paper presents a new heuristic for the fixed charge capacitatedmulticommodity network design problem. The proposed approachis a hybrid method that combines a linear programming relaxationand a local search strategy. The heuristic turns out to be very ef-fective and the results of the computational experiments show thatnear-optimal solutions for benchmark instances of large size can befound within a reasonable time.

2 - A tabu search heuristic for concave minimum costflow network problems, Marta S. R. Monteiro, Fernando A. C. C. Fontes

In this work we address the minimum cost network flow problem,where arc costs are concave and include a fixed cost that is incurredwhenever an arc is used.

For this problem we propose a tabu search heuristic embedding a lin-ear programming local search procedure that iteratively solves linearnetwork flow problems. The marginal cost of the linear problems isupdated by having into consideration the fixed costs of the previoussolution.

Research supported by FCT/FEDER/POCI-2010: ProjectPOCI/EGE/61823/2004.

66

IFORS 2008 - Sandton FB-08

3 - Neighborhoods and local search methods for sub-set selection problemsDag HauglandMany NP-hard combinatorial optimization problems can be formu-lated as the problem of selecting a subset of elements such that afunction of the subset is minimized. In this work, we study a richclass of neighborhoods, and thereby local search methods, for subsetselection problems. The class spans the range from greedy construc-tion heuristics to improvement heuristics like exchange by 1.


Military Metrics

Stream: Military Applications of ORContributed sessionChair: Camilla Andersson

1 - Military Customer Satisfaction Index for War Sup-pliesJin Sook Ahn, So Young SohnInterest is increasing in the field of quality on user’s performance andskill for war supplies in the military. The objective of this study is tosuggest the model for military customer satisfaction index that con-siders characteristics of war supplies. We use a structural equationmodel (SEM) to compare various factors that influence user satis-faction in war supplies. The proposed model can provide feedbackinformation to effectively improve user satisfaction for war suppliesin the military.

2 - Metrics for Evaluating Military StrategyGerhard N. EngelbrechtBy using a realist perspective of military strategy a hierarchy depict-ing a military strategy is proposed that forms the framework againstwhich a military’s effectiveness to conduct its strategy can be mea-sured.

From this, metrics are proposed that pronounces on a military’s abil-ity to execute its military strategy in terms of the probability that themilitary would be effective to conduct war.

3 - Health risks and safety of military divers exposedto transmitting sonar

The health risks and safety of military divers exposed to transmittingsonar, was investigated in a literature study. This complex probleminvolves underwater sound propagation, ocean acoustic noise, in-water hearing in humans and auditory and non-auditory bio-effectsof underwater sound on humans. Safe diving distances and exposurelimits are suggested.

4 - The use of Decision Analysis tools to enhance as-sessmentCamilla AnderssonThis case-study demonstrates the use of Decision Analysis toolsto support assessment and evaluation of military units. The casedemonstrates the possibilities of using Analytical Hierarchy Processto facilitate criteria weighting. Furthermore, it will show how thisenhances the accuracy in the assessment of a unit’s performance andcapability.

� FB-06Friday, 10:00 AM - 11:30 AM - Convention Centre: Boardroom2

E-CommerceStream: E-commerceContributed sessionChair:

1 - Optimization methods against spamCatherine Roucairol, didier colin, Ider Tseveendorj

Spam or unsollicited bulk email has become increasingly presentin the internet mailing traffic over the last decade. As a conse-quence, there has been a growing need to find solutions in orderto counter this seemingly inescapable phenomenon. Firstly, we givean overview of the spam filtering area, by presenting models whichhave been designed to filter spam at the user’s mailbox level. Then,we investigate how optimization methods can be used to tune andimprove the existing classification models for spam filtering andpresent some examples.

2 - Optimal allocation of overlapping inventory in on-line advertising, Deepak Agarwal

On-line advertisers typically wish to target their ads to users satis-fying a particular profile - age range, sex, interests, etc. Publisherstypically forecast the number of users in pools satisfying some ofthese requirements. We describe a model for optimally satisfyingthe advertiser requirements subject to the availabilities.


Fuzzy Decision Analysis and PreferenceStructuresStream: Fuzzy Sets and SystemsContributed sessionChair: Chung-Tsen Tsao

1 - Fuzzy Multi-Criteria Decision Making Problem ForSupplier SelectionGöksu Kaya

The objective of supplier selection decision is to find suppliers withbest fit to the firm’s needs. The aim of this paper is to use fuzzy ana-lytic hierarchy process (FAHP) to compare supplier firms for one ofthe computer manufacturer, established a new facility. FAHP tacklesefficiently both quantitative and qualitative decision factors involvedin selection of supplier firm. The triangular fuzzy numbers and thepairwise comparison matrices are used to transform linguistic com-parison of decision criteria and performance of the alternative sup-pliers. Finally, the best supplier was selected.

2 - The Portfolio Selection in Fuzzy DimensionsChung-Tsen Tsao

The statistical measures — expectation and variance, had been de-fined as the indicators of return and risk. They describe the random-ness of outcomes, yet do not manage the imprecision of data estima-tion. Fuzzy sets are here suggested as a complement. Fuzzy sets andstatistics cohere together to do the selection of efficient portfolios,considering both the randomness and vagueness of estimation. Thiswork mends some previous deficiencies by substituting the requisite-equality-constraint operation for the standard fuzzy arithmetic, anddevelops a set of algorithms for the practical use.


Decision Support Systems 2

Stream: Decision Support SystemsContributed sessionChair:

67


1 - Model development under uncertainty of the rela-tionships between waste composition and struc-tural size based on the Arcelor Mittal Poland SA,in Krakow case studyBoguslaw Bieda

The paper objective is to develop the industrial waste prognosticmodel based on the database time-series waste quantities and com-positions over the 20 years period under uncertainty using MonteCarlo (MC) simulation. In this paper is used statistical analysis (e.g.analysis of variance ANOVA, cluster analysis, box plots, etc.) toexamine relationships between waste stream composition and struc-tural size of the steel plant. Data analysis based on the Arcelor MittalPoland SA, in Krakow has been conducted using Statistica software.Uncertainty analysis is inherently a statistical process.

2 - A hybrid system combining optimization and simu-lation for improved sales personnel assignment ata public railway transportation company, , Pablo A. Rey

Metro S.A. in Santiago had the problem of determining the optimalnumber of sales personnel for their points-of-sales that assured ac-ceptable queues of passengers yet maintained low costs. A linearprogramming model was used to determine a first rough plan. Thena simulation model refined this plan using distributions of demandand service times estimated for each interval of 15 minutes. Thesystem is in use at Metro in Santiago with very positive results.


Multi-objective decision-making

Stream: Decision AnalysisContributed sessionChair: Theodor Stewart

1 - Directing an Interactive Nonlinear MultiobjectiveOptimization Procedure by Using a Trade-off Anal-ysis ToolJuha Eskelinen, Kaisa Miettinen

In interactive methods, a decision maker (DM) directs the search forthe most preferred Pareto optimal solution with his/her preferences.We propose a tool that can be used to support the DM. With thistool, the DM can conveniently learn about local tradeoffs and judgewhether they are worthwhile.

2 - A New Approach for Solving Fuzzy MCDM Prob-lemsJaroslav Ramik, Radomir Perzina

A new decision model for solving the decision making problem withfuzzy pair-wise comparisons and a feedback between the criteria isproposed. The evaluation of the weights of criteria, the variants aswell as the feedback between the criteria is based on the data givenin fuzzy pair-wise comparison matrices. An illustrating numericalexample is presented to clarify the methodology.

3 - Branch and Bound Algorithm for Tri-Objective As-signment ProblemMelih Ozlen

We consider the generation of all efficient solutions for the Tri-Objective Assignment Problem (TOAP), and propose a tri-objectivebranch and bound algorithm. We use bi-objective supported solu-tions as an initial set and populate the set using neighborhood search.As a lower bound we use single objective assignment problem solu-tions.


Modelling and Prediction of Genes,Proteins and Their StructureStream: Computational Biology and InformaticsInvited sessionChair: Gerhard-Wilhelm Weber

1 - Protein Folding Type Prediction via MILP, Metin Turkay

In this study, a three-stage mixed-integerlinear programming(MILP) based hyper-box enclosure method is presented for proteinfolding type prediction. The performance of this method is tested onsix distinct protein folding type benchmark data sets that are com-posed of 120, 138, 253, 359, 252 and 1601 protein domains, respec-tively. Leave-one-out (jack-knife) test results shows that proposedapproach is efficient.

2 - Signal Processing in Genome-Wide AssociationStudiesValentina MoskvinaWe apply the methodology of time series analysis and signal pro-cessing to genome-wide association studies taking into account thesignificance of several subsequent polymorphisms in contrast to thewidely accepted way of looking at the individual p-values. Wepresent a theoretical approach to calculating the probability of atleast one false alarm flagged by the detection statistic in a particularinterval (possibly the whole genome) under the Null Hypothesis ofno signal (association).

3 - Gene-Environment Networks in Environmental andLife SciencesErik Kropat, Gerhard-Wilhelm WeberGene-environment networks provide a conceptual framework formany phenomena in areas of medicine, environmental protection,and development under various kinds of uncertainty. We survey re-cent advances in the understanding of the mathematical foundationsand interdisciplinary implications in this field and indicate the rela-tions to applications in finance and CO2-emissions control.

� FB-11Friday, 10:00 AM - 11:30 AM - Convention Centre: CommitteeRoom 2

Combinatorial optimization: models andalgorithms

Stream: Combinatorial OptimizationContributed sessionChair:

1 - Profit-based Latency ProblemsFrits Spieksma, Sofie CoeneWe consider a latency problem with a given profit for each client.When serving a client, a certain revenue is collected. The goal is tofind routes for the servers such that total collected revenue is max-imized. We study algorithms for different variants of this problem;in particular, we consider the case where the clients are located on aline.

2 - Sequential testing under precedence constraintsTonguc Ünlüyurt, Bülent ÇatayIn this study, we consider the problem of testing a series system withthe minimum expected cost where there are precedence constraintsfor applying the tests. The tests are assumed to provide perfect infor-mation. The costs of the tests and the probobabilities regarding theresults of the tests are known. A feasible solution for the problem isa permutation of all tests obeying the precedence constraints. We re-port the performance of some heuristic algorithms that we propose.We also discuss the relation with k out of n systems.

68

IFORS 2008 - Sandton FB-14

3 - Uniformly suboptimal solutions for combinatorialoptimization problems with interval data uncer-taintyIgor Averbakh

In combinatorial optimization with interval data, it is assumed thatcoefficients defining the objective function are not known and cantake on any values in some pre-specified uncertainty intervals. Thistalk will provide a selective survey of some recent developmentson the complexity of finding uniformly suboptimal solutions (alsocalled minmax regret solutions) for such problems.


Finance: Risk management incommodity markets

Stream: FinanceInvited sessionChair:

1 - Integrated risk management in renewable conven-tional energy marketsSilvana Stefani, Paolo Falbo, Daniele Felletti

We face the problem of an energy producer willing to optimize a mixof conventional and renewable power capacities. We analyze howthe planning decisions are affected by relevant market risk sources,i.e. input and output price fluctuations, and the producer’s risk poli-cies. The producer will manage his total risk through forward con-tracts and the renewable capacity. We apply our results to the caseof two energy producers, one operating in Germany and the other inSouth Italy.

2 - The intervention of a specialist in thin stock pricesRoy Cerqueti, Rosella Castellano

The presence of a relevant number of thin stocks in financial marketscaused the introduction of market mechanisms devoted to acceleratethe convergence process of prices to a long-run equilibrium and toprovide liquidity to the market: the Specialist System. In this workwe analyze the structure of the Specialist intervention, and we de-rive the optimal bid/ask spread in an analytically closed form. Atthis aim, a dynamic stochastic optimization model is constructed andsolved, and the model will be empirically validated.

3 - Forward contract to manage market power

In electricity markets the use of derivative contracts became crucialto guarantee a competitive behavior of the spot market. Given thenon-storability of the electricity the familiar arbitrage-based meth-ods are not applicable for pricing derivative contracts. We examinean equilibrium forward contract on a nonstorable commodity whenforward market participants have market powers. We consider theintroduction of a forward market as an effective tool to increase boththe production of the commodity and the trading volume and reducemarket power.


Scheduling Independent tasks

Stream: SchedulingInvited sessionChair: Paolo Detti

1 - Algorithms for scheduling access to a shared re-sourceIsabelle Nieuwoudt, Jan van Vuuren

A maximum-degree colouring of a graph is an assignment of coloursto the vertices of the graph, one colour to a vertex, so that no colourclass induced maximum degree is more than some specified number,d. Maximum-degree colourings arise in scheduling problems wheresome threshold d of conflict between different users of a shared re-source is tolerable. Four algorithms (two heuristic and two exact)for finding maximum-degree colourings using the smallest numberof colours are presented and compared in this talk.

2 - A Scheduling Genetic AlgorithmFernando A. C. C. Fontes, Guilherme Pereira,

We propose a genetic algorithm to schedule a set of tasks, each ofwhich having several differentiating characteristics that must be ac-counted for. From the set of available tasks we must decide whichare to be performed and when. Furthermore, tasks may be sched-uled more than once and some tasks, although different may be usedto achieve the same output. The proposed scheduling heuristic ispresented in the context of a real application case-study for a majorPortuguese television station.

Research supported by FCT/FEDER/POCI2010: ProjectPOCTI/MAT/61842/2004.

3 - Sequencing unreliable jobs on parallel machinesPaolo Detti, Alessandro Agnetis, Marco Pranzo, ManbirSodhi

An allocation and sequencing problem is considered in which thereare independent jobs to be processed by a set of machines. Each jobis characterized by a certain success probability, and a reward whichis obtained if the job is successfully carried out. When a job failsduring processing, the machine is blocked and the jobs subsequentlyscheduled on that machine are blocked until the end of the unsuper-vised period. The problem is to assign and sequence the jobs on themachines so that the expected total reward is maximized. Complex-ity and algorithmic results are presented.

� FB-14Friday, 10:00 AM - 11:30 AM - Sandton Sun: Maroela North

Problem Structuring and Solving

Stream: Problem Structuring and SolvingContributed sessionChair: Roelof Coetzer

1 - Coherent Pluralism in Managing Problem Situa-tions

Pluralism in management science can be seen as a response to themultitude of methodologies, methods, techniques that can be used intackling complex problems situations. Coherent pluralism must per-mit an analysis of problem situations from multiple perspectives andfacilitate use of different methodologies in combination. Coherentpluralism: a) must encourage a flexibility in use of the widest vari-ety of methods, techniques; b) should encourage the use, together, ofdifferent methodologies based upon alternative paradigms; c) mustaccept and manage a degree of paradigm incompatibility.

2 - Lesson from Problem Structuring Methods and itsApplicationArabinda Tripathy

Various Problem Structuring methods are being used under differentproblem situations. A large number of applications of these meth-ods are also being reported. There is substantial commonality in theapproach of all these methods, though the method themselves aredifferent. An effort has been made to identify these similarities andinitiatives taken to develop appropriate methodology for each indi-vidual problem situation. Some of these applications are discussed.The areas relate to situation in parliament to technology adoption inindustry to establishment of new academic institution.

69


3 - Design and Analysis of Computer Experiments:Generating Maximum Information for the ProcessIndustryRoelof Coetzer, Diki Langley

Computer models are used extensively to predict the performanceof real life complex engineering processes. However, these com-puter codes may result in high computational costs. The method-ology of computer experiments was developed for the sampling ofthe computer code and constructing statistical meta-models of theinput-output relationships. The approximation models are then usedfor facilitating the exploration of the design space, optimization andsensitivity analysis. The theory and the application of this method-ology to mathematical programming will be discussed.

� FB-15Friday, 10:00 AM - 11:30 AM - Sandton Sun: Maroela South

Economic Modeling

Stream: Economic ModelingContributed sessionChair: Kiyoshi Yoneda

1 - On the Use of Black-Box Models for Refinery Opti-misationAninda Chakraborty

Black-boxing parts of the entire refinery become necessary whencertain business units are not allowed to view their inner workings.This may happen due to contractual obligations. This presentationwill discuss the overall philosophy of ’black-boxing’ and comparethe results of the ’black-boxed’ models with full refinery models.

2 - Dynamic prediction from elasticities in open pop-ulations: its application to the forecast of taxchargesTeofilo Valdes, Carmen Anido

We focus on the demographical and non-demographical patternsof change in open populations and their forecasting potentialities.These patterns are elasticities, defined in terms of ratios betweentwo differential variation rates: individual and global, and they al-low us to predict the individual changes from the known global ones.This situation usually occurs in a great number of tax studies. Forinstance, the prediction of future global charges of the income taxassuming that the population of tax payers is not closed but subjectto incomings and outgoings throughout time.

3 - Modeling Individual Behavior as a Box-Bounded In-verse ProblemKiyoshi Yoneda

Behavior is modeled as an inverse problem: as individual tries toadjust her operating variables to attain her objectives as closely aspossible in a stable way. The modeling consits of writing down(linear) equations. Each variable is required to lie within an inter-val deemed acceptable. Such box-bounded inverse problems maybe solved by minimizing a strictly convex loss function related tothe modern matching theory in psychology, which is easier than theconventional utility maximization under budget constraints.

� FB-16Friday, 10:00 AM - 11:30 AM - Sandton Sun: Jacaranda

Urban Transit Planning and Operations

Stream: Urban TransportationContributed sessionChair:Chair:

1 - Public transportation network design model for theCity of Tshwane Metropolitan MunicipalityManuel Fletterman, Johan Joubert

This paper presents the results of a model developed to redesign amulti-modal public transport system. The model is significant in thatit includes the optimization of bus stop placement; considers multi-ple vehicle types for route network design; and incorporates a GISinterface.

2 - What Express Services to Offer on a High DemandTransit Corridor?Juan Carlos Muñoz, Ricardo Giesen, Carola Leiva

We study the design of transit services on a high-demand corridor.We present a mathematical programming method for determiningthe best express services to provide in this corridor. We determinethe set of routes and their frequencies that would minimize the ex-pected total costs for a given origin-destination demand matrix.

3 - Robust Optimal Traffic Signal Timing, Lihui Zhang

Traditional timing approaches for pre-timed control systems do notproactively consider traffic demand uncertainty, and thus the result-ing control performance is often unstable under fluctuating trafficconditions. We propose a methodology to determine signal timingssuch that their performance will be fairly stable under any realiza-tion of uncertain traffic flows at arterials. Compared with those fromconventional timing approaches, robust timing plans are expectedto perform better for oversaturated high-demand scenarios withoutcompromising much the average performance.

� FB-17Friday, 10:00 AM - 11:30 AM - Sandton Sun: Oleander

Energy: Optimization

Stream: OR Applications in the Energy SectorInvited sessionChair: Olusola Okesola

1 - Optimisation of coal supply to power stations inSouth Africa using MILPPaul van Nierop, Alwyn van der Merwe,

A mixed integer linear program (MILP) model has been developedusing the Aimms modeling environment to support the developmentof a long term (20-50 year) coal supply plan for a South African En-ergy Utility Company. The model optimises the supply and transportof coal to power stations by conveyor, rail or road. The model meetsplanned energy production at lowest overall NPV cost. Constraintsinclude supply, transport, coal quality, investment lead times, con-tracts and mine reserve constraints. Integer variables are used tomodel investment decisions for mines and transport infrastructure.

2 - Optimization of Energy Systems with Multi-objective Optimization MethodsMetin Turkay,

Optimization of energy systems includes conflicting objectives andit is not generally possible to find a unique optimum solution that si-multaneously optimizes all objectives. Minimization of productioncost and of harmful compound emissions are examples of the multi-objective nature of energy systems. We present an algorithm forsolving mixed-integer multi-objective optimization problems. Themulti-objective optimization method is designed to generate all effi-cient solutions. We illustrate the algorithm on examples from energysupply chain systems.

70

IFORS 2008 - Sandton FC-01

3 - Application of Multi-Criteria Dynamic Linear Pro-gramming Model to Supply Chain Management ofCrude Oil in NigeriaOlusola Okesola

This study of supply chain management of Nigerian NationalPetroleum Corporation (NNPC) focuses on:

* Cost of discovery of oil and allocation of oil wells, * Cost of ac-quiring equipments for exploration, * Cost of production processes(e.g. refining), * Cost of distribution/transportation, * Cost of fore-casting the short- term, medium term, and long-term market de-mand.

This paper presents a multi-criteria decision analysis as a means ofincreasing the performance of NNPC through minimization of thecost elements stated above.

� FB-18Friday, 10:00 AM - 11:30 AM - Sandton Sun: Syringa

Optimization and Metaheuristics

Stream: OptimizationContributed sessionChair: Aaron Luntala Nsakanda

1 - A hybrid ant colony optimization algorithm appliedto the production lines design problemHicham Chehade, Farouk Yalaoui, Lionel Amodeo,

In this paper, we are considering the production lines design prob-lem consisting of determining the machines to be used in stationsand the capacities of the intermediate buffers. We present a hybridant colony optimization approach coupled with a guided local searchand we compare it to another heuristic method. The objective is tominimize the total cost of the configuration while achieving a desiredthroughput rate.

2 - Improving the Performance of Simulated Annealingfor the Travelling Salesman Problem via AdaptiveCooling ScheduleAderemi Adewumi, Montaz Ali, Ebunoluwa Fasina, OjoAyeni

The application of simulated annealing (SA) metaheuristics to theNP-hard travelling salesman problem (TSP) is studied. The choiceof cooling schedule is central to the performance of SA algorithm.A modified SA with an adaptive cooling schedule that improves theperformance of the algorithm is proposed. The performance of thetraditional SA and the modified algorithm are tested using somebenchmark data for TSP from the OR library. Simulation result ofthe experiment is presented.

3 - Population diversity computation methods in ge-netic algorithms with application to the cell forma-tion problemAaron Luntala Nsakanda

Population diversity is crucial to the genetic algorithm’s ability tocontinue fruitful exploration. It may be used in choosing an ini-tial population, in defining a stopping criterion, and in making thesearch more efficient throughout the selection of crossover opera-tors or the adjustment of various control parameters. We discusstwo measures to compute the population diversity in solving the cellformation with the genetic algorithm and investigate to what extendthese measures lead to better and faster solutions. Numerical resultsfrom an extensive computational experience will be reported.

Friday, 12:00 PM - 1:00 PM

� FC-01Friday, 12:00 PM - 1:00 PM - Convention Centre: Ballroom 1

Bringing Computers to the World

Stream: PlenariesPlenary sessionChair:

1 - The Simputer Meme: A Retrospective

This is an inside and personal account of the Simputer project pre-sented by one of the founders of the project. The Simputer, is a novelopen source hand held computer developed by a group of technolo-gists in Bangalore, India in the late 1990s. The SIMPUTER (Sim-ple Inexpensive Multilingual People’s compUTER) attracted a greatdeal of attention in the public media and open source communities.The World Resources Institute saw the Simputer as a symbol of tech-nology with "radical simplicity for universal access’ to bridge thedigital divide. The Simputer quickly became a meme* — a persis-tent idea that would represent technology innovation that addressedthe digital divide. The images of children in a school in Bastar usingSimputers as part of their school curriculum, of a village accoun-tant recording and checking agricultural records on Simputers, ofa micro-finance transaction using the smart card interface of Sim-puters were featured on BBC, New York Times, Wired, TechnologyReview and even Business Week. Calling the Simputer the most sig-nificant innovation in computer technology in 2001, Bruce Sterlingwrote in the New York Times that "This is computing as it wouldhave looked if Gandhi had invented it.’ It has been exactly ten yearssince the group of technologists in Bangalore conceived the Sim-puter and time for a serious retrospective and SWOT analysis.

* A meme as defined by memetic theory, constitutes a theoreticalunit of cultural information, the building block of culture or cul-tural evolution which spreads through diffusion propagating fromone mind to another analogously to the way in which a gene propa-gates from one organism to another as a unit of genetic informationand of biological evolution.’ [Wikipedia]

71

STREAMS

Combinatorial OptimizationSilvano MartelloUniversity of [email protected](s): 1 11

Computational Biology andInformaticsMetin TurkayKoc [email protected]

Gerhard-Wilhelm WeberMiddle East Technical [email protected](s): 10

Container PortsEk Peng ChewFaculty of Engineering, NationalUniversity of [email protected](s): 11 16

Cutting and PackingJosé Fernando OliveiraUniversity of [email protected](s): 18

DEA and PerformanceManagementAli EmrouznejadAston [email protected](s): 14 15

Decision AnalysisAhti SaloAalto University School of Scienceand [email protected]

Theodor StewartUniversity of Cape [email protected](s): 8 9

Decision Support SystemsBoguslaw BiedaAGH-University of Science [email protected](s): 8

Economic ModelingTeofilo ValdesComplutense University of [email protected](s): 15

FinanceJ. E. BeasleyBrunel University & JB [email protected](s): 12 13

Game TheoryFouad El OuardighiESSEC Business [email protected](s): 9

Graphs and NetworksSilvano MartelloUniversity of [email protected](s): 18

Humanitarian LogisticsLuk Van [email protected](s): 7

IFORS 50th AnniversaryMichael TrickCarnegie Mellon [email protected](s): 10

Location and LayoutHorst W. HamacherUniversity of [email protected]

Horst W. HamacherUniversity of [email protected](s): 16

Management InformationSystemsLidija Zadnik StirnUniversity of [email protected](s): 8

Military Applications of ORHoong Chuin LauSingapore Management [email protected](s): 3 16

Operations ManagementCandace YanoUniversity of California, [email protected](s): 10 11 16

OptimizationGeorge NemhauserGeorgia Institute of [email protected]

Chung Piaw TeoNational University of [email protected](s): 1 2 4 18

OR and EconomicDevelopmentHans W. IttmannHWI [email protected](s): 7 15

OR and EthicsLeroy WhiteUniversity of [email protected](s): 5

OR and HealthcareStefan NickelKarlsruhe Institute of Technology(KIT)[email protected](s): 7

OR and SportsMichael TrickCarnegie Mellon [email protected](s): 4

OR and the Public SectorTatsuo OyamaNational Graduate Institute forPolicy [email protected](s): 3 15

72

IFORS 2008 - Sandton STREAMS

OR Applications in the EnergySectorHans W. IttmannHWI [email protected](s): 17

OR EducationPeter BellIvey Business [email protected](s): 5

OR for SustainableDevelopmentHans W. IttmannHWI [email protected](s): 15

OR in Agriculture and NaturalResourcesSergio MaturanaP. Universidad Catolica de [email protected](s): 4

OR/MS SoftwareRobert FourerAMPL Optimization [email protected]

Bjarni KristjanssonMaximal Software, [email protected](s): 17

PlenariesJohn BartholdiGeorgia Institute of [email protected](s): 1

Production and InventorySystemsYun Fong LimSingapore Management [email protected](s): 11 16

Quality and ReliabilityMaximo BoschUniversity of [email protected]

Jeff GriffithsCardiff [email protected](s): 19

Railroad ApplicationsPaolo TothUniversity of [email protected]

Juan José Salazar GonzálezUniversidad de La Laguna (Tenerife)[email protected](s): 3

RoutingJuan José Salazar GonzálezUniversidad de La Laguna (Tenerife)[email protected](s): 9 16

Software DemonstrationsRobert FourerAMPL Optimization [email protected]

Bjarni KristjanssonMaximal Software, [email protected](s): 17

Supply chainsCandace YanoUniversity of California, [email protected](s): 11

TelecommunicationsBernard FortzUniversité Libre de [email protected](s): 18

73

Session Chair Index, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-01, FC-01, RE-03,

TC-03, RE-04, TC-04, MA-05, MC-05, FA-06, FB-06, RC-07, FB-08, MA-08, TC-08, FA-09, ME-09,TC-09, RC-10, FB-11, 12, MC-13, RC-13, MC-14,15, TC-15, FB-16, RC-16

Akume, [email protected] Science Department, University of Buea, Buea,Cameroon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-13

Alirezaee, [email protected] of Mathematics, Iran University of Science andTechnology, Tehran, Iran, Islamic Republic Of . . . RC-14

Alves, [email protected] de Economia, Universidade do Porto, Porto,Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-13

Amar, A. [email protected] Department, Seton Hall University, South Or-ange, NJ, United States . . . . . . . . . . . . . . . . . . . . . . TA-12

Andersson, [email protected] of Defence Analysis, FOI Swedish Defence Re-search Institute, Stockholm, Sweden . . . . . . . . . . . FB-03

Andreeva, [email protected] School, University of Edinburgh, Edinburgh,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-12

Arenales, [email protected] of Applied Mathematic and Statistics, universidadede São Paulo, Sao Carlos, SP, Brazil . . . . . . . . . . . RD-18

Baker, [email protected] of Business, University of Salford, Salford, Lan-cashire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-19

BARBER, [email protected] Informaticos Y Computacion, Universidad Po-litecnica De Valencia, VALENCIA, SPAIN, Spain ME-03

Bassok, [email protected] of Information and Operations ManagementScience, University of Southern California, United StatesTA-11

Beasley, J. [email protected] Sciences, Brunel University & JB Consul-tants, Uxbridge, Middlesex, United Kingdom . . . MC-12

Bell, [email protected]

Western University, Ivey Business School, London, On-tario, Canada . . . . . . . . . . . . . . . . TA-05, TD-05, TE-05

Billaut, [email protected] d’Informatique, University of Tours, Tours,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-13

Bosch, [email protected] Engineering, University of Chile, Santiago, ChileTC-19

Briand, [email protected] National Laboratories, Albuquerque, NM, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-19

Callea, [email protected] di Ingegneria Gestionale, Università degliStudi di Bergamo, Dalmine, Italy . . . . . . . . . . . . . . RA-07

Casier, Auré[email protected]épartment d’Informatique, ULB - Université Libre deBruxelles, Brussels, Belgium . . . . . . . . . . . . . . . . . TA-13

Chen, Frank [email protected] of Management Science, City University ofHong Kong, Hong Kong, China . . . . . . . . . . . . . . . ME-16

Chew, Ek [email protected] and Systems Engineering, Faculty of Engineer-ing, National University of Singapore, Singapore MA-11

Chirenje, Kelvin [email protected] Maths, National University of Science and Tech-nology, Chitungwiza, Harare, Zimbabwe . . . . . . . . RD-07

Chou, [email protected] Sciences, National University of Singapore, Sin-gapore, Singapore . . . . . . . . . . . . . . . . . . . . . . . . . . TC-11

Chu, [email protected] of Management and Information Technol-ogy, Southern Taiwan University, Yungkang City, TainanCounty, Taiwan . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-07

Cochran, [email protected] of Marketing and Analysis, Louisiana TechUniversity, Ruston, LA, United States . . . . . . . . . . TA-04

Coetzer, [email protected] and Development, Sasol Technology (Pty) Ltd.,Sasolburg, Free State . . . . . . . . . . . . . . . . . . . . . . . . FB-14

Colebrook, [email protected]

74

IFORS 2008 - Sandton SESSION CHAIR INDEX

Depto. de Estadistica, Investigacion Operativa y Computa-cion, Universidad de La Laguna, La Laguna, Santa Cruz deTenerife, Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-09

Cosma, [email protected], Université du Luxembourg, Luxembourg,Luxembourg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-12

Crook, [email protected] of Edinburgh Business School, University ofEdinburgh, Edinburgh, Lothian, United Kingdom . TD-13

Daouia, [email protected] School of Economics, University of Toulouse,Toulouse, France . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-14

de Werra, [email protected], EPFL, Lausanne, Switzerland . . . . . . . . . . . MA-01

Despontin, [email protected], Vrije Universiteit Brussel, Brussels, AfghanistanRD-15

Detti, [email protected] Engineering Department, University of Siena,Siena, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-13

Ding, [email protected] of Business, Singapore Management University,Singapore, Singapore . . . . . . . . . . . . . . . . . . . . . . . . TC-16

Dowart, [email protected] of Applied Mathematics, National Universityof Science and Technology (NUST), Harare, ZimbabweMA-16

Eglese, [email protected] Management School, Lancaster University, Lancaster,Lancashire, United Kingdom . . . . . . . . . . . . . . . . . ME-01

El Ouardighi, [email protected] Management, ESSEC Business School, CergyPontoise, France . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-09

Eleria, [email protected] Research Center, University Of Santo Tomas,Manila, Philippines . . . . . . . . . . . . . . . . . . . . . . . . . TA-14

Emrouznejad, [email protected] Business School, Aston University, Birmingham,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-15

Enns, [email protected]

Mechanical Engineering, University of Calgary, Calgary,AB, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-11

ESOGBUE, [email protected] & Systems engineering, Georgia Institute ofTechnology, Atlanta, GA, United States . . . . . . . . . RC-02

Evans, [email protected] Risk Assurance, DBSA, Midrand, South AfricaTD-15

Fatti, L. [email protected] & Actuarial Science, University of the Witwater-srand, Johannesburg, South Africa . . . . MC-07, MD-07,ME-07

Faulin, [email protected]. Statistics and Operations Research, Public Univer-sity of Navarre, Pamplona, Navarra, Spain . . . . . . . TD-09

Fehr, [email protected] (Institute for Operations Research), ETHZurich, Zurich, Switzerland . . . . . . . . . . . . . . . . . . . FA-12

Fernandez, [email protected] and Operations Research, Technical University ofCatalonia, Barcelona, Spain . . . . . . . . . . . . . . . . . . RD-03

Fernando, Babeth [email protected] Department, University of Santo Tomas, Manila,Philippines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-15

Ferreira, [email protected] & Computer Engineering, University of Camp-inas, Campinas, São Paulo, Brazil . . . . . . . . . . . . . . FA-02

Fick, [email protected] of Decision Sciences, University of SouthAfrica, Pretoria, South Africa . . . . . . . . . . . . . . . . . TD-02

Figueira, José [email protected] Superior Tecnico, Technical University of Lisbon,Lisbon, Portugal . . . . . . . . . . . . . . . . . . . MD-08, ME-08

Fourer, [email protected] Optimization Inc., Evanston, IL, United StatesME-17, TA-17, TC-17

Friedler, [email protected] of Computer Science, University of Pannonia,Veszprem, Hungary . . . . . . . . . . . . . . . . . . RC-04, RD-04

Fu, [email protected]

75

SESSION CHAIR INDEX IFORS 2008 - Sandton

The Robert H. Smith School of Business and Institute forSystems Research, University of Maryland, College Park,MD, United States . . . . . . . . . . . . . . . . . . . . . . . . . MA-09

Galagedera, [email protected] and Business Statistics, Monash University,Caulfield East, Victoria, Australia . . . . . . . . . . . . . . FA-14

Galdeano, [email protected] Aplicada, Universidad de Almeria, Almeria,Almeria, Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-14

Gendreau, [email protected] and CIRRELT, École Polytechnique, Montreal,Quebec, Canada . . . . . . . . . . . . MC-01, MD-01, TB-01

Giesen, [email protected] Engineering and Logistics, Pontificia Universi-dad Catolica, Santiago, RM, Italy . . . . . . . . . . . . . . TD-16

Gimadi, [email protected] Optimization in Operations Research, Sobolev In-stitute of Mathematics, Novosibirsk, Russian FederationRA-16

Golany, [email protected] Engineering & Management, Technion - IsraelInstitute of Technology, Haifa, Israel . . . . TA-10, TC-10

Gomes, A. [email protected] TEC, Faculdade de Engenharia, Universidade doPorto, Porto, Portugal . . . . . . . . . . . . . . . . . . . . . . . RE-18

Gonçalves, José [email protected], INESC TEC, Faculdade de Economia do Porto,Universidade do Porto, Porto, Portugal . . . . . . . . . . FA-18

Gouveia, Luí[email protected] - Departamento de Estatística e Investigação Opera-cional, Universidade de Lisboa - Faculdade de Ciências,Lisboa, Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-18

Greben, [email protected] Environment, CSIR, Pretoria, Gauteng, South AfricaTD-03

Griffiths, [email protected], Cardiff University, CARDIFF, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-19

Hallamäki, [email protected] Information Technology, University ofJyväskylä, Jyväskylä, Finland . . . . . . . . . . . . . . . . . TE-16

Hans, [email protected] of Management and Governance, University ofTwente, Enschede, Netherlands . . . . . . . . . . . . . . . TD-07

Harmse, [email protected] Resources Development & Organisational Devel-opment, Sasol Technology, Secunda, South Africa RE-05

Hochbaum, [email protected]&OR department, UC Berkeley, Berkeley, CA, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-01

Imai, [email protected] School of Maritime Sciences, Kobe University,Kobe, Japan . . . . . . . . . . . . . . . . . . . . . . . . MC-11, TA-16

Immelman, [email protected] Sciences, University of South Africa, pretoria,Gautemg, South Africa . . . . . . . . . . . . . . . . . . . . . . RD-06

Infanger, [email protected] Science and Engineering, Stanford Univer-sity, Stanford, CA . . . . . . . . . . . . . . . . . . . . . . . . . . MA-12

Ittmann, Hans [email protected] Consulting, The Willows, South Africa . . . . MB-01

Jankowitz, [email protected] Sciences, UNISA, Pretoria, South Africa FA-01

Joubert, [email protected] and systems Engineering, University of Pretoria,PRETORIA, South Africa . . . . . . . . . . . . . . . . . . . RA-11

jurczenko, [email protected], ESCP-EAP, Paris, Ile De France . . . . . . . ME-12

Kanková, [email protected], Institute of Information Theory and Au-tomation of ASCR, Prague, Czech Republic, Czech Re-public . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-02

Kempkes, Jens [email protected] & OR Lab, University of Paderborn, Germany, Pader-born, Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-13

Kirkbride, [email protected] Management School, Lancaster University, Lancaster,Lancashire, United Kingdom . . . . . . . . . . . . . . . . . RC-19

Kis, [email protected] Methods in Economy, Faculty of Economics,

76


Subotica, Serbia . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-11

Klemes, [email protected], University of Pannonia, Veszprém, HungaryRC-04, RD-04

Kristjansson, [email protected] Software, Ltd., London, United Kingdom . MC-17, TC-17

Kruger, [email protected] of Science and Technology, Federation UniversityAustralia, Ballarat, Victoria, Australia . . . . . . . . . . RD-02

kutlar, [email protected] Faculty, Cumhuriyet university, sivas, sivas,Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-14

Lamond, [email protected] & Decision Systems, Universite Laval, Quebec,Quebec, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-17

Lange, [email protected] of Law and Economics, Technische Univer-sität Darmstadt, Darmstadt, Germany . . . . . . . . . . . RE-16

Lau, Hoong [email protected] of Information Systems, Singapore ManagementUniversity, Singapore, Singapore . . . . . . . . . . . . . MD-16

Lawton, [email protected] Sustainment & Readiness, Sandia National Labo-ratories, Albuquerque, New Mexico, United States MA-19

Lee, [email protected] and Statistics, University of South Australia,Mawson Lakes, South Australia, Australia . . . . . . MD-11

Leung, [email protected] Engineering & Engineering Management Dept.,The Chinese University of Hong Kong, Shatin, Hong KongRA-03, RC-03

Lim, Yun [email protected] Kong Chian School of Business, Singapore Manage-ment University, Singapore, Singapore . . . . . . . . . MC-16

Liu, [email protected] Business School, canterbury, United Kingdom . RE-14

Lo, Hong [email protected] Engineering, Hong Kong University of Science and

Technology, Hong Kong, Hong Kong . . . . . . . . . . . FA-16

Mac Cawley, [email protected] and System Engineering, Georgia Institute ofTechnology, Atlanta, Georgia, United States . . . . . ME-04

Mareschal, [email protected] Brussels School of Economics and Management,Université Libre de Bruxelles, Brussels, Belgium . RA-08

Marques, [email protected] of Civil Engineering and Architecture,CESUR-IST, Technical Univeristy of Lisbon, Lisbon, Por-tugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-14, RE-15

Marquez, [email protected] Informatics, CSIRO, Clayton, Vic, Aus-tralia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-11

Martell, [email protected] of Forestry, University of Toronto, Toronto, On-tario, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-04

Martin-Herran, [email protected], Universidad de Valladolid, Valladolid, Spain RD-09

Masinga, [email protected] and Applied Mathematics, University of theWitwatersrand Johannesburg, Johannesburg, South AfricaRD-05

Maturana, [email protected] Industrial y de Sistemas, P. Universidad Catolicade Chile, Santiago, Chile . . . . . . . . . . . . . . . . . . . . . TE-04

Mbohwa, [email protected] and Operations Management, University of Johan-nesburg, Johannesburg, Gauteng, South Africa . . . RD-17

Mesa, Juan [email protected] of Seville, Sevilla, Spain . . . . . . . . . . . MA-03

Montibeller, [email protected]. of Management, London School of Economics, Lon-don, United Kingdom . . . . . . . . . . . . . . . . . . . . . . . MC-08

Müller-Merbach, [email protected], Universität Kaiserslautern,Kaiserslautern, Germany . . . . . . . . . . . . . . . . . . . . . RA-10

Murthy, Rajluxmi [email protected] Methods and Information Systems, Indian

77


Institute of Management, Bangalore, Karnataka, India TE-03

Nailana, Freddy [email protected] of Computer Science, University of Limpopo -Turfloop Campus, Polokwane, South Africa . . . . . MC-19

Nakai, [email protected] of Education, Chiba University, Chiba, Japan MA-02

Ncube, [email protected] School of Business Leadership, University ofSouth Africa, Pretoria, Gauteng, South Africa . . . . RC-11

Newman, [email protected] of Economics and Business, Colorado School ofMines, Golden, CO, United States . . . . . . . . . . . . . MC-04

Novaes, Antonio [email protected] Engineering, Federal University of Santa Cata-rina, Florianopolis, SC, Brazil . . . . . . . . . . . . . . . . RD-14

Nsakanda, Aaron [email protected] School of Business, Carleton University, Ottawa,Ontario, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-18

Nyamugure, [email protected], National University of Science and Technol-ogy, Bulawayo, Zimbabwe, Zimbabwe . . . . . . . . . TD-19

Odegaard, [email protected] Business School, Western University, London, On-tario, Canada . . . . . . . . . . . . . . . . TC-05, TD-05, TE-05

Okesola, [email protected] Sciences:Operations Research, University ofSouth Africa, Pretoria, South Africa, South Africa FB-17

Oliveira, José [email protected] TEC, Faculty of Engineering, University of Porto,Porto, Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-18

Osman, Ibrahim [email protected] School of Business, American University of Beirut,Beirut, Lebanon . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-01

Ozekici, [email protected] of Industrial Engineering, Koç University, Is-tanbul, Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-13

Pastor, Jesús [email protected]. de Estadística y Matemática Aplicada, Universidad

Miguel Hernández, Elche (Alicante), Spain . . . . . . TD-14

Pasupathy, [email protected] of Missouri, Columbia, United States . . TE-07

Pilgrim, [email protected] University of Sheffield, Sheffield, United KingdomRE-07

Pioro, [email protected] of Telecommunications, Warsaw University ofTechnology, Warszawa, Poland . . . . . . . . . . . . . . . . TA-18

Pizarro, [email protected]. Informática, Estadística y Telemática, UniversidadRey Juan Carlos, Mostoles, Madrid, Spain . . . . . . . TA-09

Qiang, [email protected] and Environmental Engineering, National Universityof Singapore, Singapore, Singapore . . . . . . . . . . . . ME-11

Rand, [email protected]. of Management Science, Lancaster University, Lan-caster, Lancashire, United Kingdom . . . . . TD-05, TE-05

Randall, [email protected] St, HM Treasury, London, United KingdomTA-07

Resende, [email protected] & Optimization Research, AT&T Labs Re-search, Florham Park, NJ, United States . . . . . . . . TC-18

Ryan, [email protected] Science, The University of Auckland, Auck-land, New Zealand . . . . . . . . . . . . . . . . . . . . . . . . . MD-03

Salazar González, Juan José[email protected]ística e Investigación Operativa, Universidad de LaLaguna (Tenerife), La Laguna, Tenerife, Spain . . . RD-16

Salminen, [email protected] of Business and Economics, University ofJyväskylä, Jyväskylä, Finland . . . . . . . . . . . . . . . . . TD-08

Salo, [email protected] Analysis Laboratory, Aalto University School ofScience and Technology, Aalto, Finland . . . . . . . . MA-08

Salomon, Valé[email protected], UNESP - Sao Paulo State University,Guaratingueta, Sao Paulo, Brazil . . . . . . . . . . . . . . ME-14

78


Schauten, [email protected] of Economics and Business Economics, FinanceGroup, Erasmus University Rotterdam, Rotterdam, Nether-lands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-12

Schmidt, Gü[email protected] University, Saarbrücken, Germany . . . . . RA-12

Schriber, Thomas [email protected] Information Technology, University of Michigan,Ann Arbor, Michigan . . . . . . . . . . . . . . . . . . . . . . . MC-09

Shaikh, [email protected] of Business Administration, Prince MohammadBin Fahd University, Al-Khobar, Saudi Arabia . . . FA-08

Sher, [email protected] of Planning, Israel Police, Modiin, Israel TA-03

Sillekens, [email protected] Support and OR Lab, University of Paderborn,Paderborn, Germany . . . . . . . . . . . . . . . . . . . . . . . . TE-10

Sniedovich, [email protected]. of Mathematics and Statistics, University of Mel-bourne, Melbourne, Victoria, Australia . . . . . . . . . RE-09

Sobolev, [email protected] Care & Epidemiology, University of BritishColumbia, Vancouver, BC, Canada . . . . . . . . . . . . . TC-07

Sodini, [email protected] and Applied Mathematics, University of Pisa,Pisa, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-02

Sörensen, [email protected] of Applied Economics, University of Antwerp,Antwerpen, Belgium . . . . . . . . . . . . . . . . . . . . . . . . FB-02

Speranza, M. [email protected]. of Quantitative Methods, University of Brescia,Brescia, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RB-01

Stewart, [email protected] Sciences, University of Cape Town, Ronde-bosch, South Africa . . . . . . . . . . . . . . . . . . TA-08, FB-09

Syntetos, [email protected] Business School, Cardiff University, CARDIFF,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-19

Takehara, Hitoshi

[email protected] School of Finance, Accounting and Law, WasedaUniversity, Chuo-ku, Tokyo, Japan . . . . . . . . . . . . MD-13

Talla Nobibon, [email protected] Sciences and Information Management, KU Leu-ven, Leuven, Belgium . . . . . . . . . . . . . . . . . . . . . . . TA-02

Tambovceva, [email protected] of Engineering Economics and Management, RigaTechnical University, Riga, Latvia . . . . . . . . . . . . . RC-15

Teixeira de Almeida, [email protected] Engineering, Universidade Federal de Per-nambuco - UFPE, Recife, PE, Brazil . . . . . . . . . . . ME-19

Teo, Chung [email protected] University of Singapore, Singapore, SingaporeTE-01

Toth, [email protected], University of Bologna, Bologna, Italy . . . . . TD-01,MC-03

Tran-Thanh, [email protected], Budapest University of Technologyand Economics, Budapest, Hungary . . . . . . . . . . . . TE-18

Trick, [email protected] School of Business, Carnegie Mellon University,Pittsburgh, PA, United States . . . . . . . . . . . . . . . . . RD-10

Tsao, [email protected]. of Finance, National Pingtung Institute of Com-merce, Pingtung, Taiwan . . . . . . . . . . . . . . . . . . . . . FB-07

Turkay, [email protected] of Industrial Engineering, Koc University, Is-tanbul, Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-10

van Dyk, [email protected] and Quantitative Methods, CSIR Built Environ-ment, Stellenbosch, Western Cape, South Africa . . TD-04

van Vuuren, [email protected] of Industrial Engineering, Stellenbosch Uni-versity, Stellenbosch, Western Cape, South Africa FA-03,MD-18

Vieira, Luí[email protected] Civil, Faculdade de Engenharia do Porto, Porto,Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-18

Visagie, Stephan

79


[email protected] of Logistics, University of Stellenbosch, Stel-lenbosch, South Africa . . . . . . . . . . . . . . . . . . . . . . RE-02

Wakolbinger, Lea [email protected] of Business Administration, University of Vi-enna, Vienna, Austria . . . . . . . . . . . . . . . . . . . . . . . MD-09

Wallenius, [email protected] and Service Economy, Aalto UniversitySchool of Business, Helsinki, Uusimaa, Finland . . TE-08

Weber, [email protected] de l’Innovation et de la Recherche, SNCF, Paris,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-03

Weber, [email protected] of Applied Mathematics, Middle East TechnicalUniversity, Ankara, Turkey . . . . . . . . . . . . . . . . . . . FB-10

Westermann, [email protected] Software GmbH, Koeln, Germany . . . . . . MD-17

White, [email protected]

Management Department, University of Bristol, Bristol,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-05

Wilke, [email protected] of Pretoria, Pretoria, South Africa . . . . MC-02

Yoneda, [email protected] of Economics, Fukuoka University, Fukuoka,Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-15

Zabinsky, Zelda [email protected] Engineering, University of Washington, Seattle,WA, United States . . . . . . . . . . . . . . . . . . . . . . . . . . TC-02

Zaccour, [email protected] Sciences Department, GERAD & HEC Mon-treal, Montreal, Quebec, Canada . . . . . . . . . . . . . . RC-09

Zadnik Stirn, [email protected] Faculty, University of Ljubljana, Ljubljana,Slovenia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-08

Zhigljavsky, [email protected] of Mathematics, Cardiff University, Cardiff RE-06

80

Author Index

, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-01, FC-01, MA-01, MC-01, TA-01, TD-01, TE-01, FA-02, FB-02,MC-02, ME-02, RA-02, RE-02, TC-02, TD-02, FB-03, MA-03, MC-03, MD-03, RC-03, RD-03, RE-03,TA-03, TC-03, RE-04, TC-04, TD-04, TE-04, FA-

05, MA-05, MC-05, RE-05, TA-05, TC-05, FA-06,FB-06, RD-06, RE-06, ME-07, RA-07, RC-07, TA-07, TD-07, TE-07, FB-08, MA-08, MD-08, RC-08,TC-08, TD-08, TE-08, FA-09, ME-09, RA-09, RE-09, TA-09, TC-09, TD-09, FA-10, FB-10, RC-10,RD-10, TA-10, TE-10, FA-11, MD-11, RA-11, RE-11, TC-11, TD-11, FB-12, MC-12, MD-12, ME-12,RA-12, TC-12, FA-13, FB-13, RC-13, RD-13, TC-13, FA-14, FB-14, MC-14, ME-14, RA-14, RC-14,RE-14, TE-14, MC-15, RC-15, TA-15, TC-15, FA-

16, FB-16, MC-16, MD-16, RA-16, RC-16, RD-16,TA-16, TC-16, TE-16, FB-17, ME-17, RD-17, FA-18, FB-18, MC-18, MD-18, RC-18, RD-18, RE-18,TE-18, MA-19, MD-19, ME-19, RC-19, TC-19

Abdul-Jalbar, [email protected] de La Laguna, La Laguna, Tenerife, SpainTE-09

Absi, [email protected] des Mines de Saint-Etienne - LIMOS, GARDANNE,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-11

Adendorff, [email protected], None, Pretoria, Gauteng, Afghanistan . . . . . TD-03

Adetunji, [email protected] and Systems Engineering, University of Pretoria,Pretoria, Gauteng . . . . . . . . . . . . . . . . . . . . . . . . . . MD-13

Adewumi, [email protected] of Computer Sciences, University of Lagos,Akoka,Y aba, Lagos, Nigeria . . . . . . . . . . MD-07, FB-18

Adeyefa, [email protected] Sciences, University of South Africa, Pretoria,Gauteng . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-02

Adler, [email protected] of Business Administration, Hebrew University ofJerusalem, Jerusalem, Israel . . . . . . . . . . . . . . . . . . TA-03

Agarwal, [email protected] Learning, Yahoo Research, Santa Clara, CA,United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-06

Agnetis, [email protected] dell’Informazione, Università di Siena, Siena,Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-13

Agussurja, [email protected] Logistics Institute - Asia Pacific, National Universityof Singapore, Singapore, Singapore . . . . . . . . . . . MD-16

Ahn, [email protected] für Controlling und Unternehmensrechnung, TUBraunschweig, Braunschweig, Germany . . . . . . . . TD-14

Ahn, Jin [email protected] Yonsei university, Seoul, Korea, Republic Of . FB-03

Aissaoui, [email protected] Optimization Research Group - ROI, EcolePolytechnique de Tunisie, La Marsa, Tunisia . . . . ME-07

Akume, [email protected] Science Department, University of Buea, Buea,Cameroon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-13

AL ANI, [email protected], ESIEE-PARIS, NOISY-LE-GRAND, France . MA-02

Albizuri, [email protected] Economics IV, UPV/EHU, Bilbao, Spain RD-09

Alcaide Lopez de Pablo, [email protected], Investigacion Operativa y C., Universidad deLa Laguna, La Laguna, Tenerife, Spain . . . . . . . . . RD-13

Alexandra, [email protected] Mathematics, Moldova State University, Chisinau,Moldova, Republic Of . . . . . . . . . . . . . . . . . . . . . . . TA-07

Alfandari, [email protected] Business School, Cergy-Pontoise Cedex, FranceMC-03

Ali, [email protected] of Computational and Applied Mathematics, Uni-versity of the Witwatersrand, Johannesburg, South AfricaFB-18

Ali, [email protected] of Witwatersrand, Johannesburg, South AfricaMD-07

Aliakbari, [email protected], Tehran, Iran, Islamic Republic Of . . . . . . . MD-14

Aliev, [email protected] of Mathematics, Cardiff University, Cardiff RD-02

81

AUTHOR INDEX IFORS 2008 - Sandton

Alifas Yeko, [email protected] Methods, Makerere University, Kampala,Uganda, Uganda . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-19

Alirezaee, [email protected] of Mathematics, Iran University of Science andTechnology, Tehran, Iran, Islamic Republic Of . . MD-14,RC-14

Almeida-Dias, [email protected] Superior Técnico, Technical University of Lisbon,Porto Salvo, Portugal . . . . . . . . . . . . . . . . . . . . . . . ME-08

Alonso-Ayuso, [email protected] & Operations Research Department, Rey JuanCarlos University, Mostoles, Madrid, Spain . . . . . . TA-09

alrwis, [email protected] Economics, King Saud University, Riyadh,Afghanistan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-14

Altay, [email protected], DePaul University, Chicago, Illinois, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-19

Altemeier, [email protected] Integrated Manufacturing, University of Pader-born, Paderborn, Germany . . . . . . . . . . . . . . . . . . . RE-04

Alvarez-Valdes, [email protected] and Operations Research, University of Valencia,Burjassot, Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-13

Alvelos, [email protected] de Produção e Sistemas, Universidade doMinho, Braga, Portugal . . . . . . . . . . . . . . . . . . . . . . TA-18

Alves, [email protected] de Economia, Universidade do Porto, Porto,Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-13

Amar, A. [email protected] Department, Seton Hall University, South Or-ange, NJ, United States . . . . . . . . . . . . . . . . . . . . . . TA-12

Amodeo, [email protected] Delaunay Institute, University of Technology ofTroyes, Troyes, France . . . . . . . . . . . . . . . RA-16, FB-18

Andersson, [email protected] of Defence Analysis, FOI Swedish Defence Re-search Institute, Stockholm, Sweden . . . . . . . . . . . FB-03

Andreeva, [email protected] School, University of Edinburgh, Edinburgh,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-12

Angeloudis, [email protected] & Environmental Engineering Dept, Imperial CollegeLondon, London, United Kingdom . . . . . . . . . . . . MC-11

Anido, [email protected] Analysis: Quantitatve Economy, AutonomaUniversity of Madrid, Madrid, Spain . . . . . . . . . . . FB-15

Anily, [email protected] of Management, Tel Aviv University, Tel Aviv, Is-rael . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-09

Ankamah, Johnson [email protected], TU/e, Eindhoven, Netherlands . . . . RD-03

Ansell, [email protected] of Edinburgh, Edinburgh, Midlothian, UnitedKingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-12

Arapoglu, R. [email protected] Engineering Department, Eskisehir OsmangaziUniversity, Eskisehir, Turkey . . . . . . . . . . . . . . . . . RA-16

Araujo, Marcelo Tozo [email protected] of Physics, Sao Paulo State University, SãoJosé do Rio Pret, São Paulo, Brazil . . . . . . . . . . . . . RE-18

Archibald, [email protected] Consulting Limited, Brighton, East Sussex,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-05

Arenales, [email protected] of Applied Mathematic and Statistics, universidadede São Paulo, Sao Carlos, SP, Brazil . . . . RC-18, RD-18

Arivarignan, [email protected] of Applied Mathematics and Statistics,Madurai-Kamaraj University, Madurai, Tamilnadu, IndiaMA-16, RC-19

Armentano, Viní[email protected] de Engenharia Elétrica e de Computação, Uni-versidade de Campinas, Campinas, São Paulo, Brazil RE-18

Arns Steiner, Maria [email protected], Federal University of Parana, Curitiba,Parana, Brazil . . . . . . . . . . . . . . . . . . . . . . . TE-07, TD-15

82

IFORS 2008 - Sandton AUTHOR INDEX

Aspillaga, [email protected]. Economía Agraria, Pontificia Universidad Catolicade Chile, Santaigo, Chile . . . . . . . . . . . . . . . . . . . . ME-04

Averbakh, [email protected], University of Toronto at Scarborough,Toronto, Ontario, Canada . . . . . . . . . . . . . . . . . . . . FB-11

Ayeni, [email protected] of Computer Science, University of Lagos,Akoka, Lagos, Nigeria . . . . . . . . . . . . . . . MD-07, FB-18

BABAI, Mohamed [email protected] Management School, Talence, Gironde,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-19

Babazadeh, [email protected] Iranian Gas Company, Tehran, Iran, Islamic Re-public Of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-14

Bainbridge, [email protected] University, Caulfield East, Victoria, Australia FA-14

Baker, [email protected] of Business, University of Salford, Salford, Lan-cashire . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-12, TA-19

Bamenimoghadam, [email protected] Department, Allameh Tabatabaei University,Tehran, Tehran, Iran, Islamic Republic Of . . . . . . . RA-14

BARBER, [email protected] Informaticos Y Computacion, Universidad Po-litecnica De Valencia, VALENCIA, SPAIN, Spain ME-03

Bartling, [email protected] Institute IAIS, Sankt Augustin, Germany . FA-18

Barty, [email protected]/R35, EDF, Clamart, France . . . . . . . . . . . . FA-01

Bean, [email protected] Environment, CSIR, Pretoria, Gauteng, South AfricaTD-16

Beasley, J. [email protected] Sciences, Brunel University & JB Consul-tants, Uxbridge, Middlesex, United Kingdom . . . MC-12

Beauregard, Robert

[email protected] Laval, quebec, qc, Canada . . . . . . . . . . . TC-04

Bekker, [email protected] Engineering, Stellenbosch University, Stellen-bosch, Western Cape, South Africa . . . . . . . . . . . . ME-09

Belderrain, Mischel Carmen [email protected] Engineering, Instituto Tecnologico de Aero-nautica, Sao Jose dos Campos, SP, Brazil . . . . . . . TC-05

Belgorodskiy, [email protected] Business School, The University of Salford, Sal-ford . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-12

Bell, [email protected] and Environmental Engineering, Imperial CollegeLondon, London . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-11

Bell, [email protected] College, London, United Kingdom . . . . . TE-09

Bell, [email protected] University, Ivey Business School, London, On-tario, Canada . . . . . . . . . . . . . . . . TA-05, TD-05, TE-05

Bellezza, [email protected] Business School, Barcelona, Spain . . . . . . . MA-08

Bellotti, [email protected] School and Economics, University of Edin-burgh, Edinburgh, United Kingdom . . . . . RD-06, TD-13

belouadah, [email protected] of Business & Economy, University of MohamedBoudiaf, Msila, Algeria . . . . . . . . . . . . . . . . . . . . . . TA-01

Belton, [email protected]. Management Science, University of Strathclyde,Glasgow, United Kingdom . . . . . . . . . . . . . . . . . . . MC-08

Ben-Yair, [email protected] Engineering & Management, Sami ShamoonCollege of Engineering, Beer-Sheva, Israel . . . . . . TA-10

Bensoussan, [email protected] of Management, University of Texas at Dallas,Richardson, TX, United States . . . . . . . . . . . . . . . . ME-16

Berk, [email protected] Science, Ankara, Turkey . . . . . . . . . . . . . . RC-14

Berrima, Kamel

83


[email protected] Headquarters, Tunis-Carthage, Tunisia . ME-07

Bertazzi, [email protected]. of Quantitative Methods, University of Brescia,Brescia, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-01

Bester, [email protected] Modelling, PIC Solutions, Brackenfell MC-12

Bhargava, [email protected] School of Management, UC Davis, davis, CARA-11

Bieda, [email protected], AGH-University of Science and Technology,Krakow, Poland . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-08

Bierwirth, Christianchristian.bierwirth@wiwi.uni-halle.deMartin-Luther-University Halle-Wittenberg, Halle, Ger-many . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-11, MD-11

Billaut, [email protected] d’Informatique, University of Tours, Tours,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-13

Bleys, [email protected], Vrije Universiteit Brussel, Brussels, Belgium RD-15

Bodmer, [email protected] Science Center, Technical University of Munich,Freising, Bavaria, Germany . . . . . . . . . . . . . . . . . . MD-14

Bosch, [email protected] Engineering, University of Chile, Santiago, ChileTC-19

Botton, [email protected] School of Management - CORE, Universitécatholique de Louvain, Louvain-la-Neuve, Belgium . ME-18

Bouza, [email protected] de la Habana, La Habana, Cuba . . . . MA-03

Bozhkov, [email protected] Business School, Uxbridge, United Kingdom RA-12

Bragge, [email protected] Technology, Helsinki School of Economics,Helsinki, Finland . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-08

Braude, [email protected] and Business Administration, College of Judeaand Samaria, Ariel, Israel . . . . . . . . . . . . . . . . . . . MD-12

Bräysy, [email protected] Solutions Ltd, Oulu, Finland . . . . . . . . . MC-01

Bredström, [email protected] School of Economics and Business Adminis-tration, Beregn, Norway . . . . . . . . . . . . . . . . . . . . . TC-04

Breton, Michè[email protected] and HEC Montréal, Canada . . . . . . . . . . . TC-12

Brewis, [email protected] Wholesale, British Telecom, London . . . . . . . . RE-09

Brigatti, [email protected] Information Technology, University ofJyväskylä, Jyväskylä, Finland . . . . . . . . . . . . . . . . . TE-09

Brugha, [email protected] for Business Analytics, University College Dublin,Dublin 4, Ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-05

Brunner, Daniel [email protected] Analysis, Kiva Systems, Inc., Woburn, Mas-sachusetts, United States . . . . . . . . . . . . . . . . . . . . MC-09

Buratto, [email protected] of Mathematics, University of Padova, Padova,Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-09

Bussieck, [email protected] Software GmbH, Cologne, Germany . . . . MD-17

Cakanyildirim, [email protected] Management, University of Texas at Dallas,Richardson, TX, United States . . . . . . . . . . . . . . . . ME-16

Caldentey, [email protected] School of Business, New York University, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-16

Callea, [email protected] di Ingegneria Gestionale, Università degliStudi di Bergamo, Dalmine, Italy . . . . . . . . . . . . . . RA-07

Cambini, [email protected] and Applied Mathematics, University of Pisa,Pisa, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-02

84


Canakoglu, [email protected] Engineering Department, Bahcesehir University,Istanbul, Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-13

Cangalovic, [email protected] for Operational Research, Faculty of Organiza-tional Sciences, University of Belgrade, Belgrade, Serbia,Serbia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-18

cano-sevilla, francisco josé[email protected] and Operations Research, Complutense Univer-sity, madrid, madrid, Spain . . . . . . . . . . . . . . . . . . . RA-04

Carlos Botter, [email protected] Naval and Oceanic Engineering, University ofSão Paulo, São Paulo, São Paulo, Afghanistan . . . RE-16

Caro, [email protected] Anderson School of Management, Los Angeles,CA, United States . . . . . . . . . . . . . . . . . . MC-04, ME-16

Carrillo, [email protected] Universidad Cátólica de Chile, Santiago, ChileTE-04

Casier, Auré[email protected]épartment d’Informatique, ULB - Université Libre deBruxelles, Brussels, Belgium . . . . . . . . . . . . . . . . . TA-13

Castellano, [email protected] and Law, University of Macerata, Macerata,Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-12

Catanzaro, [email protected] School of Management & CORE, UniversitéCatholique de Louvain, Mons, Belgium . . . . . . . . . TD-01

Çatay, Bü[email protected] of Eng. & Natural Sciences, Sabanci University,Istanbul, Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-11

Cavalcante, [email protected] University of Pernambuco, Recife, PE, Brazil ME-19

Cáceres, Ma [email protected]ática Aplicada I, Universidad de Sevilla, Sevilla,Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-09

Cerqueti, [email protected] of Economics and Law, University of Macer-ata, Macerata, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . FB-12

Chakraborty, [email protected] Profitability Improvement, Sasol TechnologyPty Ltd., Secunda, South Africa . . . . . . . . . . . . . . . FB-15

Chandra, [email protected], IIT Delhi, New delhi, Delhi, India . RD-06

Chang, [email protected] of Engineering Management, Minghsin Univer-sity of Science and Technology, Hsinchu County, TaiwanTC-10

Chattopadhyay, Asis [email protected], Calcutta University, Kolkata, West Bengal . FA-08

Chaudhary, [email protected] of Mathematics, Dr. B.R.Ambedkar Univer-sity, AGRA, UP, India . . . . . . . . . . . . . . . . . . . . . . MC-19

Chehade, [email protected] Losi, University of Technology of Troyes, Troyes,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-18

Chemak, [email protected] Rurale, INRAT, Ariana, Tunisia . . . . . . RD-14

Chemak, [email protected] d’économie rurale, Institut National de laRecherche Agronomique de Tunisie, Ariana, Tunisia MD-07

Chen, [email protected] Business School, University of Warwick, Coven-try, United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . TE-18

Chen, Frank [email protected] of Management Science, City University ofHong Kong, Hong Kong, China . . . . . . . . . . . . . . . ME-16

Chen, [email protected] State University, Murray, Kentucky, United StatesRD-03

Chen, [email protected] of Applied Mathematics, Academy of Mathemat-ics and Systems Science, Chinese Academy of Sciences,Beijing, China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-18

Cheng, [email protected] Engineering & Engineering Management, TheChinese University of Hong Kong, Hong Kong . . MC-16

85


Cherri, [email protected], UNESP - Bauru, Bauru, SP, Brazil . RD-18

Chevalier, [email protected], ESCP-EAP, PARIS, PARIS, France . . TA-07

Chew, Ek [email protected] and Systems Engineering, Faculty of Engineer-ing, National University of Singapore, Singapore MC-11

Chikumbo, [email protected] Forests and Environment, Rotorua, BOP, NewZealand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-15

Chikwasha, [email protected] of Mathematical Sciences, Harare Institute of Tech-nology, Harare, Zimbabwe . . . . . . . . . . . . . . . . . . . RD-17

Chilcott, [email protected], University of Sheffield, Sheffield, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-07

Chin, [email protected] Engineering and Engineering Management,City University of Hong Kong, Hong Kong, China TC-08

Chipoyera, [email protected] of Statistics & Actuarial Science, University of ofthe Witwatersrand, Johannesburg, Gauteng, South AfricaMA-16

Chirenje, Kelvin [email protected] Maths, National University of Science and Tech-nology, Chitungwiza, Harare, Zimbabwe . . . . . . . . RD-07

CHIRIPANHURA, [email protected] Mathematics ( Operations Research And AppliedStatistics), National University Of Science And Technol-ogy Zimbabwe, BULAWAYO, Zimbabwe . . . . . . . RD-07

Choi, [email protected] and Management Engineering, Hankuk Univer-sity of Foreign Studies, Yongin-shi, Kyonggi-Do, Korea,Republic Of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-19

Choo, [email protected] of Business Administration, Simon Fraser Univer-sity, Burnaby, BC, Canada . . . . . . . . . . . . . . . . . . . ME-14

Chou, [email protected] University of Singapore, Singapore, SingaporeTC-11

Chu, [email protected] of Management and Information Technol-ogy, Southern Taiwan University, Yungkang City, TainanCounty, Taiwan . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-07

Cileg, [email protected] Methods in Economy, Faculty of Economics,Subotica, Serbia . . . . . . . . . . . . . . . . . . . . MC-02, TD-11

Cinar, [email protected] Engineering, Istanbul Technical University, is-tanbul, Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-17

Ciric, [email protected] Informatics and Quantitative Methods, Faculty ofEconomics Subotica, Subotica, Serbia, Serbia . . . . FA-14

Clark, [email protected] of Engineering Design and Mathematics, Universityof the West of England, Bristol, United Kingdom . TD-07

Cochran, [email protected] of Marketing and Analysis, Louisiana TechUniversity, Ruston, LA, United States . . . . . . . . . . TA-04

Codina, [email protected] and Operational Research, UPC, Barcelona,Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-14, FA-16

Coene, [email protected], KULeuven, Leuven, Belgium . . . . . . . . . FB-11

Coetzer, [email protected] and Development, Sasol Technology (Pty) Ltd.,Sasolburg, Free State . . . . . . . . . . . . . . . . ME-02, FB-14

Colebrook, [email protected]. de Estadistica, Investigacion Operativa y Computa-cion, Universidad de La Laguna, La Laguna, Santa Cruz deTenerife, Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-09

colin, [email protected] university of Versailles, Versailles, France . FB-06

Companys, [email protected] Business Administration, Universitat Politec-nica Catalunya, Barcelona, Spain . . . . . . . . . . . . . . RE-11

Consoli, [email protected] Research Centre, European Commission, Ispra,Varese, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-18

86


Contreras, [email protected], Chile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-19

Cooper, [email protected] Environment, CSIR, Pretoria, South Africa . FA-05,TE-16

Corbett, [email protected] Anderson School of Management, Los Angeles,CA, United States . . . . . . . . . . . . . . . . . . . RA-11, TA-11

Corman, [email protected] Engineering and Logistics, Maritime and Trans-port Technology, Delft University of Technology, Delft,Netherlands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-03

Cortés, Cristiá[email protected] Engineering Department, Universidad de Chile, San-tiago, Chile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-01

Cosma, [email protected], Université du Luxembourg, Luxembourg,Luxembourg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-12

Crespo, [email protected] University, Spain . . . . . . . . . . . . . . . . . . . . FA-13

Crook, [email protected] of Edinburgh Business School, University ofEdinburgh, Edinburgh, Lothian, United Kingdom . TD-13

Cruz-Casas, [email protected] and Associates, Inc., Ft. Lauderdale, FL,United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-09

Dag, [email protected], Ireland . . . . . . . . . . . . . . . . . . . . . . . . . . TC-11

Dahlmann, André[email protected] für Technische Dienstleistungen und OperationsManagement, Technische Universität München, München,Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-18

Daniell, [email protected] for Resource and Environmental Studies, AustralianNational University, Canberra, ACT . . . . . . . . . . . . TA-08

Daouia, [email protected] School of Economics, University of Toulouse,Toulouse, France . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-14

Darby-Dowman, [email protected]

Mathematical Sciences, Brunel University, Uxbridge, Mid-dlesex, United Kingdom . . . . . . . . . . . . . . . . . . . . MD-18

Dasu, [email protected] of Southern California, Los Angeles, CA,United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-11

Dawidson, [email protected], DELICOM, Paris, France, France . . FA-05

de Almeida Filho, [email protected] of Southampton, Southampton, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-12

De Almeida, [email protected] & Research Department, SNCF, Paris CEDEX08, France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-03

De Brucker, [email protected] of Economics and Business - Research Centre forEconomics and Corporate Sustainability (CEDON), KULeuven - University of Leuven (Campus HU Brussel), Bel-gium, Brussels, Belgium . . . . . . . . . . . . . . . . . . . . . RD-15

de Frutos, [email protected], Universidad de Valladolid, Valladolid, Spain TC-12

de Koster, René[email protected] School of Management, Erasmus UniversityRotterdam, Rotterdam, Netherlands . . . . . . . . . . . . RD-16

de Rooij, [email protected], Haarlem, Netherlands . . . . . . . . . . . . . . . TC-17

De Smet, [email protected] - Code, Université Libre de Bruxelles, Bruxelles, Bel-gium . . . . . . . . . . . . . . . . . . . . . . ME-08, RA-08, TA-13

de Sousa, [email protected] de Aveiro, Instituto de Telecomunicações,AVEIRO, Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . TA-18

de Werra, [email protected], EPFL, Lausanne, Switzerland . . . . . . . . . . . . TD-01

De Witte, [email protected] for Economic Studies, University of Leuven (KUL),Leuven, Afghanistan . . . . . . . . . . . . . . . . . TC-14, RE-15

Dekker, [email protected] University Rotterdam, Rotterdam, NetherlandsMD-19

87


Demange, [email protected] Business School and LAMSADE UMR 7243,Paris, France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-03

Deo, [email protected] School of Management, Northwestern University,Evanston, IL, United States . . . . . . . . . . . . . . . . . . . TA-11

Despontin, [email protected], Vrije Universiteit Brussel, Brussels, AfghanistanRD-15

Detta-Silveira, Jose [email protected] & Operations Engineering, Instituto Technologi-coautonomo de Mexico, Mexico city, Mexico . . . MC-07

Detti, [email protected] Engineering Department, University of Siena,Siena, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-13

Di Stefano, [email protected] of L’Aquila, Monteluco di Roio, L’Aquila, ItalyMC-03

Diatha, Krishna [email protected] and Operations Management, Indian Instituteof Management, Bangalore, Karnataka, India . . . . TE-03

Ding, [email protected] of Business, Singapore Management University,Singapore, Singapore . . . . . . . . . . . . . . . . . . . . . . . . TC-16

Dodin, [email protected] Graduate Scool of Management, University ofCalifornia, Riverside, California, United States . . . TC-10

Doerner, [email protected] of Business Studies, University of Vienna, Vi-enna, Vienna, Austria . . . . . . . . . . . . . . . . . . . . . . . TD-09

Dohn, [email protected] Management Engineering, Technical University ofDenmark, Kgs. Lyngby, Denmark . . . . . . TE-07, RC-13

Dominguez, [email protected]. of Computer Science, E.T.S.I.Informatica - Univer-sity of Malaga, Malaga, Spain . . . . . . . . . . . . . . . . . TA-09

Dowart, [email protected] of Applied Mathematics, National Universityof Science and Technology (NUST), Harare, ZimbabweMA-16

du Plooy, [email protected], South Africa . . . . . . . . . . . . . . . . . . RD-17

Du Plooy, [email protected] Sciences, University of South Africa, Pretoria,Gauteng, South Africa . . . . . . . . . . . . . . . . . . . . . . . TA-05

du Toit, [email protected] of Logistics, University of Stellenbosch, Stel-lenbosch, South Africa . . . . . . . . . . . . . . . . . . . . . . MD-16

du toit, [email protected] of Mathematical Sciences, University of Stel-lenbosch, Stellenbosch, Afghanistan . . . . . . . . . . . . FA-03

Duarte, [email protected] University of Rio de Janeiro, Rio de Janeiro,Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-04

Duenas, [email protected], University of Sheffield, Sheffield . . . . . RE-07

Dullaert, [email protected] of Economics and Business Administration, VUUniversity Amsterdam, Amsterdam, Netherlands MC-01

Dunstall, [email protected], Informatics and Statistics, CSIRO, SouthClayton, Victoria, Australia . . . . . . . . . . . . . . . . . . TD-04

Durbach, [email protected] Sciences, University of Cape Town, Cape Town,Western Cape, South Africa . . . . . . . . . . . . . . . . . . TD-08

Dymond, [email protected] and Aeronautical Engineering, University ofPretoria, Pretoria, Gauteng . . . . . . . . . . . . . . . . . . . RA-02

Dzida, [email protected] of Telecommunications, Warsaw University ofTechnology, Warsaw, Poland . . . . . . . . . . . . . . . . . ME-18

Eglese, [email protected] Management School, Lancaster University, Lancaster,Lancashire, United Kingdom . . . . . . . . . . . . . . . . . ME-01

Ehrgott, [email protected] Science, Lancaster University, Lancaster,United Kingdom . . . . . MD-03, MD-08, TA-08, TC-16

el Dsoki, [email protected]

88


Fraunhofer Institute for Structural Durability LBF, Darm-stadt, Hessia, Germany . . . . . . . . . . . . . . . . . . . . . . TC-01

El Ouardighi, [email protected] Management, ESSEC Business School, CergyPontoise, France . . . . . . . . . . . . . . . . . . . . RA-09, RC-09

Eleria, [email protected] Research Center, University Of Santo Tomas,Manila, Philippines . . . . . . . . . . . . . . . . . . . . . . . . . TA-14

Elimam, [email protected] Engineering, Professor American Universityin Cairo, New Cairo, Cairo, Egypt . . . . . . . . . . . . . TC-10

Elphinstone, [email protected] Built Environment, Logistics and Quantitative Meth-ods, Pretoria, South Africa . . . . . . . . . . . . . . . . . . . FA-05

Emrouznejad, [email protected] Business School, Aston University, Birmingham,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-15

Engelbrecht, Andries [email protected] Science, University of Pretoria, Pretoria, SouthAfrica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-04, FA-06

Engelbrecht, Gerhard [email protected], Clubview, South Africa . . . . . . . . . . . . . . . . FB-03

Eppe, [email protected] Faculty, CoDE-SMG„ Brussels, BelgiumME-08

Ernst, [email protected] and Information Sciences, CSIRO, ClaytonSouth, Vic, Australia . . . . . . . . . . . . . . . . . RC-11, RE-11

Escudero, Laureano [email protected]. de Estadística e Investigación Operativa, UniversidadRey Juan Carlos, Mostoles (Madrid), Spain TA-09, TE-10

Eskelinen, [email protected] University School of Business, Aalto, Finland . FB-09

ESOGBUE, [email protected] & Systems engineering, Georgia Institute ofTechnology, Atlanta, GA, United States . . . . . . . . . RC-02

Esser, [email protected] for Industrial Production, Universität Karlsruhe(TH), Karlsruhe, Germany . . . . . . . . . . . . . . . . . . . . FA-12

Evans, [email protected] Risk Assurance, DBSA, Midrand, South AfricaTD-15

Ewusi-Mensah, [email protected] Information Systems, Loyola Marymount Uni-versity, Los Angeles, CA, United States . . . . . . . . . TC-15

Ezen, [email protected] Engineering, Halic University, Istanbul, TurkeyTC-19

Falbo, [email protected] of Economics and Management, University ofBrescia, Brescia Bs, Italy . . . . . . . . . . . . . . . . . . . . . FB-12

Fan, L. [email protected], Institute of Systems Design and Optimization,Kansas State University, Manhattan, Kansas, United StatesRD-04

Fasina, [email protected] of Computer Science, University of Lagos,Akoka, Lagos, Nigeria . . . . . . . . . . . . . . . MD-07, FB-18

Fatti, L. [email protected] & Actuarial Science, University of the Witwater-srand, Johannesburg, South Africa . . . . . MC-07, TE-13

Faulin, [email protected]. Statistics and Operations Research, Public Univer-sity of Navarre, Pamplona, Navarra, Spain TA-05, TD-09

Fehr, [email protected] (Institute for Operations Research), ETHZurich, Zurich, Switzerland . . . . . . . . . . . . . . . . . . . FA-12

Felletti, [email protected] Quantitativi, Facoltà di Economia, Università diMilano-Bicocca, Milano, Italy . . . . . . . . . . . . . . . . FB-12

Feng, [email protected] Engineering, Université Laval, Quebec, QC,Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-04

Fernandes Muritiba, Albert [email protected], Univeristy of Bologna, Bologna, Italy . . . MA-01,TD-01

Fernandez, [email protected] and Operations Research, Technical University ofCatalonia, Barcelona, Spain . . . . . . . . . . . RD-03, TA-09

89


Fernando, Babeth [email protected] Department, University of Santo Tomas, Manila,Philippines . . . . . . . . . . . . . . . . . . . . . . . . . FA-14, TA-15

Ferreira, [email protected] Mathematics, Universidade Federal do TriânguloMineiro, Uberaba, Minas Gerais, Brazil . . . . . . . . . FA-11

Ferreira, Fernanda [email protected], ESEIG - Instituto Politécnico do Porto, Vilado Conde, Portugal . . . . . . . . . . . . . . . . . . RA-09, RC-09

Ferreira, Flá[email protected], ESEIG - Instituto Politécnico do Porto, Vilado Conde, Portugal . . . . . . . . . . . . . . . . . . . . . . . . . RA-09

Ferreira, [email protected] & Computer Engineering, University of Camp-inas, Campinas, São Paulo, Brazil . . . . . . . . . . . . . . FA-02

Ferrer, [email protected] Industrial y de Sistemas, P. Universidad Catolicade Chile, Santiago, Chile . . . . . . . . . . . . . . . . . . . . . TA-12

Ferris, [email protected] Sciences Department, University of Wisconsin,Madison, WI, United States . . . . . . . . . . . . . . . . . . ME-17

Fiala, [email protected]. of Econometrics, University of Economics Prague,Prague 3, Czech Republic . . . . . . . . . . . . . . . . . . . . RA-11

Fick, [email protected] of Decision Sciences, University of SouthAfrica, Pretoria, South Africa . . . . . . . . . . . . . . . . . TD-02

Figueira, José [email protected] Superior Tecnico, Technical University of Lisbon,Lisbon, Portugal . . . . . . . . . . . . . . . . . . . . ME-08, TD-08

Fleischmann, [email protected]&SCM, Universität Augsburg, Augsburg, Ger-many . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-16

Fleszar, [email protected] School of Business, American University of Beirut,Beirut, Lebanon . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-01

Fletcher, [email protected] of Business Leadership, UNISA, East London,South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-11

Fletterman, [email protected] and Systems Engineering, University of Pretoria,Pretoria, South Africa . . . . . . . . . . . . . . . . . . . . . . . FB-16

Florens, [email protected] of Toulouse, Toulouse School of Economics,Toulouse, France . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-14

Fonseca i Casas, [email protected] University of Catalonia, Barcelona, Catalunya, SpainTA-05

Fontes, Fernando A. C. [email protected]. of Electrical and Computing Engineering, Universityof Porto, Faculty of Engineering, Porto, Portugal . FB-02,FB-13

Forbrig, [email protected] of Computer Science, University of Rostock,Rostock, Germany . . . . . . . . . . . . . . . . . . . . . . . . . . RE-18

Forradellas, [email protected] de Ingenieria, Universidad Nacional de Cuyo,Mendoza, Mendoza, Argentina . . . . . . . . . . . . . . . . TE-04

Fortz, [email protected]épartement d’Informatique, Université Libre de Brux-elles, Bruxelles, Belgium . . . . . . . . . . . . . . . . . . . . ME-18

Fouche, [email protected] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-01

Fourer, [email protected] Optimization Inc., Evanston, IL, United StatesMC-17, ME-17, TA-17

Fourie, [email protected] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-19

França, Paulo [email protected]. Matemática, Estatística e Computação, UniversidadeEstadual Paulista, Presidente Prudente, SP, Brazil . RD-03

Franquesa, [email protected] Politècnica de Catalunya, Barcelona, SpainRD-03

Fransoo, Jan [email protected] of Technology Management, Technische Uni-versiteit Eindhoven, Eindhoven, Netherlands . . . . RA-11

Freitas Rodrigues, [email protected]

90


CEFET/RJ, Rio De Janeiro, Brazil . . . . . . . . . . . . . RE-14

Friedler, [email protected] of Computer Science, University of Pannonia,Veszprem, Hungary . . . . . . . . . . . . . . . . . . . . . . . . . RD-04

Fröhling, [email protected] for Industrial Production (IIP), Karlsruhe Instituteof Technology (KIT), Karlsruhe, Germany . . . . . . RC-15

Froyland, [email protected] of Mathematics and Statistics, University of NewSouth Wales, Sydney, NSW, Australia . . . . . . . . . . RC-13

Fu, [email protected] Robert H. Smith School of Business and Institute forSystems Research, University of Maryland, College Park,MD, United States . . . . . . . . . . . . . . . . . . . . . . . . . MA-09

Fu, [email protected] of Traffic and Transportation Engineering, CentralSouth University, Changsha, Hunan, China . . . . . . ME-01

Fügenschuh, [email protected] Engineering, Helmut Schmidt University,Hamburg, Germany . . . . . . . . . . . . . . . . . . . . . . . . . TC-01

Fwa, Tien [email protected] University of Singapore, Singapore, SingaporeME-11

Galagedera, [email protected] and Business Statistics, Monash University,Caulfield East, Victoria, Australia . . . . . . . . . . . . . . FA-14

Galdeano, [email protected] Aplicada, Universidad de Almeria, Almeria,Almeria, Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-14

Gandibleux, [email protected] - Umr Cnrs 6241, The University of Nantes, Nantes,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-08

Gans, Jan [email protected] Computation, esp. CIM, Heinz Nixdorf Institute,Paderborn, Germany . . . . . . . . . . . . . . . . . . . . . . . . RE-04

García Quiles, [email protected] of Mathematics, University of Edinburgh, Edin-burgh, United Kingdom . . . . . . . . . . . . . . . . . . . . . . TA-09

Garcia Marquez, Fausto [email protected]ón de Empresas, Universidad de Castilla-La

Mancha, Ciudad Real, Spain . . . . . . . . . . . . . . . . . . TC-18

Garcia, Fernanda [email protected] de Ingenieria, Universidad Nacional de Cuyo,Mendoza, Mendoza, Argentina . . . . . . . . . . . . . . . . TE-04

Garcia, Viní[email protected] University of Santa Maria, Santa Maria, RS, BrazilRD-03

Gardes, [email protected] Rhône-Alpes, Saint-Ismier, France . . . . . . . TE-14

Gay, David [email protected] Optimization LLC, Albuquerque, NM, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-17, TA-17

Geipele, [email protected] Technical University, Riga, Latvia . . TA-07, RC-15

Geleijnse, [email protected] Interaction (MI) group, Philips Research, Eind-hoven, Netherlands . . . . . . . . . . . . . . . . . . . . . . . . . MD-18

Gemayqzel, [email protected] University, Havana, Cuba . . . . . . . . . . . . . MA-03

Gendreau, [email protected] and CIRRELT, École Polytechnique, Montreal,Quebec, Canada . . . . . . . . . . . . . . . . . . . MC-01, MD-01

Genoese, [email protected] for Industrial Production, University of Karlsruhe,Karlsruhe, Germany . . . . . . . . . . . . . . . . . . . . . . . . . FA-12

Germain, [email protected] Business School, Toulouse, France . . . . . TD-14

Gevers, [email protected] School, University of Stellenbosch, Bellville,South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-04

Ghanmi, [email protected], DND, Ottawa, Ontario, Canada . . . . . . . . RE-03

Gharote, [email protected] School of Management, Indian Institute of Technol-ogy, Bombay, Mumbai, Maharashtra, India . . . . . . TD-11

Ghosh, [email protected] Administration, Swami Vivekananda Institute OfManagement & Computer Science, Kolkata, West Bengal,

91


India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-14

Giesen, [email protected] Engineering and Logistics, Pontificia Universi-dad Catolica, Santiago, RM, Italy . . . . . . . FB-16, TD-16

Gilibert, [email protected] Oberta de Catalunya, Barcelona, Spain TD-09

Girard, [email protected] Rhône-Alpes, projet Mistis, Saint-Ismier, FranceTE-14

Girardeau, [email protected], Edf R&d, Clamart, France . . . . . . . . . . . . FA-01

Glazebrook, [email protected] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-19

Golany, [email protected] Engineering & Management, Technion - IsraelInstitute of Technology, Haifa, Israel . . . . . . . . . . . TA-10

Golden, [email protected] & Information Technologies, University of Mary-land, College Park, MD, United States . . . . . . . . . . RD-03

Golenko-Ginzburg, [email protected] Engineering & Management, Academic Collegeof Judea and Samaria, Ariel, Israel . . . . . . . . . . . . . TA-10

Gomes, A. [email protected] TEC, Faculdade de Engenharia, Universidade doPorto, Porto, Portugal . . . . . . . . . . . . . . . . FA-18, RC-18

Gomez Gonzalez, [email protected] e Investigacion Operativa III, Complutense Uni-versity, Madrid, Spain . . . . . . . . . . . . . . . . FA-07, RD-09

Gomez Padilla, [email protected] Engineering, University of Guadalajara,Guadalajara, Jalisco, Mexico . . . . . . . . . . . . . . . . . RC-12

Gonçalves, José [email protected], INESC TEC, Faculdade de Economia do Porto,Universidade do Porto, Porto, Portugal . . . . . . . . . . FA-18

Gonen, [email protected] of Technology, Holon Institute of Technology- HIT, Holon, Israel . . . . . . . . . . . . . . . . . . . . . . . . . RE-03

González-Arangüena, [email protected]

Statistics and Operation Research III, Complutense Uni-versity of Madrid, Madrid, Madrid, Spain . . . . . . . RD-09

González-Martín, [email protected] de La Laguna, La Laguna. Tenerife, SpainME-01, RD-13

Gouveia, Luí[email protected] - Departamento de Estatística e Investigação Opera-cional, Universidade de Lisboa - Faculdade de Ciências,Lisboa, Portugal . . . . . . . . . . . . . . . . . . . . ME-18, TA-18

Gouws, [email protected] of Chemical Engineering, University of Preto-ria, Pretoria, South Africa . . . . . . . . . . . . . . . . . . . . RC-04

Greben, [email protected] Environment, CSIR, Pretoria, Gauteng, South AfricaTD-03

Griffiths, [email protected], Cardiff University, CARDIFF, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-19

Grobler, [email protected] of Industrial and Systems Engineering, Uni-versity of Pretoria, Pretoria, South Africa . . . . . . . RE-04

Groth, Julie [email protected] planning, DSB S-tog a/s, Copenhagen V, Den-mark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-03

Guddat, [email protected] Berlin, Berlin, Germany . . . . . . . . . . . . . . . . . MC-15

Guenatri, Mohamed [email protected], UPC, Barcelona, Catalonia, Spain . . . . . . . . . TA-14

Guerrin, [email protected] Recyclage & Risque, Inra & Cirad, MontpellierCedex 5, France . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-04

Guibetlafaye, [email protected], Paris, France . . . . . . . . . . . . . . . . . . . . FA-05

Günther, [email protected] of Business Administration and Economics,Bielefeld University, Bielefeld, Germany TC-05, MD-09

Gutierrez, Jose [email protected]ística, Investigación Operativa y Computación, Uni-versidad de La Laguna, La Laguna, Tenerife, Spain TE-09

92


Hakkert, [email protected] and Environmental Engineering, Technion, Israel,Haifa, Israel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-03

Haley, K. [email protected] & Mechanical Engineering, University ofBirmingham, Birmingham, United Kingdom . . . . RA-10

Hallamäki, [email protected] Information Technology, University ofJyväskylä, Jyväskylä, Finland . . . . . . . . . . TE-09, TE-16

HAMAM, [email protected] - Tut, Pretoria, South Africa . . . . . MA-02, TE-18

Hansen, [email protected] and HEC Montreal, Montreal, Quebec, CanadaMA-01

Haouari, [email protected] Polytechnique de Tunisie, Tunis . MD-01, ME-07

Harmse, [email protected] Resources Development & Organisational Devel-opment, Sasol Technology, Secunda, South Africa RE-05

Hasson, [email protected] Sciences, IRAQ / Faculty of sciences / Univer-sity of Babylon, Babylon, Babylon, Iraq . . . . . . . . RC-19

Hattingh, [email protected] Science & Information Systems, North-WestUniversity, Potchefstroom, South Africa . . . . . . . . ME-18

Hattingh, [email protected] of Computer, Statistical and Mathematical Sci-ences, North-West University, Potchefstroom, South AfricaFA-02, RA-02, RE-02, RD-18, RE-18, TE-18

Haugland, [email protected], Norway . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-02

Hayez, [email protected] - Faculty of Engineering, ULB - Brussels Free Uni-versity, Brussels, Belgium . . . . . . . . . . . . . . . . . . . . RA-08

Hämäläinen, Raimo [email protected] Analysis Laboratory, Aalto University, School ofScience, Aalto, Finland . . . . . . . . . . . . . . . . . . . . . . TE-08

Heavey, [email protected] of Limerick, Limerick, Ireland . . . . . . . TC-11

Hederstierna, [email protected] of Management, Blekinge Institute of Technology,Ronneby, Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . RC-08

Hegyhati, [email protected] of Pannonia, Veszprem, Hungary . . . . . RD-04

Helberg, [email protected] of Electrical, Electronic and Computer Engineer-ing, North-West University, Potchefstroom, South AfricaTE-18

Hemmelmayr, [email protected] University of Economics and Business (WU), Vi-enna, Austria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-09

Hendriks, Maarten [email protected] of Mechanical Engineering, Eindhoven Uni-versity of Technology, Eindhoven, Netherlands . . MD-11

Hennig, [email protected] of Industrial Economics and Technology Man-agement, Norwegian University of Science and Technol-ogy, Trondheim, Norway . . . . . . . . . . . . . . . . . . . . RD-16

Hernandez, [email protected] and Sales, Megaworld Corporation, Makati City,Philippines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-14

Herroelen, [email protected] Sciences and Information Management,Katholieke Universiteit Leuven, Leuven, Belgium TA-10

Hertz, [email protected] Polytechnique, Montreal, Canada . . . . . . . . MC-01

Hinojosa, [email protected]ía Aplicada I, Universidad de Sevilla, Sevilla,Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-09

Hirotsu, [email protected] University, Inzai, Chiba, Japan . . . . . . . . TA-04

Hirsch, [email protected] of Production and Logistics, University of NaturalResources and Life Sciences, Vienna, Wien, Austria . RD-07

Hochbaum, [email protected]&OR department, UC Berkeley, Berkeley, CA, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-01

93


Hodge, [email protected], University of Nottingham, Nottingham, Not-tinghamshire, United Kingdom . . . . . . . . . . . . . . . RA-19

Holczinger, [email protected] of Computer Science, University of Pannonia,Veszprem, Hungary . . . . . . . . . . . . . . . . . . . . . . . . . RD-04

Holmes, [email protected], University of Sheffield, Sheffield, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-07

Hotokka, [email protected] of Mathematical Information Technology, Uni-versity of Jyväskylä, Jyväskylä, Finland . . TE-09, TE-16

Huang, [email protected] of Engineering, University of Technology, Sydney,Broadway, NSW, Australia . . . . . . . . . . . . . . . . . . MD-11

Hujala, [email protected] Unit, Finnish Forest Research Institute, Vantaa, Fin-land . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-04, RE-09

Hunter, [email protected], Oceana Group, Stompneusbaai, Western Cape,South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-16

Hurink, [email protected] of Applied Mathematics, University of Twente,Enschede, Netherlands . . . . . . . . . . . . . . . . . . . . . . TC-10

Imai, [email protected] School of Maritime Sciences, Kobe University,Kobe, Japan . . . . . . . . . . . . . . . . MC-11, ME-11, TA-16

Immelman, [email protected] Sciences, University of South Africa, pretoria,Gautemg, South Africa . . . . . . . . . . . . . . . TA-05, RD-06

INGOLOTTI, [email protected] InformÁticos Y ComputaciÓn, UniversidadPolitÉncia De Valencia, VALENCIA, Spain . . . . . ME-03

Inhetvin, [email protected] Science, Universidade Presbiteriana Mackenzie,São Paulo, SP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-02

Iori, [email protected], University of Modena and Reggio Emilia, ReggioEmilia, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-01

Ishii, Hiroaki

[email protected] University, Japan . . . . . . . . . . . . . . . . . . . . . RE-06

Ittmann, Hans [email protected] Consulting, The Willows, South Africa . . . . . FA-05

Jablonsky, [email protected]. of Econometrics, University of Economics Prague,Prague 3, Czech Republic . . . . . . . . . . . . RD-05, MD-17

Jagla, [email protected] Software GmbH, Cologne, Germany . . . . ME-17,TA-17

jalila, [email protected], Laboratoire d’informatique de Paris Nord, Villeta-neuse, France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-03

James, [email protected], Louisiana Tech University, Ruston, LouisianaTA-04

Jankowitz, [email protected] Sciences, UNISA, Pretoria, South Africa FA-01

Joubert, [email protected] and systems Engineering, University of Pretoria,PRETORIA, South Africa . . . . ME-01, RA-11, FB-16,TD-16

Juan, Angel [email protected] Science, Fundació per a la Universitat Oberta deCatalunya, Barcelona, Spain . . . . . . . . . . . TA-05, TD-09

Jurado, [email protected]ón Industrial y Gestión de Empresas, Universi-dad de Sevilla, Sevilla, Spain . . . . . . . . . . . . . . . . . TE-09

Justesen, [email protected](DORS), Valby, Denmark . . . . . . . . . . . . . . . . . . . . TE-07

Kabongo Kantu, [email protected] Sciences, University of Cape Town, Cape Town,South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-14

Kah, Jainaba [email protected] of Business & Entrepreneurship, American Univer-sity of Nigeria, YOLA, Adamawa State . . RD-05, RE-06

Kah, Muhammadou [email protected] of Information Technology & Communications,American University of Nigeria, YOLA, Adamawa State,Nigeria . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-05, RE-06

94


Kana Zeumo, [email protected], University Paris Dauphine, Paris, Paris, FranceTA-08

Kangas, [email protected] of Forest Resources Management, Universityof Helsinki, Helsinki, Finland . . . . . . . . . . TC-04, RE-09

Kangas, [email protected] Forest Research Institute, Helsinki, Finland FA-17

Kangaspunta, [email protected] of Mathematics and Systems Analysis, AaltoUniversity School of Science, Aalto, Finland . . . . MA-08

Kanková, [email protected], Institute of Information Theory and Au-tomation of ASCR, Prague, Czech Republic, Czech Re-public . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-02

Kaparias, [email protected] for Transport Studies, Imperial College London,London, United Kingdom . . . . . . . . . . . . . . . . . . . . TE-09

KAPLAN, [email protected] Engineering, Koc University, ISTANBUL,Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-10

Kaya, Gö[email protected] Engineering, Halic University, Istanbul, TurkeyFB-07

Kayakutlu, [email protected], Turkey . . . . . . . . . . . . . . . . . . . . . TA-15, FA-17

Kazemi Matin, [email protected] of Applied Mathematics, Islamic Azad Uni-versity, Karaj Branch, Tehran, Iran, Islamic Republic OfMC-15

Kedad-Sidhoum, [email protected] - Upmc, Paris, France . . . . . . . . . . . . . . . . . . . FA-11

Keeney, [email protected] School of Business, Duke University, San Francisco,CA, United States . . . . . . . . . . . . . . . . . . . . . . . . . . MC-08

Kempkes, Jens [email protected] & OR Lab, University of Paderborn, Germany, Pader-born, Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-13

Kiesling, Elmar

[email protected] of Business Administration, University of Vi-enna, Vienna, Austria . . . . . . . . . . . . . . . . . . . . . . . TC-05

Kim, [email protected] School of Management, KAIST, Seoul, Korea,Republic Of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-09

Kim, [email protected] Department, Santa Clara University, Santa Clara,CA, United States . . . . . . . . . . . . . . . . . . . . . . . . . . RA-12

Kim, [email protected] of Business, Yonsei University, Seoul, Korea, Re-public Of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-07

Kinoshita, [email protected] Science Department, Meijo University, Kani, Gifu,Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-14

Kirchner, [email protected] Studies, Berlin School of Economics and Law,Berlin, Germany . . . . . . . . . . . . . . . . . . . . FA-06, RC-08

Kirkbride, [email protected] Management School, Lancaster University, Lancaster,Lancashire, United Kingdom . . ME-19, RA-19, RC-19

Kis, [email protected] Methods in Economy, Faculty of Economics,Subotica, Serbia . . . . . . . . . . . . . . . . . . . . MC-02, TD-11

Klemes, [email protected], University of Pannonia, Veszprém, HungaryRC-04, RD-04

Kleywegt, [email protected] of Industrial and Systems Engineering, Georgia In-stitute of Technology, Atlanta, GA, United States . RC-09

Koberstein, [email protected] Administration, European University Viadrina,Frankfurt (Oder), Germany . . . . . . . . . . . . . . . . . . . TE-10

Koen, [email protected] and Quantitative Methods, CSIR, Pretoria, SouthAfrica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-03

Kok, [email protected], University of Stellenbosch, Stellenbosch, West-ern Province, South Africa . . . . . . . . . . . . . . . . . . . FA-03

Kolisch, [email protected]

95


TUM School of Management, Technische UniversitaetMuenchen, Muenchen, Germany . . . . . . . . . . . . . . TC-18

Kopa, [email protected] of Probability and Mathematical Statistics,Charles University in Prague, Faculty of Mathematics andPhysics, Prague, Czech Republic . . . . . . . . . . . . . . ME-12

Korach, [email protected] Engineering and Management, Ben-Gurion Uni-versity, Beer-Sheva, Israel . . . . . . . . . . . . . . . . . . . . TC-01

Kornbluth, Jonathan [email protected] of Business Administration, Hebrew University ofJerusalem, Jerusalem, Israel . . . . . . . . . . . . . . . . . . TA-03

Korst, [email protected] Interaction (MI) group, Philips Research, Eind-hoven, Netherlands . . . . . . . . . . . . . . . . . . . . . . . . . MD-18

Kotze, [email protected] Profitability Improvement, Sasol Technology,Johannesburg, Gauteng, South Africa . . . . . . . . . . MD-09

Kristjansson, [email protected] Software, Ltd., London, United Kingdom . MC-17, ME-17, TC-17

Kropat, [email protected] of Computer Science, Universität der Bun-deswehr München, Neubiberg, Germany . . . . . . . . FB-10

Kruger, [email protected] of Science and Technology, Federation UniversityAustralia, Ballarat, Victoria, Australia . . . . . . . . . . RD-02

Kruger, [email protected] of Computer, Statistical and Mathematical Sci-ences, North-West University, Potchefstroom, South AfricaFA-02, RA-02

Kruger, [email protected] for BMI, North-West University, Potchefstroom,South Africa . . . . . . . . . . . . . . . . . . . . . . . TC-13, MC-17

Kubiak, [email protected] of Business Administration, Memorial University,St. John’s, NL, Canada . . . . . . . . . . . . . . . . . . . . . MA-01

Kumar, [email protected], Dr. B.R. Ambedkar University, Agra, Agra,UP, India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-07

Kurttila, Mikko

[email protected] Research Unit, Finnish Forest Research Institute,Joensuu, Finland . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-09

kwak, N. [email protected] Sciences & MIS, St. Louis University, St. Louis,MO, United States . . . . . . . . . . . . . . . . . . . . . . . . . . TA-11

Lackes, [email protected] of Business Information Management, Univer-sity of Dortmund, Dortmund, Germany . . . . . . . . . RC-16

Lahdelma, [email protected] of Energy Technology, Aalto University, Es-poo, Finland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-08

Lam, Hon [email protected], University of Pannonia, Veszprem, Veszprem,Hungary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-04

Lam, [email protected] University of Hong Kong, Kowloon, HongKong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-03

Lam, [email protected] Polytechnic University of Hong Kong, Hong Kong,Hong Kong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-03

Lamond, [email protected] & Decision Systems, Universite Laval, Quebec,Quebec, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-17

Landete, [email protected] de Estadística y Matemática Aplicada, Uni-versity Miguel Hernández of Elche, Elche, Alicante, SpainTE-10

Lange, [email protected] of Law and Economics, Technische Univer-sität Darmstadt, Darmstadt, Germany . . . . . . . . . . . RE-16

Langley, [email protected] . . . . . . . . . . . . . . . . . . . . . . . . . . ME-02, FB-14

Larsen, [email protected] of Management Engineering, Technical Uni-versity of Denmark, Kgs. Lyngby, Denmark . . . . MD-03,TE-07, RC-13

Larson, [email protected], Boston, United States . . . . . . . . . . . . . . . . . . . RC-07

Laturi, [email protected]

96


Finnish Forest Research Institute Metla, Helsinki, FinlandRA-04

Lau, [email protected] of Engineering, University of Technology, Sydney,Broadway, NSW, Australia . . . . . . . . . . . . . . . . . . MD-11

Lau, Hoong [email protected] of Information Systems, Singapore ManagementUniversity, Singapore, Singapore . . . . . . . . . . . . . MD-16

Létocart, [email protected] Umr Cnrs 7030, Institut Galilée - Université Paris 13,Villetaneuse, France . . . . . . . . . . . . . . . . . . . . . . . . . TA-02

le Roux, [email protected] Sciences, University of South Africa, Pretoria,Gauteng, South Africa . . . . . . . . . . . . . . . . . . . . . . RD-07

Lebkowski, [email protected] of Management, AGH University of Science andTechnology, Krakow, Micronesia, Federated States Of RE-11

Lee, Chang [email protected] School of Health Management and Policy, TheCatholic University of Korea, Seoul, Korea, Republic OfTA-11

Lee, [email protected] and Statistics, University of South Australia,Mawson Lakes, South Australia, Australia . . . . . MC-11,MD-11

Leiva, [email protected] Engineering and Logistics, Pontificia Universi-dad Catolica, Santiago, RM, Chile . . . . . . . . . . . . . FB-16

Leon, [email protected]ía Industrial, P. Universidad Católica de Chile,Santiago, Chile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-12

Leskinen, [email protected], Finland . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-04

Leskinen, [email protected] Programme for Production and Consumption,Finnish Environment Institute, Joensuu, Finland . TC-04,RE-09

Leus, [email protected] Sciences and Information Management, KULeu-ven - FBE, Leuven, Belgium . . . . . . . . . . . TA-02, TA-10

Li, [email protected] of Logistics & Maritime Studies, The HongKong Polytechnic University, Hong Kong . . . . . . . ME-01

Li, [email protected] Division, Institute of Policy and Management,CAS, Beijing, China . . . . . . . . . . . . . . . . . . . . . . . . TC-08

Li, [email protected] of Management, Fudan University&The Polytech-nic University of Hong Kong, Kowloon, Hong Kong RC-03

Liao, [email protected] School of Business, University of California - Berke-ley, Berkeley, CA, United States . . . . . . . . . . . . . . . TC-10

Lidouh, [email protected] and Decision Engineering (CoDE), UniversitéLibre de Bruxelles, Brussels, Belgium . . . . . . . . . . RA-08

Liesiö, [email protected] of Information and Service Economy, AaltoUniversity, Helsinki, Finland . . . . . . . . . . . . . . . . . MA-08

Lim, Yun [email protected] Kong Chian School of Business, Singapore Manage-ment University, Singapore, Singapore . . . . . . . . . MC-16

Lin, Ka Yuk [email protected]. of Management Sciences, City University of HongKong, Hong Kong, Hong Kong . . . . . . . . . . . . . . . MD-01

Lin, [email protected] of Computing & Technology, Cameron Uni-versity, Lawton, Oklahoma, United States . . . . . . MD-12

Lin, [email protected] of Technology Management, National Universityof Tainan, Tainan, Taiwan . . . . . . . . . . . . . . . . . . . . FA-07

Lintunen, [email protected] Forest Research Institute Metla, Helsinki, FinlandFA-17

Liston, [email protected] of Limerick, Limerick, Ireland . . . . . . . TC-11

Liu, [email protected] of Engineering, University of Technology, Sydney,Broadway, NSW, Australia . . . . . . . . . . . . . . . . . . MD-11

Liu, Jiyin

97


[email protected] of Business and Economics, Loughborough Uni-versity, Loughborough, Leicestershire, United KingdomMA-11

Liu, [email protected] Business School, canterbury, United Kingdom . RE-14

Lloyd-Jones, [email protected], University of Sheffield, Sheffield, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-07

Lo, Hong [email protected] Engineering, Hong Kong University of Science andTechnology, Hong Kong, Hong Kong . . . . RC-03, FA-16

Loiseau, [email protected] de Computación-, Facultad de Ciencias Ex-actas y Naturales, Universidad de Buenos Aires, BuenosAires, Argentina . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-02

Loo, Becky [email protected], The University of Hong Kong, Pokfulam,Hong Kong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-03

Lourens, [email protected]??, Johannesburg, South Africa . . . . . . . . . . . . . . . RD-17

LOVA, [email protected] de Estadística e Investigación Operativa,Universidad Politécnica de Valencia, Spain . . . . . ME-03

Love, [email protected] Graduate School of Business, Simon Fraser Univer-sity, Vancouver, British Columbia, Canada . . . . . . TA-19

Lozano, Antonio [email protected] of Mathematics, University of Huelva, Huelva,Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-03

Lusby, [email protected] of Management Engineering, Technical Uni-versity of Denmark, Kgs Lyngby, Denmark . . . . . MD-03

Lüthi, [email protected], ETHZ, Zürich, Switzerland . . . . . . . . . . FA-12

Lutu, [email protected], University of Pretoria, Pretoria, Gauteng,South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-06

Lysgaard, [email protected]

Department of Economics and Business, Aarhus Univer-sity, Aarhus V, Denmark . . . . . . . . . . . . . . . . . . . . . TC-02

Ma, [email protected] of Information Management, National UnitedUniversity, Miaoli, Taiwan, Taiwan . . . . . . . . . . . . TC-13

Ma, [email protected] Engineering, Hong Kong University of Science andTechnology, Hong Kong, Hong Kong . . . . . . . . . . RC-03

Mac Cawley, [email protected] and System Engineering, Georgia Institute ofTechnology, Atlanta, Georgia, United States . . . . . ME-04

Machado Casali, [email protected] de Engenharia de Producao e Sistemas, Uni-versidade Federal de Santa Catarina, Florianopolis, SantaCatarina, Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-18

Macharis, [email protected], Vrije Universiteit Brussel, Brussels, Bel-gium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-15

Maddulapalli, Anil [email protected] Science Lab, GMTCI, General Motors R&D, Banga-lore, India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-08

Maden, [email protected] and Logistics, University of Huddersfield, Hud-dersfield, United Kingdom . . . . . . . . . . . . . . . . . . . ME-01

Mahdi, [email protected] CAILLAU, Issy Les Moulineaux, France . . . . RA-16

Majozi, [email protected] of Pretoria, Pretoria, South Africa . . . . RC-04,RD-04

Makui, [email protected] University of Science and Technology, Tehran, Iran,Islamic Republic Of . . . . . . . . . . . . . . . . . . . . . . . . MD-14

Malaguti, [email protected], University of Bologna, Bologna, Italy . . . . . . TD-01

Malherbe, [email protected] Technology, Secunda, Mpumalanga, South AfricaRE-11

Mansour, Farah [email protected] Optimization Research Group - ROI, EcolePolytechnique de Tunisie, La Marsa, Tunisia . . . . ME-07

98


Manuel, Conrado [email protected] and Operation Research III, Complutense Uni-versity of Madrid, Madrid, Spain, Spain . . . . . . . . RD-09

Maquera, [email protected] de Ingeniería y Arquitectura, Universidad PeruanaUnión, Juliaca, Puno, Peru . . . . . . . . . . . . . . . . . . . TD-09

Marais, [email protected] Meraka Institute, CSIR, Pretoria, South Africa . . RE-15, TD-15

Marín, [email protected] de Estadística e Investigación Operativa,University of Murcia, Murcia, Spain . . . . . . . . . . . TA-09

Marc, Almiñ[email protected]ística y Matemática Aplicada, Universidad MiguelHernández de Elche, Elche, Alicante, Spain . . . . . . TE-10

Marchetta, [email protected] de Ingenieria, Universidad Nacional de Cuyo,Mendoza, Mendoza, Argentina . . . . . . . . . . . . . . . . TE-04

Mareschal, [email protected] Brussels School of Economics and Management,Université Libre de Bruxelles, Brussels, Belgium . RA-08

Marques, [email protected] of Civil Engineering and Architecture,CESUR-IST, Technical Univeristy of Lisbon, Lisbon, Por-tugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-14, RE-15

Marquez, [email protected] Informatics, CSIRO, Clayton, Vic, Aus-tralia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-11, RE-11

Martell, [email protected] of Forestry, University of Toronto, Toronto, On-tario, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-04

Martin, [email protected], University of Chicago, Chicago, Illinois, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-17

Martin-Herran, [email protected], Universidad de Valladolid, Valladolid, Spain RD-09

Masinga, [email protected] and Applied Mathematics, University of theWitwatersrand Johannesburg, Johannesburg, South Africa

RD-05

Masterson, [email protected], Cape Town, Western Cape, South Africa MC-13

Mateo, [email protected] Business Administration, Universitat Politec-nica Catalunya, Barcelona, Spain . . . . . . . . . . . . . . RE-11

Maturana, [email protected] Industrial y de Sistemas, P. Universidad Catolicade Chile, Santiago, Chile . . . . . . ME-04, TD-04, TE-04

Mayerle, Sérgio [email protected] de Produção e Sistemas, Universidade Federalde Santa Catarina, Florianópolis, SC, Brazil . . . . . FA-18

Mazri, [email protected], INERIS, Verneuil - en - Halatte, France . . . TA-08

Mbohwa, [email protected] and Operations Management, University of Johan-nesburg, Johannesburg, Gauteng, South Africa . . . RD-17

Medvedeva, [email protected] Federal University, Ekaterinburg, Russian FederationTC-12

Meeraus, [email protected] Development Corporation, Washington, DC,United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-17

Meisel, [email protected], Kiel, Germany . MA-11,MD-11

Melachrinoudis, [email protected] and Industrial Engineering, Northeastern Uni-versity, Boston, MA, United States . . . . . . . . . . . . RA-16

Merchant, [email protected] Link Consulting, Rickmansworth, Hertfordshire,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-05

Mesa, Juan [email protected] of Seville, Sevilla, Spain . . . . . . . . . . . MA-03

Meyer, [email protected], Pretoria, South Africa . . . . . . . . . . . . . . . . TD-15

Meyer, [email protected], Sasol Technology, Sasolburg, Freestate . ME-09

99


Miettinen, [email protected]. of Mathematical Information Technology, Universityof Jyväskyla, University of Jyväskyla, Finland . . . FB-09

Milioni, Armando [email protected] de Organizacao, Instituto Tecnologico deAeronautica, Sao Jose dos Campos, Sao Paulo, Brazil TC-05, RE-14

Mishina, [email protected] Science and Technology, Akita Perfectural Uni-veristy, Akita, Japan . . . . . . . . . . . . . . . . . . . . . . . . RC-12

Mishra, [email protected] of Computer Science and IT, University of Notting-ham, Nottingham . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-07

Mitchell, [email protected] Science Dept., University of Auckland, Auck-land, New Zealand, New Zealand . . . . . . . . . . . . . . TC-04

Mittal, Ashok [email protected], IIT Kanpur, Kanpur, India . . . . . . . . . . . . . . MC-05

Mohr, [email protected] Management, University of Mannheim,Mannheim, Germany . . . . . . . . . . . . . . . . . . . . . . . RA-12

Monge, Juan [email protected] de Investigación Operativa, Universidad MiguelHernández, Elche, Alicante, Spain . . . . . . . . . . . . . TE-10

Monteiro, Marta S. [email protected] de Economia and LIAAD-INESC Porto L.A.,Universidade do Porto, Porto, Portugal . . . . . . . . . . FB-02

Montero, [email protected] of Mathematics, Complutense University, Madrid,Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-07

Montero, Lí[email protected] and Operations Research Department, UPCBarcelonaTech, Barcelona, Catalonia, Spain . . . . . FA-16

Montibeller, [email protected]. of Management, London School of Economics, Lon-don, United Kingdom . . . . . . . . . . . . . . . . . . . . . . . MC-08

Moolman, [email protected] Development, Traffic Management Technologies,Capetown, Western Cape, South Africa . . . . . . . . . TE-03

Morabito, [email protected]. of Production Engineering, Federal University ofSão Carlos, Sao Carlos, Sao Paulo, Brazil TE-01, FA-11

Moreira, M. [email protected] de Economia, Universidade do Porto, Porto,Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-13

Morelato, André [email protected], Universidade Estadual de Campinas, Campinas, SP,Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-03

Morison, [email protected] and Profitability Improvement, Sasol Technol-ogy (Pty) Ltd, Secunda, Mpumalanga . . . . . . . . . . RE-16

Moskvina, [email protected] Medicine, School of Medicine, Cardiff Uni-versity, Cardiff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-10

Mossberg, [email protected]örsvarsanalys, FOI, Stockholm, Sweden . . . . . . . MC-05

Möst, [email protected] of Energy Economics, Technische Universität Dres-den, Dresden, Germany . . . . . . . . . . . . . . . . . . . . . . FA-12

Moura, [email protected], University of Lisbon, Lisbon, Portugal . . . . TA-18

Movahedi, Mohammad [email protected] Department, Islamic Azad University,Firoozkooh Branch, Firoozkooh Iran., Tehran, Tehran, Iran,Islamic Republic Of . . . . . . . . . . . . . . . . . . . . . . . . RA-14

Muñoz, David [email protected] & Operations Engineering, Instituto TecnologicoAutonomo de Mexico, Mexico City, Mexico . . . . MC-07

Muñoz, Juan [email protected] Engineering and Logistics, Pontificia Universi-dad Catolica, Santiago, RM, Chile . . . . . . FB-16, TD-16

Muñoz, [email protected] de Estadística e Investigación Operativa,Universidad Complutense de Madrid, Madrid, Spain . FA-07

Mues, [email protected] School, University of Southampton,Southampton, United Kingdom . . . . . . . . . . . . . . . MD-12

Mukeru, S

100


[email protected] Sciences, UNISA, Pretoria, South Africa FB-01

Müller-Merbach, [email protected], Universität Kaiserslautern,Kaiserslautern, Germany . . . . . . . . . . . . . . . . . . . . . RA-10

Munoz, David [email protected] & Operations Engineering, Instituto Techno-logico Autonomo de Mexico, Mexico City, Mexico . MC-07

Murthy, Rajluxmi [email protected] Methods and Information Systems, IndianInstitute of Management, Bangalore, Karnataka, India TE-03

Mustajoki, [email protected] Analysis Laboratory, Helsinki University of Tech-nology, HUT, Finland . . . . . . . . . . . . . . . . . . . . . . . TE-08

Myers, [email protected] of Forestry, University of Toronto, Toronto, On-tario, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-04

Nagaraj, [email protected] Science, CMR Institute of Technology, Banga-lore, Karnataka, India . . . . . . . . . . . . . . . . . . . . . . . . FA-16

Nagih, [email protected] - Ufr Mim, Universite Paul Verlaine, Metz Cedex 1,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-02, MC-03

Nagy, Gá[email protected] Business School, University of Kent, Canterbury,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-16

Naidoo, [email protected] research Institute, Pretoria, South Africa . . RD-06

Nailana, Freddy [email protected] of Computer Science, University of Limpopo -Turfloop Campus, Polokwane, South Africa . . . . . MC-19

Nakai, [email protected] of Education, Chiba University, Chiba, Japan MA-02

Nalchigar, [email protected] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-14

Nalpas, [email protected] & Finance, Toulouse Business School,

Toulouse, France . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-14

Narasimhan, [email protected] & Quantitative Methods Area, Indian Instituteof Management, Ahmedabad, Ahmedabad, Gujarat, IndiaTD-15

NARCISO PEREIRA, [email protected] Naval And Oceanic Engineering, UniversityOf SÃo Paulo, SÃo Paulo, SÃo Paulo, Afghanistan TD-16

Nascimento, [email protected]ática, Universidade Federal de Goiás, Goiânia,Goiás, Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-18

Nascimento, [email protected] tecnológico de Aeronáutica, São José dos Cam-pos, São Paulo, Brazil . . . . . . . . . . . . . . . . . . . . . . . TC-05

Natarajan, [email protected] University of Technology and Design, Singa-pore, Singapore . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-01

Núñez, [email protected] Engineering Department, Universidad de Chile,Santiago, Chile . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-01

NCUBE, [email protected] And Rolling Stock, National Railways Of Zim-babwe, KWEKWE, MIDLANDS . . . . . . . . . . . . . RD-17

Ncube, [email protected] School of Business Leadership, University ofSouth Africa, Pretoria, Gauteng, South Africa . . . . RC-11

Ndiaye, Malick [email protected] Engineering, King Fahd University of PetroleumAnd Minerals, Dhahran, Saudi Arabia . . . . . . . . . . FA-08

Negrotto, [email protected]ón, Universidad de Buenos Aires - Facultad deCiencias Exactas y Naturales, Buenos AIres, AfghanistanTC-02

Neiva de Figueiredo, Joã[email protected] of Management, Saint Joseph’s University,Haub School of Business, Philadelphia, PA, United StatesTA-15, FA-18

Nel, [email protected] of Logistics, Stellenbosch University, Stellen-bosch, Western Province, South Africa . . . . . . . . . . FA-16

Nemery, Philippe

101


[email protected] of Mathematics, University of Portsmouth,Portsmouth, United Kingdom . . . . . . . . . ME-08, RA-08

Newman, [email protected] of Economics and Business, Colorado School ofMines, Golden, CO, United States . . . . . . . . . . . . . MC-04

Ng, Ada Suk [email protected] of Transport and Logistics Studies, University ofSydney, University of Sydney, NSW, Australia . . . RE-13

Ng, Szu [email protected], Singapore, Singapore . . . . . . . . . . . . . . . . . . MC-11

Nieuwoudt, [email protected] of Stellenbosch, Stellenbosch, South AfricaFB-13

Nishimura, [email protected] School of Maritime Sciences, Kobe University,Kobe, Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-16

Nogueira, Christiane [email protected]ção, Sociedade Educacional de Santa Catarina,Florianópolis, Santa Catarina, Brazil . . . . . . . . . . . ME-14

Novaes, Antonio [email protected] Engineering, Federal University of Santa Cata-rina, Florianopolis, SC, Brazil . . . . . . . . . . . . . . . . . TA-14

Nsakanda, Aaron [email protected] School of Business, Carleton University, Ottawa,Ontario, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-18

Numminen, [email protected] of Management, Blekinge Institute of Technology,Ronneby, Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . RC-08

Nyamugure, [email protected], National University of Science and Technol-ogy, Bulawayo, Zimbabwe, Zimbabwe . . . . . . . . . TD-19

Nygreen, Bjø[email protected] of Industrial Economics and Technology Man-agement, Norwegian University of Science and Technol-ogy, Trondheim, Norway . . . . . . . . . . . . . . . . . . . . RD-16

Odegaard, [email protected] Business School, Western University, London, On-tario, Canada . . . . . . . . . TC-05, TD-05, TE-05, TA-13

Ogryczak, [email protected]

Institute of Control & Computation Engineering, WarsawUniversity of Technology, Warsaw, Poland . . . . . . ME-18

Okesola, [email protected] Sciences:Operations Research, University ofSouth Africa, Pretoria, South Africa, South Africa FB-17

Olivares, [email protected] and Operational Research, Rey Juan Carlos Uni-versity, Móstoles, Madrid, Spain . . . . . . . . . . . . . . . FA-02

Oliveira, [email protected] & Applied Mathematics, State Universityof Campinas, Campinas, SP, Brazil . . . . MC-02, ME-02

Oliveira, José [email protected] TEC, Faculty of Engineering, University of Porto,Porto, Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-18

Oliveira, [email protected] of Mathematics, ISEP, Porto, Portugal FA-18

Oliveira, Rú[email protected] de Matemática, Computação e Cognição, Univer-sidade Federal do ABC, Santo André, São Paulo, BrazilFA-02

Orlovich, [email protected] Mathematics and Algorithmics, Belarus State Uni-versity, Minsk, Belarus . . . . . . . . . . . . . . . . . . . . . . MC-18

Oron, [email protected] and Business Statistics, The University ofSydney, Sydney, New South Wales, Australia . . . . RE-13

Ortmann, [email protected] of Logistics, University of Stellenbosch, Stel-lenbosch, Western Cape, South Africa . . . . . . . . . . RD-18

Ortuño, M. [email protected]ística e Investigación Operativa, Universidad Com-plutense de Madrid, Madrid, Spain . . . . . . . . . . . . . RC-07

Osei-Bryson, [email protected] Commonwealth University, Richmond, Virginia,United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-06

Osei-Bryson, [email protected] Systems, Virginia Commonwealth University,Richmond, United States . . . . . . . . . . . . . . . . . . . . RD-05

Osman, Ibrahim [email protected] School of Business, American University of Beirut,

102


Beirut, Lebanon . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-01

Oum, [email protected] Procurement, Pacific Gas and Electric Company,San Francisco, CA, United States . . . . . . . . . . . . . . TC-13

Oyama, [email protected] for Policy Studies, National Graduate Institute forPolicy Studies, Tokyo, Japan . . . . . . . . . . . . . . . . . . TA-03

Ozekici, [email protected] of Industrial Engineering, Koç University, Is-tanbul, Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-13

Özkazanç, Ümit [email protected], Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-16

Ozlen, [email protected] of Mathematical and Geospatial Sciences, RMITUniversity, Melbourne, VIC, Australia . . . . . . . . . . FB-09

Pacciarelli, [email protected] di Ingegneria, Università Roma Tre, Roma,Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-03

Pagac, [email protected] Technology and Systems, Botany, NSW, AustraliaMD-11

Paiva, [email protected]. of Production Engineering, Federal University ofSao Carlos, SaoCarlos, Brazil . . . . . . . . . . . . . . . . . TE-01

Papamichail, K. [email protected] Business School, University of Manchester,Manchester, United Kingdom . . . . . . . . . . . . . . . . . RE-09

Paradi, [email protected] Engineering and Applied Chemistry, Univeresityof Toronto, Toronto, Ontario, Canada . . . . . . . . . . . TD-13

Park, [email protected] School of Management, KAIST, Seoul, Korea,Republic Of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-09

Park, [email protected] University, Seoul, Korea, Republic Of . . . . TC-07

Pasupathy, [email protected] of Missouri, Columbia, United States . . TE-07

Paulus, Jacob [email protected]

Cqm B.v., Eindhoven, Netherlands . . . . . . . . . . . . . TC-10

Pauws, [email protected] Interaction (MI) group, Philips Research Europe,Eindhoven, Netherlands . . . . . . . . . . . . . . . . . . . . . MD-18

Pelser, [email protected] Institute, Armscor, Pretoria, Gauteng, SouthAfrica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-03

Perea, [email protected]ística e Investigación Operativa Aplicadas y Calidad,Universidad Politécnica de Valencia, Valencia, Spain MA-03

Pereira, [email protected] and Systems, University of Minho, Braga, Por-tugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-13

Perin, [email protected] Mathematics, State University of Campinas,Campinas, São Paulo, Brazil . . . . . . . . . . . . . . . . . ME-02

Perl, [email protected] and Systems Engineering, Pontificia UniversidadCatólica de Chile, Santiago, Chile . . . . . . . . . . . . . ME-04

Persson, [email protected] of Systems and Software Engineering,Blekinge Institute of Technology, Karlshamn, AfghanistanTC-07

Persson, [email protected] of Software and Systems Engineering,Blekinge Institute of Technology, Ronneby, Sweden . TC-07

Perzina, [email protected] of Mathematical Methods in Economics, Sile-sian University, School of Business, Karvina, Czech Re-public . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-09

Pesenti, [email protected] - Dept. of Management, University of Venezia,Venezia, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-01

Petrovic, [email protected] of Engineering and Computing, Coventry Univer-sity, Coventry, United Kingdom . . . . . . . . . . . . . . . TD-07

Petrovic, [email protected] of Operations Management and Information Sys-tems, Nottingham University Business School, Notting-ham, United Kingdom . . . . . . . . . . . . . . . . . . . . . . . RE-07

103


Pilgrim, [email protected] University of Sheffield, Sheffield, United KingdomRE-07

Pino, José [email protected]ística e Investigación Operativa, Universidad deSevilla, Sevilla, Spain . . . . . . . . . . . . . . . . . . . . . . . TE-09

Pinto Ferreira, Ma [email protected] of Mathematics, ISEP, PORTO, Portugal FA-18

Pioro, [email protected] of Telecommunications, Warsaw University ofTechnology, Warszawa, Poland . . . . . . . . ME-18, TA-18

Pizarro Romero, [email protected]. de Estadística e Investigación Operativa, Universi-dad Rey Juan Carlos, Móstoles, Spain . . . . . . . . . . TA-09

Pla, [email protected], University of Lleida, Lleida, Spain . TA-05

Plateau, Agnè[email protected] d’Étude et de Recherche en Informatique du Cnam,Paris cedex 03, France . . . . . . . . . . . . . . . . . . . . . . MC-03

Poldaru, [email protected] of Economics and social sciences, Estonian Uni-versity of Life Sciences, Tartu, Estonia . . . . . . . . . RD-14

Poli, [email protected] Engineering, Universidade Federal de Sao Car-los, Sao Carlos, Sao Paulo, Brazil . . . . . . . . . . . . . . TA-16

Porkka, [email protected] of Industrial Management and Logistics, LundUniversity, LUND, Sweden . . . . . . . . . . . . . . . . . . MC-01

Potgieter, Petrus [email protected] of Decision Sciences, University of SouthAfrica, Unisarand . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-12

Potts, [email protected] of Mathematics, University of Southampton,Southampton, Hampshire, United Kingdom . . . . . MC-18

Pranzo, [email protected] di Ingegneria dell’Informazione, Universitàdi Siena, Siena, Italy . . . . . . . . . . . . . . . . . MC-03, FB-13

Preciado, Luis Javier

[email protected], Universidad de Sonora, Hermosillo, Sonora,Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-14

Puranen, [email protected] of Mathematical Information Technology, Uni-versity of Jyväskylä, Jyväskylä, Finland . . . . . . . . . TE-09

Pureza, [email protected] Engineering, Universidade Federal de Sao Car-los, Sao Carlos, Sao Paulo, Brazil . . . . . . . . . . . . . . TA-16

Puterman, [email protected] for Operations Excellence - University of BritishColumbia, Canada . . . . . . . . . . . . . . . . . . . . . . . . . . TA-13

Qiang, [email protected] and Environmental Engineering, National Universityof Singapore, Singapore, Singapore . . . . . . . . . . . . ME-11

Quadrifoglio, [email protected] Engineering, Texas A&M University, College Sta-tion, Texas, United States . . . . . . . . . . . . . . . . . . . . TE-16

Raad, [email protected], University of Stellenbosch, Cape Town, SouthAfrica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-07

Rabasa, [email protected] de Investigación Operativa, Universidad MiguelHernández, Elche, Alicante, Spain . . . . . . . . . . . . . TE-10

Raith, [email protected] Science, The University of Auckland, Auck-land, New Zealand . . . . . . . . . . . . . . . . . . MD-08, TA-08

Ramón, [email protected] de Investigación Operativa, Universidad MiguelHernández, Elche, Spain . . . . . . . . . . . . . . . . . . . . . TC-14

Ramik, [email protected]. of Math. Methods in Economics, Silesian University,School of Business, Karvina, Czech Republic . . . . FB-09

Ramokgadi, [email protected] sciences, Unisa, Pretoria, RSA, South Africa FA-08

Rand, [email protected]. of Management Science, Lancaster University, Lan-caster, Lancashire, United Kingdom . . . . TA-05, TD-05,TE-05, RA-10

Randall, Paul

104


[email protected] St, HM Treasury, London, United KingdomTA-07

Rangel, [email protected] - Sao Paulo State University, S.J. do Rio Preto, SãoPaulo, Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-11

Rapp, [email protected] of Management, IMIT, Ljusterö, Sweden . RC-08

Rasmussen, Matias [email protected] of Computer Science, University of Copen-hagen, Copenhagen, Denmark . . . . . . . . . . . . . . . . . TE-07

Rath, [email protected] of Statistics and Decision Support Systems,University of Vienna, Vienna, Austria . . . . . . . . . . RC-07

Ratliff, H. [email protected] of Industrial and Systems Engineering, Georgia In-stitute of Technology, Atlanta, GA, United States . TB-01

Raubenheimer, [email protected] for BMI, North-West University, Potchefstroom,South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-13

Ray, Ajit [email protected] of Commerce, University of North Bengal,Siliguri, West Bengal, India . . . . . . . . . . . . . . . . . . MC-14

Reber, [email protected], Paris, France . . . . . . . . . . . . . . . . . . . . FA-05

Reeves, Colin [email protected] of Mathematical Sciences, Coventry University,Coventry, United Kingdom . . . . . . . . . . . . . . . . . . . TD-07

Rentz, [email protected] Engineering, Institute Of Industrial Production,Kalsruhe, Germany . . . . . . . . . . . . . . . . . . FA-12, RC-15

Resende, [email protected] & Optimization Research, AT&T Labs Re-search, Florham Park, NJ, United States . . . . . . . . TC-18

Rey, Pablo [email protected] Engineering, University of Chile, Santiago, ChileMD-01, FB-08

Ribas, [email protected] Business Administration, Universitat Politec-nica de Catalunya, Barcelona, Spain . . . . . . . . . . . . RE-11

Ribeiro, [email protected] of Computer Science, Universidade FederalFluminense, Rio de Janeiro, RJ, Brazil . . . . . . . . . . TA-04

Ribeiro, [email protected], ITA, Sao Jose dos Campos, Sao Paulo, BrazilTC-05

Rodríguez Martín, [email protected], Universidad de La Laguna, La Laguna, Tenerife,Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-02

Rodrigues, [email protected] Management University, Singapore, SingaporeTC-16

Rodrigues, Rita de Cássia [email protected] for Computing and Applied Mathematics(LAC), Brazilian Institute for Space Research (INPE), SaoJose dos Campos, Sao Paulo, Brazil . . . . . . . . . . . . RC-13

Rodriguez-Sanchez, Sara [email protected] Program in Systems Engineering, UniversidadAutonoma de Nuevo Leon, San Nicolas de los Garza,Nuevo Leon, Mexico . . . . . . . . . . . . . . . . . . . . . . . . TA-05

Rojas-Hernandes, [email protected] of Limerick, Limerick, Ireland . . . . . . . RC-04

Romero, [email protected] Engineering, ITAM - Instituto Tecnologico Au-tonomo de Mexico, Mexico City, DF, Mexico . . . MC-07

Rönnqvist, [email protected]épartement de génie mécanique, Québec, Canada TC-04

Roots, [email protected] University of Life Sciences, Tartu, Estonia . RD-14

Rosenhead, [email protected], London School of Economics, London MC-08

Rostamy-Malkhalifeh, [email protected] of Mathematics, Science and ResearchBranch,Islamic Azad University, Tehran, Iran, Islamic Re-public Of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-15

Roth, [email protected] and Biosystems Engineering, University of Pretoria,Pretoria, Gauteng, South Africa . . . . . . . . . . . . . . MA-03

105


Roucairol, [email protected] lab. - CNRS, Univ. of Versailles Saint Quentin enYvelines, Versailles Cedex, France . . . . . . . . . . . . . FB-06

Roux, [email protected] Radar Systems, Somerset West, South Africa FA-03, MD-16, MD-18

Roy, [email protected], Université Paris-Dauphine, Paris Cedex 16,France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-08

Ruhland, [email protected] of Jena, Faculty of Business and Economics,Jena, Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-06

Ruiz, José [email protected] de Investigacion Operativa, Universidad MiguelHernández, Elche, Alicante, Spain . . . . . . . . . . . . . TC-14

Ruiz, [email protected] de Estadistica e Investigación OperativaAplicadas y Calidad, Universitat Politècnica de València,Valencia, Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-09

Ruiz-Hernandez, [email protected] Methods, CUNEF, Madrid, Spain . . . ME-19

Ryan, [email protected] Science, The University of Auckland, Auck-land, New Zealand . . . . . . . . . . . . . . . . . . MD-03, TC-16

SABWA, [email protected] Sciences, University of Cape Town, Cape Town,Western Cape, South Africa . . . . . . . . . . . . . . . . . . TC-15

Salazar González, Juan José[email protected]ística e Investigación Operativa, Universidad de LaLaguna (Tenerife), La Laguna, Tenerife, Spain . . TD-01,FB-02, RD-16

Saldanha-da-Gama, [email protected]/DEIO, University of Lisbon, Lisbon, Portugal TA-09

Salhi, [email protected] Business School, University of Kent, Canterbury,Kent, United Kingdom . . . . . . . . TA-01, FA-08, RC-16

SALIDO, MIGUEL [email protected] Informaticos Y Computacion, Universidad Po-litecnica De Valencia, Valencia, Spain . . . . . . . . . . ME-03

Salminen, [email protected] of Business and Economics, University ofJyväskylä, Jyväskylä, Finland . . . . . . . . . . . . . . . . . TD-08

Salo, [email protected] Analysis Laboratory, Aalto University School ofScience and Technology, Aalto, Finland . MA-08, TE-08

Sanchez, [email protected] Universidad Católica de Chile, Santiago, ChileTD-04

Sant?Anna, Annibal [email protected] de Pós graduação em Engenharia de Produção,Universidade Federal Fluminense, Rio de Janeiro, Rio deJaneiro, Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-09

Santiváñez, José[email protected] and Management Engineering, Universidad delTurabo, Gurabo, Puerto Rico . . . . . . . . . . . . . . . . . RA-16

Santos, [email protected] de Telecomunicações - Pólo de Aveiro, Aveiro,Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-18

Saraç, [email protected] of Industrial Engineering, Eskisehir Os-mangazi University, Eskisehir, Turkey . . . . . . . . . . RE-02

Sargent, [email protected] of Electrical Engineering & Computer Sci-ence, Syracuse University, Syracuse, NY, United StatesMA-09

Sauma, [email protected] and Systems Engineering Department, PontificiaUniversidad Catolica de Chile, Santiago, Chile . . MA-12

Savelsbergh, [email protected] of Industrial and Systems Engineering, Georgia In-stitute of Technology, Atlanta, GA, United States . TD-09

Sawik, [email protected] Research & Information Technology, AGHUniversity of Science & Technology, Krakow, Poland RD-13

Saxsena, [email protected], DRDO Delhi, New Delhi, Delhi, India . . . . RD-06

Sáez, [email protected] Engineering Department, Universidad de Chile,Santiago, Chile . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-01

106


Sällberg, [email protected] of Management, Blekinge Institute of Technology,Ronneby, Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . RC-08

Sánchez-Soriano, [email protected] de Investigación Operativa, Universidad MiguelHernández, Elche, Alicante, Spain . . . . . . . . . . . . . TE-10

Scarpin, Cassius [email protected] Federal do Paraná, Curitiba, Paraná, BrazilTE-07

Schauten, [email protected] of Economics and Business Economics, FinanceGroup, Erasmus University Rotterdam, Rotterdam, Nether-lands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-12

Schittekat, [email protected], SINTEF, Oslo, Norway . . . . . . . . . . . . . . . . . MC-01

Schmid, [email protected] Chain Management, Europa-Universität Viadrina,Frankfurt, Germany . . . . . . . . . . . . . . . . . . . . . . . . . TD-09

Schmidt, Gü[email protected] University, Saarbrücken, Germany . . . . . RA-12

Schmitz, [email protected] and Quantitative Methods, CSIR Built Environ-ment, Pretoria, Gauteng, Afghanistan . . . FA-05, TD-11,TE-16

Schoen, [email protected] Business School, Leibniz University Hannover,Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-12

Schorpp, [email protected]&Logistics, Universität Augsburg, Augsburg,Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-16

Schriber, Thomas [email protected] Information Technology, University of Michigan,Ann Arbor, Michigan . . . . . . . . . . . . . . . . . . . . . . . MC-09

Schultmann, [email protected] for Industrial Production, Karlsruhe Institute ofTechnology (KIT), Karlsruhe, Germany . . . . . . . . . RE-13

Scott, [email protected] Sciences, University of Cape Town, Cape Town,Western Cape, South Africa . . . . . . . . . . . . . . . . . . RE-15

Scurr, [email protected] Operational Risk, Barclays, London, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-12

Sedeño-Noda, [email protected] de La Laguna, La Laguna. Tenerife, SpainME-01, RD-13

Sell Dyminski, [email protected] Federal do Paraná, Curitiba, Paraná, BrazilTD-15

Sen, [email protected] University, New Delhi, Delhi, India . . . . . . MC-13

Senne, [email protected], UNESP/FEG, Guaratingueta, SP, Brazil RC-13

Serafini, [email protected]. of Mathematics and Computer Science, Universityof Udine, Udine, Italy . . . . . . . . . . . . . . . . . . . . . . . TA-03

Serguieva, [email protected] for the Analysis of Risk and Optimisation Mod-elling Applications, Brunel University, Uxbridge, London,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-12

Serrato, [email protected], Consulting & Continuing Education Programs,Tecnologico de Monterrey, Morelia, Michoacan, MexicoME-07

Sethi, [email protected] School of Management - ISOM, University of Texasat Dallas, Richardson, TX, United States . . . . . . . ME-16

Sezer, [email protected] d’Evry, Evry, France . . . . . . . . . . . . . . . RD-02

Shaikh, [email protected] of Business Administration, Prince MohammadBin Fahd University, Al-Khobar, Saudi Arabia . . . FA-08

Shao, [email protected], Houston, Tx, United States . . . . . . . MD-08

sharma, [email protected] Broadcasting Engineering Services, Government ofIndia, New Delhi, India . . . . . . . . . . . . . . . . . . . . . . TE-03

Sher, [email protected]

107


Department of Planning, Israel Police, Modiin, Israel TA-03

Sherali, [email protected] Tech Industrial & Systems Engineering, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-01, ME-07

Shingwenyana, [email protected] of Chemical Engineering, University of Preto-ria, Pretoria, South Africa . . . . . . . . . . . . . . . . . . . . RC-04

Shintani, [email protected] of Marine Science and Technology, Tokai Univer-sity, Shizuoka, Shizuoka, Japan . . . . . . . . . . . . . . . ME-11

Shiode, [email protected] Gakuin University, Kobe, Japan . . . . . . . . . . RE-06

Shrikant, [email protected], DRDO Delhi, New Delhi, Delhi, India . . . . RD-06

Sicilia, Joaquí[email protected]. de Estadística, Investigación Operativa y Com-putación, Universidad de La Laguna., La Laguna. Tener-ife., Canary Islands, Spain . . . . . . . . . . . . . . . . . . . . TE-09

Sier, [email protected] and Information Sciences, CSIRO, Clayton,Victoria, Australia . . . . . . . . . . . . . . . . . . . RC-11, RE-11

Sigauke, [email protected] Mathematics, National University of Science andTechnology, Bulawayo, Zimbabwe . . . . . . . . . . . . MC-13

Sigue, [email protected] of Business, Athabasca University, Athabasca,Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-09

Sillekens, [email protected] Support and OR Lab, University of Paderborn,Paderborn, Germany . . . . . . . . . . . . . . . . . . . . . . . . TE-10

Silva, [email protected]ção, Instituto Tecnológico de Aeronáutica, Sao Josedos Campos, Sao Paulo, Brazil . . . . . . . . . . . . . . . . TC-05

Sim, [email protected] University, Seoul, Korea, Republic Of . . . . TC-07

Simar, Lé[email protected] de statistique, Universite Catholique de Louvain,Louvain la Neuve, Belgium, Belgium . . . . . . . . . . . TE-14

Simões, [email protected], IST, Lisbon, Portugal . . . . . . . . . . . . . . . . RE-15

Simon de Blas, [email protected] & Operations Research, Rey Juan Carlos Univer-sity, Mostoles, Madrid, Spain . . . . . . . . . . . . . . . . . RA-04

Simpson, [email protected], University of Sheffield, Sheffield, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-07

Singh, Manu [email protected] Science, Dr. B.R. Ambedkar University, Agra,UP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-02

Sinske, [email protected] Software (Pty) Ltd, Stellenbosch, WC, South AfricaMC-07

Sinuany-Stern, [email protected] Engineering and Management, Ben Gurion Uni-versity, Beer-Sheva, Israel . . . . . . . . . . . . . . . . . . . MD-12

Sipahioglu, [email protected] Engineering, Osmangazi University, Eskisehir,Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-02, RE-16

Sirvent, [email protected] de Investigación Operativa, Universidad MiguelHernández, Elche, Alicante, Spain . . . . . . . . . . . . . TC-14

Sivakumar, [email protected] of Mathematics, Madurai-Kamaraj University,Madurai, Tamilnadu, India . . . . . . . . . . . MA-16, RC-19

Skocdopolova, [email protected]. of Econometrics, University of Economics Prague,Prague 3, Czech Republic . . . . . . . . . . . . . . . . . . . MD-17

Sladky, [email protected] of Econometrics, Institute of Information The-ory and Automation, Academy of Sciences of the CzechRepublic, Prague 8, Czech Republic . . . . . . . . . . . MA-02

Sliwinski, [email protected] of Control and Computation Engineering, WarsawUniversity of Technology, Warsaw, Poland . . . . . . ME-18

Soares, [email protected], Campinas, SP, Brazil . . . . . . . . . . . . . MC-02

Sodhi, [email protected]

108


Industrial & Manufacturing Engineering, University ofRhode Island, Kingston, RI, United States . . . . . . . FB-13

Sodini, [email protected] and Applied Mathematics, University of Pisa,Pisa, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-02

Sohn, So [email protected], Yonsei university, Seoul, Korea, Republic Of FB-03

Soriano, [email protected], HEC Montréal, Montreal, Quebec, CanadaMC-01

Speranza, M. [email protected]. of Quantitative Methods, University of Brescia,Brescia, Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-01

Spieksma, [email protected] Research and Business Statistics, KU Leuven,Leuven, Belgium . . . . . . . . . . . . . . . . . . . . TA-02, FB-11

Spronk, [email protected] School of management, Erasmus University,Rotterdam, Netherlands . . . . . . . . . . . . . . . . . . . . . . RC-12

Staffetti, [email protected] of Statistics and Operations Research, Univer-sity Rey Juan Carlos, Mostoles, Madrid, Spain . . . FA-02

Stefani, [email protected] Methods for Economics and Business Sci-ences, Università Milano Bicocca, Milano, Italy . . FB-12

Stein, [email protected] of Operations Research, Karlsruhe Institute ofTechnology, Karlsruhe, Germany . . . . . . . . . . . . . MC-15

Steiner Neto, [email protected], Federal University at Paraná, Curitiba, Pr., BrazilTE-07, TD-15

Stern, [email protected] College of Tel-Aviv Yaffo, Tel-Aviv, Israel TC-01

Stevens, [email protected], University of Sheffield, Sheffield, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-07

Stewart, [email protected] Sciences, University of Cape Town, Ronde-bosch, South Africa . . . . . . . . . . . TA-08, RE-14, TC-15

Steyn, [email protected] of Computer, Statistical and Mathematical Sci-ences, North-West University, Potchefstroom, South AfricaRD-18, RE-18

Stojanovic, [email protected] of Mathematics, Faculty of Natural Sciencesand Mathematics, Kragujevac, Serbia . . . . . . . . . . MD-18

Stopher, [email protected] of Transport & Logistics Studies, The Universityof Sydney, The University of Sydney, NSW, Australia TE-16

Strusevich, [email protected] of Computing and Mathematical Sciences, Univer-sity of Greenwich, London, United Kingdom . . . . MC-18

Strydom, [email protected] sciences, Unisa, Pretoria, South Africa . . TC-01

Stummer, [email protected] of Business Administration and Economics,Bielefeld University, Bielefeld, Germany TC-05, MD-09

Stylianides, [email protected] for Logistics and Decision Support, Pretoria, SouthAfrica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-05

Sumalee, [email protected] Polytechnic University of Hong Kong, Kowloon, HongKong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-03

Sundar, [email protected]. of Oncology, City Hospital Campus, NottinghamUniversity Hospitals NHS Trust, Nottingham, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-07

Sunke, [email protected] of Business Administration, Construction Manage-ment and Economics, University of Siegen, Siegen, Ger-many . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RE-13

Sunter, [email protected] American, Johannesburg, South Africa . . . MB-01

Suzuki, [email protected] of Engineeing, Information and Systems, Univer-sity of Tsukuba, Tsukuba, Ibaraki, Japan . . . . . . . . RC-15

Swanepoel, [email protected] Sciences, University of South Africa, Pretoria,

109


Gauteng, South Africa . . . . . . . . . TD-02, TA-05, TC-05

Syntetos, [email protected] Business School, Cardiff University, CARDIFF,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-19

Sze, San [email protected] & Business Statistics, University of Sydney,Sydney, New South Wales, Australia . . . . . . . . . . . RE-13

Takehara, [email protected] School of Finance, Accounting and Law, WasedaUniversity, Chuo-ku, Tokyo, Japan . . . . . . . . . . . . MD-13

Talla Nobibon, [email protected] Sciences and Information Management, KU Leu-ven, Leuven, Belgium . . . . . . . . . . . . . . . . . . . . . . . TA-02

tamaki, [email protected] Administration, Aichi Univeristy, Nishikamo,Aichi, Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-02

Tamarit, [email protected]. Statistics and Operations Research, University of Va-lencia, Burjassot, Spain . . . . . . . . . . . . . . . . . . . . . . FA-13

Tambovceva, [email protected] of Engineering Economics and Management, RigaTechnical University, Riga, Latvia . . . . . . . . . . . . . RC-15

Taylan, [email protected], Dicle University, Diyarbakır, Turkey RD-02

Teixeira de Almeida, [email protected] Engineering, Universidade Federal de Per-nambuco - UFPE, Recife, PE, Brazil . . . . . . . . . . . ME-19

Tello, [email protected] de Monterrey, Campus Morelia, Morelia, Mi-choacan, Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-07

Teo, Chung [email protected] University of Singapore, Singapore, SingaporeTE-01

Terblanche, Fanie (SE)[email protected] for Business Mathematics and Informatics, North-West University, Potchefstroom, South Africa . . . ME-18

Thomas, [email protected]. of Electrical and Electronic Engineering, Universityof Melbourne, Melbourne, Victoria, Australia . . . MC-04

Thomas, [email protected] of Managemento, University of Southampton,Southampton, Hants, United Kingdom . . . . . . . . . MD-12

Tikkanen, [email protected] of Eastern Finland, Joensuu, Finland . . TC-04,RE-09

Toloo, [email protected], Iran, Islamic Republic Of . . . . . . . . . . . . . RC-14

Tomaszewski, [email protected] of Telecommunications, Warsaw University ofTechnology, Warszawa, Poland . . . . . . . . . . . . . . . . TA-18

Tong, [email protected] Engineering, The University of Hong Kong, HongKong, Hong Kong . . . . . . . . . . . . . . . . . . . . . . . . . . RA-03

Tong, [email protected] School, Loughborough University, Loughbor-ough, United Kingdom . . . . . . . . . . . . . . . . . . . . . . MA-11

Toppila, [email protected] Analysis Laboratory, Helsinki University of Tech-nology, TKK, Finland . . . . . . . . . . . . . . . . . . . . . . MA-08

TORMOS, [email protected] And Operations Research, Universidad Politec-nica De Valencia, VALENCIA, SPAIN, Spain . . . ME-03

Toth, [email protected], University of Bologna, Bologna, Italy . . . . MA-01,TD-01

Touati, [email protected] UMR CNRS 7030, Institut Galilée - Université Paris13, Villetaneuse, France . . . . . . . . . . . . . . . . . . . . . . TA-02

Tran, [email protected] Engineering and Applied Chemistry, Universityof Toronto, Toronto, Ontario, Canada . . . . . . . . . . . TD-13

Tran-Thanh, [email protected], Budapest University of Technologyand Economics, Budapest, Hungary . . . . . . . . . . . . TE-18

Trick, [email protected] School of Business, Carnegie Mellon University,Pittsburgh, PA, United States . . . . . . . . . . TA-04, RD-10

Tripathy, [email protected]

110


Vinod Gupta School of Management, Kharagpur, WestBengal, India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-14

Troncoso, Juan José[email protected] of Industrial Engineering, University of Chile,Santiago, Chile . . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-04

Tsao, [email protected]. of Finance, National Pingtung Institute of Com-merce, Pingtung, Taiwan . . . . . . . . . . . . . . . . . . . . . FB-07

Tseveendorj, [email protected], University of Versailles, Versailles, 78, FranceFB-06

Tsoukiàs, [email protected] - Lamsade, Paris Cedex 16, France . TA-08, RD-10

Turkay, [email protected] of Industrial Engineering, Koc University, Is-tanbul, Turkey . . . . . . . . . . . . . . . . FA-10, FB-10, FB-17

Turpin, [email protected], University of Pretoria, Pretoria, South AfricaRE-15

Udding, Jan [email protected] of Mechanical Engineering, Eindhoven Uni-versity of Technology, Eindhoven, Netherlands . . MD-11

Ueda, [email protected] of Science and Technology, Seikei University,Musashino-Shi, Tokyo, Japan . . . . . . . . . . . . . . . . . TA-04

UNAL, [email protected] Engineering, Gazi University, ANKARA, TR,Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-14

Ünlüyurt, [email protected] Systems/Industrial Engineering, SabanciUniversity, Ýstanbul, Turkey . . . . . . . . . . . . . . . . . . FB-11

Usberti, Fá[email protected], Universidade Estadual de Campinas, Campinas,SP, Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-03

Uusivuori, [email protected] Forest Research Institute Metla, Helsinki, FinlandRA-04, FA-17

Valdes, [email protected] and Operations Research, Complutense Univer-sity of Madrid, Madrid, Spain . . . . . . . . . . . . . . . . . FB-15

Valdivia, [email protected] Engineering and Logistics, Pontificia Universi-dad Catolica, Santiago, RM, Chile . . . . . . . . . . . . . TD-16

Valeiras, [email protected]ática Aplicada I, Universidad de Sevilla, Sevilla,Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TE-09

Valente, [email protected] de Economia - LIAAD - INESC Porto L.A.,Universidade do Porto, Porto, Portugal . . . . . . . . . . FA-13

Vallim, [email protected] Science, Universidade Presbiteriana Mackenzie,Sao Paulo, SP, Brazil . . . . . . . . . . . . . . . . . . . . . . . . TD-02

van der Mei, [email protected] Wiskunde & Informatica, Amstedam, Nether-lands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-04

van der Merwe, [email protected], Parklands, South Africa . . . . . . . . . . . . . . . . . FB-17

Van der Merwe, [email protected] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-09

van der Walt, [email protected], South Africa . . . . . . . . . . . . . . . . . . RD-17

van Dieman, [email protected], Stellenbosch University, Atlantis, Western Cape,South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-18

van Dyk, [email protected] and Quantitative Methods, CSIR Built Environ-ment, Stellenbosch, Western Cape, South Africa . . TD-04

Van Hoepen, [email protected] Sciences, University of South Africa, Pretoria,Gauteng, South Africa . . . . . . . . . . . . . . . . . . . . . . . TA-05

van Nierop, [email protected] Chain Software, Barloworld Logistics, SouthAfrica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-17

van Staden, [email protected] of Computer, Statistical and Mathematical Sci-ences, North-West University, Potchefstroom, South AfricaRE-18

van Vuuren, [email protected]

111


Department of Industrial Engineering, Stellenbosch Uni-versity, Stellenbosch, Western Cape, South Africa FA-03,TD-04, MC-07, RD-07, FB-13, MD-16, MD-18,RD-18

Van Wassenhove, [email protected] and Operations Management Area, INSEAD,Fontainebleau cedex, France . . . . . . . . . . . . . . . . . . RB-01

Van Zyl, [email protected] and Systems Engineering, University of Pretoria,Pretoria, South Africa . . . . . . . . . . . . . . . . . . . . . . . RA-11

Vanags, [email protected] of Engineering Economics, Riga Technical univer-sity, Riga, Latvia . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-07

Vanhems, [email protected], Toulouse Business School, Toulouse, FranceTD-14

Varbanov, [email protected], University of Pannonia, Veszprem, HungaryRC-04, RD-04

Vargas, [email protected], Universidad de Sonora, Hermosillo, Sonora,Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-14

Vargas, [email protected] Graduate School of Business, University of Pitts-burgh, Pittsburgh, PA, United States . . . . . . . . . . . . FA-09

Varghese, [email protected], IIT Delhi, New Delhi, Delhi, India . RD-06

Vasconcelos Silva, Jackson [email protected], Centre for Management Studies of IST, Techni-cal University of Lisbon /, Universidade Estadual do Ceará,Fortaleza, Ceará, Brazil . . . . . . . . . . . . . . . . . . . . . . TD-03

Vaughan, David [email protected] of Services Management, Bournemouth University(UK), Poole, Dorset, United Kingdom . . . . . . . . . MD-14

Velasquez, [email protected] Ltda., Bogotá . . . . . . . . . . . . . . . . . MD-17

Veldhuizen, [email protected] Or, Sasol Technology, Sasolburg, South Africa . MC-04

Venter, [email protected]

Institute for Maritime Technology, Simon’s Town, SouthAfrica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-16

Venter, [email protected] Support, Sasol Technology, Secunda, AfghanistanMC-16

Verbeke, [email protected] School of Business, University of Calgary, Cal-gary, Alberta, Canada . . . . . . . . . . . . . . . . . . . . . . . RD-15

Vianna, André[email protected], UNESP - Bauru, Bauru, São Paulo, BrazilRD-18

Vieira, Luí[email protected] Civil, Faculdade de Engenharia do Porto, Porto,Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-18

Villwock, [email protected] Federal do Paraná, Curitiba, Paraná, BrazilTD-15

Visagie, [email protected] of Logistics, University of Stellenbosch, Stel-lenbosch, South Africa . . . . . . . . . . . . . . . RE-02, MC-16

Vitoriano, Begoñ[email protected]ística e Investigación Operativa I, Fac. Matemáticas,Universidad Complutense de Madrid, Madrid, Spain RC-07

Vivian, [email protected] of Economic and Business Sciences, University ofthe Witwatersrand, Johannesburg, Gauteng, South AfricaRE-09

Wagner, [email protected] for Agricultural and Nutrition Sciences; Chairof Agricultural Economics, Martin-Luther University ofHalle-Wittenberg, Halle (Saale), Germany . . . . . . MD-14

Wakolbinger, Lea [email protected] of Business Administration, University of Vi-enna, Vienna, Austria . . . . . . . . . . . . . . . . . . . . . . . MD-09

Wallenius, [email protected] Engineering and Management, Aalto UniversitySchool of Science, Espoo, Finland . . . . . . . . . . . . . TE-08

Wallenius, [email protected] and Service Economy, Aalto UniversitySchool of Business, Helsinki, Uusimaa, Finland . . TE-08

112


Wang, Judith [email protected] and Environmental Engineering, University of Auck-land, Auckland, New Zealand . . . . . . . . . . . . . . . . . TA-08

Wang, [email protected] Engineering, National University of Singapore, Sin-gapore, Singapore . . . . . . . . . . . . . . . . . . . . . . . . . . ME-11

Wang, [email protected] school of Maritime Sciences, Kobe University,Kobe, Hyogo, Japan . . . . . . . . . . . . . . . . . . . . . . . . ME-11

Wang, [email protected] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-08

Washburn, [email protected] and Coastal Engineeering, University of Florida,Gainesville, FL . . . . . . . . . . . . . . . . . . . . . . . . . . . . TC-09

Wassan, [email protected] Business School, University of Kent, United King-dom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TA-01

Weber, [email protected] de l’Innovation et de la Recherche, SNCF, Paris,France . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-03, ME-03

Weber, [email protected]ät Essen, Essen, Germany . . . . . . . . . . . . . FA-17

Weber, [email protected] of Applied Mathematics, Middle East TechnicalUniversity, Ankara, Turkey . . . . . . . . . . . . RD-02, FB-10

Wedley, [email protected] of Business Administration, Simon Fraser Univer-sity, Burnaby, BC, Canada . . . . . . . . . . . . . . . . . . . ME-14

WEE, Kwan [email protected] Management University, Singapore, SingaporeTC-16

Weintraub, André[email protected] engineering, University of Chile, Santiago, ChileMC-04, RA-04, RD-10

Welch, [email protected] National Laboratories, Albuquerque, NM, UnitedStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MA-19

Wessäly, [email protected] GmbH, Berlin, Germany . . . . . . . . . . . . . . . ME-18

Westermann, [email protected] Software GmbH, Koeln, Germany MD-17, TA-17

White, [email protected] Department, University of Bristol, Bristol,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . FA-05

Wilke, [email protected] of Pretoria, Pretoria, South Africa . . . . MC-02

Willemse, Elias [email protected] Environment, CSIR, Pretoria, Gauteng, South AfricaME-01

Wong, [email protected] Parcel Service, Timonium, Maryland, AfghanistanRD-03

Wong, [email protected] University of Hong Kong, Hong Kong, Hong KongRC-03

Wood, [email protected] Research Department, Naval PostgraduateSchool, Monterey, Ca, United States . . . . . . . . . . . MC-04

Wu, [email protected] of Economics, The University of Queensland, Bris-bane, Queensland, Australia . . . . . . . . . . . . . . . . . . TC-14

Xie, [email protected] and Logistic College, Dalian Maritime Uni-versity, Dalian, China . . . . . . . . . . . . . . . . . . . . . . . TC-08

Y S, [email protected] of Computer Applications, Dayanand Sagar Col-lege of Engg, Bangalore, Karnataka, India . . . . . . . FA-16

Yadavalli, [email protected] & Systems Engineering, University of Pretoria,Pretoria, Gauteng, South Africa RE-04, RC-11, MD-13,MA-16, RC-19

Yalaoui, [email protected] Charles Delaunay, ICD LOSI, University of Tech-nology of Troyes, Troyes, France . . . . . . . RA-16, FB-18

Yan, [email protected] University of Hong Kong, Hong Kong, ChinaME-16

Yanasse, Horacio

113


[email protected], INPE, São José dos Campos, SP, Brazil . . . . RC-13

Yang, [email protected] Division, Institute of Policy and Manage-ment,Chinese Academy of Sciences, Beijing, China . TC-08

Yano, [email protected] Dept. and Haas School of Business, University ofCalifornia, Berkeley, Berkeley, CA, United States ME-16

yekkalam, [email protected] Of Mathematics, Islamic Azad University Cen-tral Tehran Branch, tehran, tehran . . . . . . . . . . . . . TD-14

Yeung, [email protected] of Industrial Engineering & Automatic Con-trol, Ecole des Mines de Nantes / IRCCyN, Nantes, FranceMA-12

Yi, [email protected] College, Shanghai Ocean University, Shang-hai, China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MD-07

Ying-qi, [email protected] Fisheries University, Shanghai, China . MD-07

Yon -Woo, [email protected] Div., Korea Atomic Energy Research Institute, Tae-jeon, Taejeon, Korea, Republic Of . . . . . . . . . . . . . RE-05

Yoneda, [email protected] of Economics, Fukuoka University, Fukuoka,Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FB-15

Yu, [email protected] for the Analysis of Risk and Optimisation Mod-elling Applications, Brunel University, Uxbridge, London,United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . RA-12

YU, [email protected] School of Management, Erasmus University,Rotterdam, Netherlands . . . . . . . . . . . . . . . . . . . . . RD-16

Zabinsky, Zelda [email protected] Engineering, University of Washington, Seattle,WA, United States . . . . . . . . . . . . . . . . . . . . . . . . . . TC-02

Zaccour, [email protected] Sciences Department, GERAD & HEC Mon-treal, Montreal, Quebec, Canada . . . . . . . RC-09, RD-09

Zadnik Stirn, [email protected] Faculty, University of Ljubljana, Ljubljana,Slovenia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RC-08

Zagozdzon, [email protected] of Telecommunications, Warsaw University ofTechnology, Warsaw, Poland . . . . . . . . . . . . . . . . . ME-18

Zarzuelo, José [email protected] Economics IV, Basque Country University, Bil-bao, Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-09

Zhang, [email protected] of Mathematics and Systems Science, Beijing,China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ME-16

Zhang, [email protected] of Florida, Gainesville, FL, United States . FB-16

Zhelev, [email protected] of Limerick, Limerick, Ireland . . . . . . . RC-04

zheng, [email protected] College, Shanghai Fisheries University,Shanghai, China . . . . . . . . . . . . . . . . . . . . . . . . . . . RD-14

Zhigljavsky, [email protected] of Mathematics, Cardiff University, Cardiff . MC-02, RE-06

Zhong, [email protected] Parcel Service, Timonium, MD, United States RD-03

Ziegler, [email protected] for Management Sciences and Energy Economics,University Duisburg-Essen, Essen, Germany . . . . . FA-17

Zineb, [email protected] Opérationnelle, Univerite des sciences et de latechnologie Houari Boumedienne, Algeiers, Algerie, Alge-ria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MC-18

114

SESSION INDEX

Monday, 8:00 AM - 9:30 AM

MA-01: Some Models of Combinatorial Optimisation (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1MA-02: Optimization Under Uncertainty (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1MA-03: Railway Network Design (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1MA-05: Lessons from OR Consultancy Practice 1 (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2MA-08: Decision Analysis (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2MA-09: Simulation 1 (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2MA-11: Berth Allocation in Container Ports (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3MA-12: Finance: Dynamic models (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3MA-15: DEA Tutorial and Software Demo (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3MA-16: Production and Inventory Management (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3MA-19: Condition Based Maintenance and Prognostics (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Monday, 9:45 AM - 11:15 AM

MB-01: Official opening of IFORS 2008 Triennial Conference (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Monday, 11:45 AM - 1:15 PM

MC-01: Combinatorial Optimization: Vehicle Routing 1 (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5MC-02: Optimization: Gradients and Constraints (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5MC-03: Robust and On-line Problems in Railways (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5MC-04: Applications of OR in Mining (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6MC-05: Lessons from OR Consultancy Practice 2 (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6MC-07: OR-in-Development 1 (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6MC-08: Facilitated Multi-Criteria Decision Analysis (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7MC-09: Discrete Event Simulation (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7MC-11: Container Port Design and Optimization (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7MC-12: Finance: Optimisation models (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8MC-13: Finance: Applications (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8MC-14: DEA: Applications 1 (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8MC-15: DEA Theory: New developments 1 (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9MC-16: Production Lines (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9MC-17: OR Software Applications (Sandton Sun: Oleander) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9MC-18: Graphs and Networks 1 (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10MC-19: Reliability (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10


MD-01: Combinatorial Optimization: Vehicle Routing 2 (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11MD-03: Optimisation and Railroad Scheduling Problems (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11MD-07: OR-in-Development 2 (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11MD-08: Multiobjective Optimization Methods (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12MD-09: Simulation 2 (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12MD-11: Resource Scheduling and Dispatching in Container Ports (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . 12MD-12: Finance: Financial modelling (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13MD-13: Finance: Stocks and bonds (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13MD-14: DEA: Applications 2 (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13MD-16: Military Optimization Models (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14MD-17: Optimization Software (Sandton Sun: Oleander) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14MD-18: Graphs and Networks 2 (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

115

SESSION INDEX IFORS 2008 - Sandton

MD-19: Maintenance: Spare parts and logistics (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15


ME-01: Combinatorial Optimization: Routing Problems (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15ME-02: Linear Programming (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15ME-03: Timetabling and Scheduling in Railways (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16ME-04: Applications of OR in Agriculture (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16ME-07: OR-in-Development 3 (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16ME-08: Muliple Criteria Decision Aiding: Sorting problems (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . 17ME-09: Simulation 3 (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17ME-11: Container Shipping Flow (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17ME-12: Finance: Portfolio optimisation (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17ME-14: DEA: AHP / MCDM (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18ME-16: Operations Management: Inventory risk, assortment, and pricing (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . 18ME-17: Modeling Languages and Solver Interfaces (Sandton Sun: Oleander) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18ME-18: Survivable Telecommunications Network Design (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19ME-19: Maintenance 1 (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Tuesday, 8:00 AM - 9:30 AM

TA-01: Combinatorial Optimization: Recent Advances in Scheduling (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . 20TA-02: Integer Programming (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20TA-03: Optimization Modeling for the Public Sector (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20TA-04: Sports and OR (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20TA-05: OR Education 1 (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21TA-07: Economic Development: Modelling and Finance (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21TA-08: Multicriteria Applications (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22TA-09: Facility Location and Routing (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22TA-10: Project Scheduling: Resource Allocation (Convention Centre: Boardroom 6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22TA-11: Operations Management: Health Care (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23TA-12: Finance: Revenue management and pricing (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23TA-13: Finance: Auctions (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23TA-14: DEA: Applications in transport (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24TA-15: DEA: Applications in industry (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24TA-16: Container Storage Planning (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24TA-17: Software Demos I (GAMS/AMPL) (Sandton Sun: Oleander) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25TA-18: Telecommunications Trees and Flows (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25TA-19: Maintenance 2 (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Tuesday, 10:00 AM - 11:00 AM

TB-01: Supply Chain Management (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Tuesday, 11:30 AM - 1:00 PM

TC-01: Combinatorial Optimization: Data mining and clustering (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . 26TC-02: Discrete Optimization (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26TC-03: Multicriteria Public Sector Decision-Making (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27TC-04: Applications of OR in Forestry I (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27TC-05: OR Education 2 (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27TC-07: Healthcare policy analysis using simulation (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28TC-08: MCDA and Performance Assessment (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28TC-09: Urban Transportation (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28TC-10: Project Scheduling: Scheduling of Activities (Convention Centre: Boardroom 6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29TC-11: Operations Management: Supply Chains (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29TC-12: Finance: Option pricing (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

116

IFORS 2008 - Sandton SESSION INDEX

TC-13: Finance: Portfolio management (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30TC-14: DEA Theory: New developments 2 (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30TC-15: Economic Development: Strategic Planning and Management (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . 30TC-16: Air Transportation (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31TC-17: Software Demos II (AIMMS/MPL) (Sandton Sun: Oleander) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31TC-18: Meta-heuristics for optimizing telecommunications networks (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . 31TC-19: Quality and Reliability (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32


TD-01: Advanced Topics in Combinatorial Optimization (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32TD-02: Optimization Heuristics and Applications (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32TD-03: Complex Public Sector Social Problems (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33TD-04: Wine Supply Chain (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33TD-05: Bringing the real world of OR into the classroom 1 (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34TD-07: Efficient use of core hospital resources (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34TD-08: MCDA/SMAA (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34TD-09: Heuristic Approaches to Routing (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34TD-11: Supply Chain Optimization (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35TD-13: Finance: Risk analysis and management (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35TD-14: DEA Theory: New developments 3 (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35TD-15: Economic Development: General Topics (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36TD-16: Maritime logistics (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36TD-19: Quality and Availability (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36


TE-01: Minimax Stochastic Optimization with Applications (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37TE-03: Public Sector Transport and Management (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37TE-04: Agribusiness Supply Chain (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38TE-05: Bringing the real world of OR into the classroom 2 (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38TE-07: Advanced Modeling Methods in Health Care (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38TE-08: Multiple Criteria Decision Analysis (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39TE-09: Routing Applications (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39TE-10: Scheduling Applications 1 (Convention Centre: Boardroom 6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39TE-13: Finance: Markov models and Valuation modelling (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . 40TE-14: DEA Theory: Recent developments in nonparametric frontier estimation (Sandton Sun: Maroela North) . . . . . . . . . . . . 40TE-16: Information technology in logistics (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40TE-18: Wireless Telecommunications Networks (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41


RA-02: Optimization Methods (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42RA-03: Transportation Systems: Traffic Assignment Modelling (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . 42RA-04: Applications of OR in Forestry II (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42RA-07: Health Care for the Community (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43RA-08: Visual Aspects of MCDA (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43RA-09: Game Theory and Applications 1 (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43RA-10: Creating and Building IFORS (Convention Centre: Boardroom 6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44RA-11: Supply Chain Risk, Pricing, and Coordination (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44RA-12: Finance: Finance and banking (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44RA-14: DEA: Applications in education (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45RA-16: Location and Layout (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45RA-19: Queues 1 (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45


117

SESSION INDEX IFORS 2008 - Sandton

RB-01: Humanitarian Logistics (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

Thursday, 11:30 AM - 1:00 PM

RC-02: Dynamic Programming (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46RC-03: Transportation Systems: Supply and Demand (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46RC-04: Optimization for sustainable production 1 (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46RC-07: Humanitarian Logistics (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47RC-08: Decision support systems for economic or environmental issues (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . 47RC-09: Game Theory and Applications 2 (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48RC-10: The Present State of IFORS (Convention Centre: Boardroom 6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48RC-11: Supply Chain Applications (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48RC-12: Finance: Financial modelling (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48RC-13: Scheduling Applications 2 (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49RC-14: DEA Theory: New developments 4 (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49RC-15: OR for Sustainable Development 1 (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49RC-16: Locational Planning (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50RC-18: Cutting and Packing 1 (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50RC-19: Queues 2 (Sandton Sun: Acacia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50


RD-02: Convex Optimization (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51RD-03: Arc Routing (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51RD-04: Optimization for sustainable production 2 (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52RD-05: OR Applications to Education (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52RD-06: Data Mining 1 (Convention Centre: Boardroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52RD-07: Humanitarian Logistics (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53RD-09: Game Theory and Applications 3 (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53RD-10: The Future of IFORS (Convention Centre: Boardroom 6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53RD-13: Multicriteria Scheduling (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54RD-14: DEA: Applications in farms and agricultural (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54RD-15: OR for Sustainable Development 2 (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54RD-16: Pickup and Delivery Routing (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55RD-17: Energy: General (Sandton Sun: Oleander) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55RD-18: Cutting and Packing 2 (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55


RE-02: Knapsack Problems and Extensions (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56RE-03: Military Simulation and Stochastic Models (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56RE-04: Job Shops and Assembly Lines (Convention Centre: Ballroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57RE-05: Human Resources Management (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57RE-06: Data Mining 2 (Convention Centre: Boardroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57RE-07: Improving policy in healthcare (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58RE-09: Decision Analysis: Methods and Processes (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58RE-11: Production Planning & Control (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58RE-13: Logistic Scheduling (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59RE-14: DEA Theory: New developments 5 (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59RE-15: Systems Approaches to Public Sector Problems (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59RE-16: Applications in logistics (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60RE-18: Cutting and Packing 3 (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Friday, 8:00 AM - 9:30 AM

FA-01: Stochastic Models (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61FA-02: Nonlinear Optimization (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

118

IFORS 2008 - Sandton SESSION INDEX

FA-03: Military Threat Evaluation / DSS (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61FA-05: OR and Ethics (Convention Centre: Boardroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62FA-06: Data Mining and Knowledge Management (Convention Centre: Boardroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62FA-07: Fuzzy Computing and Its Applications (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62FA-08: Decision Support Systems 1 (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63FA-09: Group decision making and voting (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63FA-10: Regulatory Networks, Inhibitors and Drug Design (Convention Centre: Boardroom 6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63FA-11: Production Lot Sizing (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63FA-12: Finance: Emissions trading and power markets (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . 64FA-13: Just-in-Time Scheduling (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64FA-14: DEA: Applications in finance and banking (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64FA-16: Urban transportation: planning and modeling (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65FA-17: Energy: Policy and Planning (Sandton Sun: Oleander) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65FA-18: Cutting and Packing 4 (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Friday, 10:00 AM - 11:30 AM

FB-01: Markov Models and Brownian Motion (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66FB-02: Optimization and Local Search (Convention Centre: Ballroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66FB-03: Military Metrics (Convention Centre: Ballroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67FB-06: E-Commerce (Convention Centre: Boardroom 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67FB-07: Fuzzy Decision Analysis and Preference Structures (Convention Centre: Boardroom 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67FB-08: Decision Support Systems 2 (Convention Centre: Boardroom 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67FB-09: Multi-objective decision-making (Convention Centre: Boardroom 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68FB-10: Modelling and Prediction of Genes, Proteins and Their Structure (Convention Centre: Boardroom 6) . . . . . . . . . . . . . . . . 68FB-11: Combinatorial optimization: models and algorithms (Convention Centre: Committee Room 2) . . . . . . . . . . . . . . . . . . . . . . . 68FB-12: Finance: Risk management in commodity markets (Convention Centre: Committee Room 4) . . . . . . . . . . . . . . . . . . . . . . . . 69FB-13: Scheduling Independent tasks (Convention Centre: Committee Room 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69FB-14: Problem Structuring and Solving (Sandton Sun: Maroela North) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69FB-15: Economic Modeling (Sandton Sun: Maroela South) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70FB-16: Urban Transit Planning and Operations (Sandton Sun: Jacaranda) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70FB-17: Energy: Optimization (Sandton Sun: Oleander) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70FB-18: Optimization and Metaheuristics (Sandton Sun: Syringa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Friday, 12:00 PM - 1:00 PM

FC-01: Bringing Computers to the World (Convention Centre: Ballroom 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

119

Monday, 8:00 AM - 9:30 AMStream: Combinatorial Optimization Invited session Chair: Dominique de...

Documents

Transcript of Monday, 8:00 AM - 9:30 AMStream: Combinatorial Optimization Invited session Chair: Dominique de...