COMPUTERIZED MULTIPLE CRITERIA DECISION MAKING MODEL FOR PROJECTS PLANNING AND IMPLEMENTATION By...

COMPUTERIZED MULTIPLE COMPUTERIZED MULTIPLE CRITERIA DECISION CRITERIA DECISION MAKING MODEL FOR MAKING MODEL FOR

PROJECTS PLANNING AND PROJECTS PLANNING AND IMPLEMENTATIONIMPLEMENTATION

By By MOHAMMED FASIUDDIN MOHAMMED FASIUDDIN

ID # 220324ID # 220324Semester-021Semester-021

1111thth January, 2003 January, 2003

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

Presented toPresented toDr. S. ASSAFDr. S. ASSAF

ContentContent

Introduction Introduction Problem StatementProblem StatementObjectives, Scope & LimitationObjectives, Scope & LimitationMethodologyMethodology

Previous StudiesPrevious Studies Analytical Hierarchy Process (AHP) Analytical Hierarchy Process (AHP) Decision Criteria For The Selection Of Project Decision Criteria For The Selection Of Project

Alternatives Alternatives Computerized Decision Making Model Computerized Decision Making Model Project Alternatives Evaluation Problem- Case Study Project Alternatives Evaluation Problem- Case Study Conclusion And Recommendation Conclusion And Recommendation

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

IntroductionIntroduction

Projects are implemented to achieve certain goals, which Projects are implemented to achieve certain goals, which includesincludesIncreasing company’s profit Increasing company’s profit Enhancing its competitiveness to survive in future Enhancing its competitiveness to survive in future

For the projects to be successful the company must come For the projects to be successful the company must come up with best alternatives during preliminary engineering up with best alternatives during preliminary engineering study when making decision, but the decision making study when making decision, but the decision making team is faced with a problem as more than one objective team is faced with a problem as more than one objective needs to be satisfied needs to be satisfied

To satisfy these objectives simultaneously order of To satisfy these objectives simultaneously order of preference is a major factor preference is a major factor

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Problems that might be encountered during decision Problems that might be encountered during decision making process may include: making process may include:

Complexity of the decisionComplexity of the decision

Inconsistency of the decision makerInconsistency of the decision maker

Political favor and hidden agendaPolitical favor and hidden agenda

Overlooking objective of the projectOverlooking objective of the project

Conflicting between the individualsConflicting between the individuals

Variation of perception from one individual to Variation of perception from one individual to anotheranother

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


To avoid these problems and improve the decision making To avoid these problems and improve the decision making process, a structured and comprehensive Computerized process, a structured and comprehensive Computerized Multi-Criteria Decision Making Model based on Multi-Criteria Decision Making Model based on Analytical Hierarchy Process (AHP) that assist in Analytical Hierarchy Process (AHP) that assist in selecting the best alternativeselecting the best alternative

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

Problem StatementProblem Statement

During preliminary engineering phase of any project, there During preliminary engineering phase of any project, there are an array of possible alternative solutions to any are an array of possible alternative solutions to any project, deciding on which alternative to select is difficult.project, deciding on which alternative to select is difficult.

Owners often do not consider alternatives to their Owners often do not consider alternatives to their

investment projects, if they do; they may compare only investment projects, if they do; they may compare only costcost

Projects involve environment, political and other intangible Projects involve environment, political and other intangible factors, which are usually ignored in the cost benefit factors, which are usually ignored in the cost benefit analysis. analysis.

Conflict may arise between the decision maker because of Conflict may arise between the decision maker because of their different concerns and goals their different concerns and goals A

RE

-520

A

dvan

ce c

onst

ruct

ion

and

Mai

nten

ance

Mod

elin

g

Objective, Scope & Limitation Objective, Scope & Limitation

The objective of this research is to develop a Computerized The objective of this research is to develop a Computerized Decision Support Model based on Multi-criteria Decision Decision Support Model based on Multi-criteria Decision Making, to assist in decision making. Making, to assist in decision making.

This model will be applied to a case study to demonstrate This model will be applied to a case study to demonstrate its feasibility. Factors considered in the decision model its feasibility. Factors considered in the decision model will be the factors that influence the decision making with will be the factors that influence the decision making with regard to the system selection and project regard to the system selection and project implementation.implementation.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

MethodologyMethodology

Determine the evaluation criteria for selecting Determine the evaluation criteria for selecting projects alternatives from literature review and projects alternatives from literature review and documentation of previous projects. documentation of previous projects.

Present the selected method and discuss its Present the selected method and discuss its mathematics and applications mathematics and applications

Develop a computerized model based on the Develop a computerized model based on the selected method selected method

Apply the computerized modelApply the computerized model on real project on real project

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

Previous Studies Previous Studies

Al-Sughaiyer conducted a study, on public construction Al-Sughaiyer conducted a study, on public construction projects in Saudi Aramco. In the value engineering projects in Saudi Aramco. In the value engineering study, alternatives were compared by using weighted study, alternatives were compared by using weighted evaluation to help in selecting the best alternative evaluation to help in selecting the best alternative

This method comprised of two steps:This method comprised of two steps:Paired comparison criteria weighting process: Under Paired comparison criteria weighting process: Under this criteria influencing the decision making are listed & this criteria influencing the decision making are listed & assigned a letter depending upon its importance to the assigned a letter depending upon its importance to the decision maker. decision maker.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

Previous StudiesPrevious Studies

When selecting between two criteria, degree of importance When selecting between two criteria, degree of importance of one over another can be:of one over another can be:Major Major (3 points)(3 points)Minor Minor (1 point)(1 point)Minor Minor (1 point)(1 point)No Preference No Preference (0 point)(0 point)

Raw score of all the criteria is then adjusted to a scale of Raw score of all the criteria is then adjusted to a scale of 1-10 with, 10 assigned to the criteria with the highest 1-10 with, 10 assigned to the criteria with the highest raw score raw score

The evaluation matrix: Once the criteria elements and The evaluation matrix: Once the criteria elements and their weights have been established, they are entered in their weights have been established, they are entered in Evaluation Matrix as in Evaluation Matrix as in fig2fig2, then each criterion is , then each criterion is ranked against each alternative.ranked against each alternative.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


A scoring of 1-5 is used as follows:A scoring of 1-5 is used as follows:Excellent Excellent = 5= 5Very GoodVery Good = 4= 4GoodGood = 3= 3FairFair = 2= 2PoorPoor = 1= 1

Rank of each alternative is multiplied with weight of criterion Rank of each alternative is multiplied with weight of criterion and the result is entered in the matrix. The total score is and the result is entered in the matrix. The total score is summed up for each alternative and ranked for selection. summed up for each alternative and ranked for selection. The alternative with the highest total score is selected The alternative with the highest total score is selected

This method has a drawback because of its 5-point scale, This method has a drawback because of its 5-point scale, therefore can handle only a few or limited number of therefore can handle only a few or limited number of decision criteriadecision criteria

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Zedeh in his Fuzzy Set Theory using Fuzzy Multi-criteria Zedeh in his Fuzzy Set Theory using Fuzzy Multi-criteria concept comparing projects, by assigning values a concept comparing projects, by assigning values a membership from 0 to 1 in the sets, where 1 indicates membership from 0 to 1 in the sets, where 1 indicates membership in the set and 0.5 means that it is equally membership in the set and 0.5 means that it is equally likely to be in set or out of the set. likely to be in set or out of the set.

Grandzol and Gershhon in their study entitled Multiple Grandzol and Gershhon in their study entitled Multiple Criteria Decision Making; developed Criteria for Criteria Decision Making; developed Criteria for evaluating the alternatives by assigning weights and evaluating the alternatives by assigning weights and ranks. The team used Electra Technique which ranks. The team used Electra Technique which compares a pair of alternative actions and ranks them by compares a pair of alternative actions and ranks them by weighted scores for criteria for which a given alternative weighted scores for criteria for which a given alternative action is better. action is better.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Alternatives actions that are better in the weighted Alternatives actions that are better in the weighted criterion and not too much worse in the other criteria criterion and not too much worse in the other criteria rank highest. The study team decided on 0.8 level of rank highest. The study team decided on 0.8 level of concordance and a 0.2 level of discordance for concordance and a 0.2 level of discordance for alternative action to qualify. alternative action to qualify.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

ANALYTICAL HIERARCHY PROCESS ANALYTICAL HIERARCHY PROCESS (AHP) (AHP)

Introduction: It is difficult to assign and estimate the Introduction: It is difficult to assign and estimate the weight of each attribute, as the decision making is weight of each attribute, as the decision making is based on subjective criteria, which cannot be based on subjective criteria, which cannot be qualified in the utility function. qualified in the utility function.

Goal Programming lacks a systematic approach to set Goal Programming lacks a systematic approach to set priorities and trade off among objectives. Fuzzy set priorities and trade off among objectives. Fuzzy set theory is difficult to implement. theory is difficult to implement.

To overcome short-comings from Goal Programming & To overcome short-comings from Goal Programming & Fuzzy set theory, the AHP was developed. Fuzzy set theory, the AHP was developed.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

ANALYTICAL HIERARCHY PROCESS ANALYTICAL HIERARCHY PROCESS (AHP)(AHP)

The Decision Model OutlineThe Decision Model OutlineUnderstand clearly the scope of the project Understand clearly the scope of the project Define the main objective of the project Define the main objective of the project Determine the project alternativesDetermine the project alternatives

Literature review Literature review Market surveys Market surveys

Determine all the criteria that influence the decisionDetermine all the criteria that influence the decisionBrainstorming sessions Brainstorming sessions Questionnaires Questionnaires

Group the criteria that are related Group the criteria that are related Use the AHP methodology to rank the project Use the AHP methodology to rank the project alternative alternative

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


The AHP Steps The AHP Steps

Construction of decision hierarchy by breaking Construction of decision hierarchy by breaking down the decision problem into a hierarchy of inter-down the decision problem into a hierarchy of inter-related elements.related elements.

Performing pair-wise comparisons of the decision Performing pair-wise comparisons of the decision elements.elements.

Estimating the weights if the decision elements by Estimating the weights if the decision elements by using Eigen-value method. using Eigen-value method.

Aggregating the weights of the decision elements Aggregating the weights of the decision elements to provide a set of ratings for the decision to provide a set of ratings for the decision alternatives.alternatives.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


The first step, the decision problem is formulated in a The first step, the decision problem is formulated in a hierarchical structure and is broken into a hierarchy of hierarchical structure and is broken into a hierarchy of interrelated decision elements. interrelated decision elements. Figure illustrates Figure illustrates

At the top of the hierarchy lies the most general At the top of the hierarchy lies the most general objective. Number of levels depends on the objective. Number of levels depends on the complexity of the problem and on the degree of complexity of the problem and on the degree of detail. detail.

Each level of the hierarchy contains attributes or Each level of the hierarchy contains attributes or objectives that influence the decision. objectives that influence the decision.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Each relationship is weighted according to the Each relationship is weighted according to the strength of influence an alternative or criterion at strength of influence an alternative or criterion at same level exerts on alternative or criterion same level exerts on alternative or criterion

The most general risky and uncertain the decision The most general risky and uncertain the decision element the higher the level. The elements in element the higher the level. The elements in each level are influenced or controlled by the each level are influenced or controlled by the elements in the level immediately above. elements in the level immediately above.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Influence is distributed downward from the top on nine-Influence is distributed downward from the top on nine-point scale.point scale.

One (1) for equal importance One (1) for equal importance Three (3) for moderate importance of one over the Three (3) for moderate importance of one over the other. other. Five (5) for strong importance of one over the other Five (5) for strong importance of one over the other Seven (7) for very strong importance of one Seven (7) for very strong importance of one elements over the other elements over the other Nine (9) for extreme importance of one elements Nine (9) for extreme importance of one elements over the other. over the other. 2, 4, 6, 8 for compromise.2, 4, 6, 8 for compromise.Reciprocals for the inverse comparison. Reciprocals for the inverse comparison. A

RE

-520

A

dvan

ce c

onst

ruct

ion

and

Mai

nten

ance

Mod

elin

g


The second step involves the pair-wise comparison of The second step involves the pair-wise comparison of the decision elements for each group headed by a main the decision elements for each group headed by a main criterion (node). The comparison is done in pairs and criterion (node). The comparison is done in pairs and placed in matrix A placed in matrix A

In the 3rd step the Eigen-value method is used to In the 3rd step the Eigen-value method is used to estimate the relative weights of the decision elements. estimate the relative weights of the decision elements.

In the last step of the AHP the relative weights of various In the last step of the AHP the relative weights of various levels are aggregated. The results produce a vector of levels are aggregated. The results produce a vector of composite weights, which will serve as a ranking of the composite weights, which will serve as a ranking of the decision alternatives. decision alternatives.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

DECISION CRITERIA FOR THE SELECTION DECISION CRITERIA FOR THE SELECTION OF PROJECT ALTERNATIVES OF PROJECT ALTERNATIVES

Project-Related Criteria Project-Related Criteria Time: defined as the time required to place the Time: defined as the time required to place the system in operation. system in operation. Permit & Approval :This criterion includes Permit & Approval :This criterion includes

waivers for using non-standard equipment or waivers for using non-standard equipment or non-standard installations non-standard installations equipment import permits equipment import permits land use permits land use permits

Performance & Acceptance: This criterion includes Performance & Acceptance: This criterion includes acceptance of the system by the owner, the project acceptance of the system by the owner, the project management team, the end user, and the public.management team, the end user, and the public.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

DECISION CRITERIA FOR THE SELECTION DECISION CRITERIA FOR THE SELECTION OF PROJECT ALTERNATIVESOF PROJECT ALTERNATIVES

Cost : This cost will be broken down into:Cost : This cost will be broken down into:Initial costInitial costOperating and Maintenance costOperating and Maintenance costReplacement costReplacement costSystem upgrading cost System upgrading cost Leasing costLeasing costDecommissioning cost Decommissioning cost

Equipment Related FactorsEquipment Related FactorsOperation Characteristics: Equipment operation Operation Characteristics: Equipment operation characteristics include reliability, availability, and characteristics include reliability, availability, and protection during failure, heat dissipation, power, protection during failure, heat dissipation, power, and security of the system equipment. and security of the system equipment.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Mechanical Characteristics: This constitutes the Mechanical Characteristics: This constitutes the dimensions, physical configuration and the weight of dimensions, physical configuration and the weight of the equipment.the equipment.Compliance : Compliance is the ability of the system Compliance : Compliance is the ability of the system to interface with existing and future communications to interface with existing and future communications systems. systems.

Mandatory features Mandatory features Operational features Operational features

Life & Technology : This consists of Life & Technology : This consists of Working lifeWorking lifeEconomical life Economical life Technological life Technological life

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Vendor Related Factors: This criterion includes the Vendor Related Factors: This criterion includes the vendor’s experience and reputation; the vendor’s ability vendor’s experience and reputation; the vendor’s ability to support its products through warranties, site to support its products through warranties, site maintenance, hot-line support, user training, maintenance, hot-line support, user training,

consulting, and documentation.consulting, and documentation.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

COMPUTERIZED DECISION MAKING COMPUTERIZED DECISION MAKING

MODELMODEL IntroductionIntroduction

The computerized decision making model was created in The computerized decision making model was created in visual basic. The program consist of six modules. visual basic. The program consist of six modules. These are:These are:

Start ModuleStart Module

Initial Data ModuleInitial Data Module

Criteria Pair-Wise Comparison Module Criteria Pair-Wise Comparison Module

Initial Alternatives Data Module Initial Alternatives Data Module

Alternative Pair-Wise Comparison Module Alternative Pair-Wise Comparison Module

Synthesis ModuleSynthesis Module

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

COMPUTERIZED DECISION MAKING COMPUTERIZED DECISION MAKING MODELMODEL

All the input data and output data are automatically All the input data and output data are automatically saved in a micro access file called Alirezam.mbd. saved in a micro access file called Alirezam.mbd.

The input data is plotted immediately after performing The input data is plotted immediately after performing the pair-wise comparison and obtaining the weights the pair-wise comparison and obtaining the weights for the criteria. for the criteria.

The consistency check is done every time the pair-wise The consistency check is done every time the pair-wise

comparison is performedcomparison is performed

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Program Limitations Program Limitations The program is limited to decision problems that have: The program is limited to decision problems that have:

Hierarchies of five level or less Hierarchies of five level or less Criteria groups with ten sub-criteria or less Criteria groups with ten sub-criteria or less

There is no restriction on the no of alternatives, but it is There is no restriction on the no of alternatives, but it is recommended that should not exceed more than ten.recommended that should not exceed more than ten.The Developed Program The Developed Program Flow Chart of the program : The following is the Flow Chart of the program : The following is the

flow chart of the programflow chart of the program. It has 12 steps. Once the . It has 12 steps. Once the program is loaded it initializes the output date files program is loaded it initializes the output date files and then proceeds with the calculations after the and then proceeds with the calculations after the user has input the data. user has input the data.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing


Program Step By Step CalculationProgram Step By Step Calculation

The different The different modules of the computerized model are shown modules of the computerized model are shown

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

PROJECT ALTERNATIVES EVALUATION PROJECT ALTERNATIVES EVALUATION PROBLEM- CASE STUDY PROBLEM- CASE STUDY

Scope In this section, a Scope In this section, a case study case study is presented to is presented to demonstrate the application of the developed demonstrate the application of the developed Computerized Decision Making Model Computerized Decision Making Model

A major industrial company has decided to replace its A major industrial company has decided to replace its current mobile communication system, infrastructure and current mobile communication system, infrastructure and end user equipments with a new state of Art Mobile end user equipments with a new state of Art Mobile Radio System. Radio System.

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

PROJECT ALTERNATIVES EVALUATION PROJECT ALTERNATIVES EVALUATION PROBLEM- CASE STUDYPROBLEM- CASE STUDY

Alternatives: Three mobile radio system alternatives were Alternatives: Three mobile radio system alternatives were investigated. These are investigated. These are

To replace the current system with an analog To replace the current system with an analog systemsystem

To replace the current system with open To replace the current system with open architecture architecture

To replace the current system with proprietary To replace the current system with proprietary system architecturesystem architecture

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

PROJECT ALTERNATIVES EVALUATION PROJECT ALTERNATIVES EVALUATION PROBLEM- CASE STUDYPROBLEM- CASE STUDY

Methodology Applications and DiscussionMethodology Applications and Discussion

Criteria Criteria Pair wise ComparisonPair wise Comparison: In this step the pair wise : In this step the pair wise

comparison was carried out for all the criteriacomparison was carried out for all the criteria

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

CONCLUSION AND CONCLUSION AND RECOMMENDATION RECOMMENDATION

Conclusion: The developed Computerized Multiple Conclusion: The developed Computerized Multiple Criteria Decision-Making Model gives the uses a Criteria Decision-Making Model gives the uses a structured and systematic decision making approach for structured and systematic decision making approach for evaluating and selecting projects alternatives. evaluating and selecting projects alternatives.

Additional research on the developed Computerized Additional research on the developed Computerized Multiple Criteria Decision-Making Model should be Multiple Criteria Decision-Making Model should be conducted. This can include application of this model on conducted. This can include application of this model on the other aspects of project management the other aspects of project management

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

THANK YOUTHANK YOU

AR

E-5

20

Adv

ance

con

stru

ctio

n an

d M

aint

enan

ce M

odel

ing

COMPUTERIZED MULTIPLE CRITERIA DECISION MAKING MODEL FOR PROJECTS PLANNING AND IMPLEMENTATION By...

Documents

Transcript of COMPUTERIZED MULTIPLE CRITERIA DECISION MAKING MODEL FOR PROJECTS PLANNING AND IMPLEMENTATION By...