M. Tyndel, Graduate Lecture – Project managementFeb 21 st, RAL Project Management Final student...

M. Tyndel, Graduate Lecture – Project management Feb 21st , RAL

Project Management

Final student lecture in RAL series ‘Port & cigars after the meal?’

What is project management Text book definition Relevance to big science

Examples of projects PhD thesis Detector R&D

e.g. Calorimetry RD

Detector construction e.g. ATLAS SCT

Project management concepts & tools Scope, Cost, Time Risk

Aim: – introduce the concepts a

nd the language

Disclaimer –

I have no PM qualifi

cations (apart f

rom some experie

nce)


Who needs project management - Eurotunnel?


Who needs project management - Eurofighter?

85. MoD's memorandum stated:[129] Typhoon (formerly Eurofighter) is an agile fighter aircraft that will serve as the cornerstone of the RAF's future fighting capability…..

The in-service date for Typhoon (defined as the date of delivery of the first aircraft to the RAF ) was achieved in 2003—some 54 months late. The current forecast cost of Typhoon is £19,018 million, compared to £16,670 million approved at Main Gate.

http://www.publications.parliament.uk/pa/cm200304/cmselect/cmdfence/572/57206.htm#note129


Who needs project management - ATLAS?

There are scientific & technical uncertainties with scientific projects. Large projects with many partners or stakeholders are difficult to control It is “not acceptable” to screw up large projects involving public money


What is project management?..if you have no idea start with

http://en.wikipedia.org/wiki/Project_management

Project management is the discipline of organizing and managing resources in such a way that these resources deliver all the work required to complete a project within defined scope, time, and cost constraints. A project is a temporary and one-time endeavor undertaken to create a unique product or service. This property of being a temporary and a one-time undertaking contrasts with processes, or operations, which are permanent or semi-permanent ongoing functional work to create the same product or service over and over again. The management of these two systems is often very different and requires varying technical skills and philosophy, hence requiring the development of project management.

The first challenge of project management is ensuring that a project is delivered within the defined constraints. The second, more ambitious, challenge is the optimized allocation and integration of the inputs needed to meet those pre-defined objectives. The project, therefore, is a carefully selected set of activities chosen to use resources (time, money, people, materials, energy, space, provisions, communication, quality, risk, etc.) to meet the pre-defined objectives.

http://en.wikipedia.org/wiki/Project_management


What is project management?

Time

Scope

Cost

Risk


Risk

Consider what the risks are & document A “risk register”

Analyse the impact Documented in risk register

Mitigate Have an escape route

Control Regular reviews & reporting

Note – Auditors like to ‘quantify risk’ Definition = probability x impact


Example 1 – PhD thesis

Scope Gain PhD - become qualified as a researcher

Further ones education through post-graduate courses Undertake original research and publish results Pass final exam (“viva”) to gain PhD

Time Externally imposed constraints

Total time = 3 ( or 4 years) Course work for 1 year in first year Research time ~ 18months can depend on several factors

1. Others providing equipment, data ….2. Having the required knowledge and expertise etc

Cost Well defined for this example with

Salary agreed in advance for the 3 years Research tools (computing, lab space, travel…) provided by the University

Risks Having an “inadequate supervisor” or an “inappropriate” project Not knowing how much work is needed to complete research Personal issues – illness


PhD example - Gantt

Gantt – A tool to list tasks, show dependencies & make resources explicit Tool = Microsoft Project

ID Task Name Duration Start Predecessors

1 PhD start 0 mons Mon 02/10/06

2 Lectures 9 mons Mon 02/10/06 1

3 Technical project 12 mons Mon 19/03/07 2FS-3 mons

4 1rst year exam/review 0 mons Fri 31/08/07 1FS+12 mons

5 Physics analysis 18 mons Mon 03/09/07 4

6 Write-up 21 mons Mon 26/11/07

7 Introduction 1 mon Mon 26/11/07 3SS+3 mons,5SS+3 mons

8 Technical project 3 mons Mon 18/02/08 3

9 Analysis 4 mons Mon 19/01/09 5

10 Final corrections 2 mons Mon 11/05/09 9

11 "Float" 3 mons Mon 06/07/09 6

12 PhD complete 0 mons Fri 25/09/09 11


PhD example – Gantt/Critical path

ID Task Name

1 PhD start

2 Lectures

3 Technical project

4 1rst year exam/review

5 Physics analysis

6 Write-up

7 Introduction

8 Technical project

9 Analysis

10 Final corrections

11 "Float"

12 PhD complete

02/10

Me

Me

31/08

Me

Me

Me

Me

Me

Me

25/09

Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1Year 1 Year 2 Year 3 Year 4 Year 5 Year 6

Critical path analysis – shows in red the tasks which determine the end date


PhD example Gantt/Resource summary

50%

100%

150%

200%

250%

300%

Peak Units:

Me Overallocated: Allocated:

Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4Year -1 Year 1 Year 2 Year 3 Year 4 Year 5

100% 200% 200% 200% 300% 200% 200% 100% 100% 100% 100% 100%


PhD example - summary

This project is relatively “simple” because of the small number of independent tasks and people involved.

In “project management speak”

1. Analysed the project and split it into work-packages (WP)

2. Estimated the time needed for each WP and the overall time

3. Documented the project to enable the stakeholders to agree to the plan Stakeholders are – the student, the supervisor/university and the funding

agency/PPARC

4. Estimated the resources needed ~5 fte years of student effort (3 years available!)

The full economic cost would = salary, equipment, computing, travel (typically 2-3 x salary)

5. Introduced contingency by planning to complete the work early


PhD example - summary

Risks: Analysed the risks (examples)

An “inadequate supervisor” or an “inappropriate” project Probability low; impact high

Action: Review at end of first year

Research can be open-ended i.e. difficult to estimate how much work is needed Probability high; impact medium

Actions:

2 research topics for thesis (1 technical; 1 analysis)

Agree to restrict scope of research to time available

Avoid “mission creep” i.e. stop investigating at appropriate time

Factor in a time contingency

Personal issues – illness Probability low; impact medium

Action: None planed – escape route would be to apply for more funding


return to Wikipedia Project Management Activities

Project Management is composed of several different types of activities such as:

1. Planning the work or objectives 2. Analysis & Design of objectives 3. Assessing and controlling risk (or Risk management) 4. Estimating resources 5. Allocation of resources 6. Organizing the work 7. Acquiring human and material resources 8. Assigning tasks 9. Directing activities 10. Controlling project execution 11. Tracking and Reporting progress 12. Analyzing the results based on the facts achieved 13. Defining the products of the project 14. Forecasting future trends in the project 15. Quality Management 16. Issues Management 17. Issues solving 18. Defect prevention 19. Project Closure meet


Wikipedia

Project management artefacts

Most projects, to be successful, must adequately document objectives and deliverables. These documents are a mechanism to align sponsors, clients, and project team's expectations.

1. Project Charter 2. Business case / Feasibility study 3. Scope statement / Terms of reference 4. Project Management plan / Project Initiation Document 5. Work Breakdown structure 6. Change Control Plan 7. Risk management plan 8. Communications Plan 9. Governance Model 10. Risk Register 11. Issue Log 12. Action Item List 13. Resource Management Plan 14. Project schedule 15. Status Report 16. Responsibility assignment matrix 17. Database of risks 18. Database of lessons learned 19. Stakeholder Analysis

These documents are normally hosted on a shared resource (i.e., Intranet web page) and are available for review by the project's stakeholders. Changes or updates to these documents are explicitly outlined in the project's configuration management (or change control plan).


Project control variables Wikipedia

Project Management tries to gain control over variables such as risk:

1. Risk is defined as potential points of failure. Most negative risks (or potential failures) can be overcome or resolved, given enough planning capabilities, time, and resources. According to some definitions (including PMBOK Third Edition) risk can also be categorized as "positive--" meaning that there is a potential opportunity, e.g., complete the project faster than expected.

2. Customers (either internal or external project sponsors), external organizations (such as government agencies and regulators) can dictate the extent of three variables: time, cost, and scope. The remaining variable (risk) is managed by the project team, ideally based on solid estimation and response planning techniques. Through a negotiation process among project stakeholders, an agreement defines the final objectives, in terms of time, cost, scope, and risk, usually in the form of a charter or contract.

3. To properly control these variables a good project manager has a depth of knowledge and experience in these four areas (time, cost, scope, and risk), and in six other areas as well: integration, communication, human resources, quality assurance, schedule development, and procurement.


Example 2 – Calorimetry R&D for Linear collider

Scope Develop ‘active pixel sensors’ as a tool for a ‘particle flow’ approach to calorimetry Break down the project into work-packages

1. Physics requirement and specification

2. Active pixel design

3. Active pixel evaluation

4. Evaluate prototype calorimeter module in test-beam

Time Defined by requirement for concept to be proven for LC TDR in 2010

Limits scope of R&D

Cost Cost = procurement, manpower (measured in fte) and travel

Risks Failure or delay in any one work-package causes the project to fail

Procurement costs exceed estimates Management contingency (held by PPARC)

Insufficient or loss of expert manpower Regular progress reviews


Particle Flow Algorithm for calorimetry

ECAL

HCAL

Tracker


Active Pixel Sensors for Calorimetry

CMOS active pixel sensors are fully integrated sensors and electronics

RD project is to develop a device which is sensitive to tracks and has very fine granularity:

Provide calorimetry in the usual way by counting tracks and all single tracks to be identified and measured precisely


Example 2 RD for silicon sensors for CALICE


Example 2 – CALICE example

Project Management activities: PPARC requirements

1. Planning the work or objectives Project description and plan

2. Analysis & Design of objectives 3. Assessing and controlling risk Risk register

4. Estimating resources Grant resource request with FEC

5. Allocation of resources Defined resource sharing between WPs & institutes

6. Organizing the work Set up both a WP & an institute organisation

7. Acquiring human and material resources 8. Assigning tasks 9. Directing activities Some combination of - PI, Spokesman, PM

10. Controlling project execution 11. Tracking and Reporting progress Regular reporting to Over-sight committee 12. Analyzing the results based on the facts achieved 13. Defining the products of the project 14. Forecasting future trends in the project 15. Quality Management ISO 9000 for engineering 16. Issues Management 17. Issues solving 18. Defect prevention 19. Project Closure meet


CALICE example - summary

Differences from Example-1 (PhD)

Scope of project is initially defined from within project Scope can be modified by funding body

Project is explicitly broken into sub-projects or work-packages Different people in the individual work-packages Understand interfaces between work-packages Introduce reviews to monitor and control work-packages

Several institutes/groups involved Needs an organisational structure Needs a decision taking mechanism

Project resources are controlled externally (but managed internally)


Example 3 – The ATLAS Silicon Tracker (SCT)

Scope Design and build tracker for a general purpose detector for LHC

Again scope was initially defined from within the ATLAS project Scope evolved and was modified on the basis of R&D Scope modified in the context of overall detector optimisation Scope modified by resources and expertise available.

Time Schedule evolved over the first few years

Bottom-up: Time needed to develop technical solutions specifically for SCT Top-down: Constraints from the LHC framework

Cost Total funding & resources available were a complicated constraint

Funding from 11 separate funding agencies Individual profiles and procedures to be followed

Risks Technical e.g. at start-up no radiation sensors or readout available Organisational – many work-packages and funding agencies Financial – no margin for cost over-runs People: Maintaining coherence with a large team over a long time


SCT tracker projects – difference from above examples

Scale of the projectsPhysically large

~ 105 separate components

Technically complex Many R&D programs – sensors, ASICs, Readout, Materials….

But with strong interfaces

Resources required £20M of purchases ~2,000 fte of in-house effort from 40 institutes

Management complexReporting to 11 funding agencies and to the overall ATLAS projectTaking technical decisions between 40 institutes (200 physicists)Sub-dividing and organising workPeople!


Example 3 – The ATLAS Silicon Tracker


Design choices are fixed by physics requirements.

Sounds simple but, in general, an increase in performance improves the physics and an increase in performance costs….

Performance variables include: Number , size & position of the

detecting elements Measurement precision Transparency of the tracker

(%X0)

Cost & constraints include: Resources – finite and fixed Time available - fixed Technology available (or likely

to be available) - constraint

Tracker Design Choices

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

120.0%

0 0.5 1 1.5 2 2.5

Cost

Pe

rfo

rma

nc

e

Res

ourc

e co

nstr

aint

9. “The best is the enemy of the good” – Voltaire


System optimisation 1. No. of measurements fixed = 4 2. Layout to get evenly spaced points

with barrel/endcap split at 450

3. Opted for 4 perfect, hermetic layers. 4. Detailed design was “Bottom-up”

starting from sensors/ASICs

Advantages: Minimised silicon area Provided overlaps for alignment

Cost: Complexity of the design &

assembly High cost of perfect components

(>99%) High cost of building ‘perfect (i.e.

99% good channels)’ modules Complexity of services

Example-1 System choices : Layout & material

“Let no one ignorant of geometry enter” – Plato


In 1990s no sensors had the required performance:

GaAs investigated because of anticipated radiation tolerance

MSGCs investigated because of anticipated lower cost

Silicon strip options considered n-in-p (inversion) double-sided (material, cost) p-in-n

DC coupled AC coupled

6” wafers or 4” wafers Oxygenated

Close collaboration with industry was the key to success.

Sensors with strip yield close to 100% & delivered to agreed schedule

Example-2 Technical choices : Sensors


In 1990s no proven radiation hard technology available with the required performance.

Analogue de-convolution to get speed

Digital 2 chip-set

Binary 2 chip set Bi-CMOS ABCD

ABCD3-T

During 1990s may radiation hard foundries closed and there was the great discovery that deep sub-micron processes were radiation hard.

Production of ASICs on specialised process was ‘tough’ and yield ~ 26%

Example-3 Technical choices : ASICs


ATLAS SCT tracker


ATLAS SCT Schedule

1997: resources fixed

Evolution of schedule end-date complicated decision making

“The success of most things depends on knowing how long it will take to succeed” – Montesquieu


CMS tracker assembly organisation


Once upon a time there was a Red rowing team.

People


This Red team agreed to hold an annual rowing race with a Green team. Each team would contain 8 men.


Both teams worked really hard to get in the best shape. On the day of the fi rst race, both teams were ready to win.


FINISHFINISH

The Green team won by 1 mile!


The Red team was crushed in their defeat, but they were determined to win the race next year. So they established a panel of auditors to observe the situation and ascertain if there were any diff erences

between the teams.


Af ter several weeks of detailed intelligence gathering, the auditors could fi nd only one diff erence; the Green team had 7 rowers and 1

captain...


… and the Red team had 7 captains and 1 rower!


Un-perplexed by the raw data, upper management showed unexpected wisdom: they hired a consulting company to analyze the data and suggest a

solution that would enable the Red team to win next year.


Like sharks getting the scent of reorganization blood, upper management wasted no time in restructuring the Red team into

4 Captains, led by 2 Managers, reporting to 1 Senior Director with a dotted line to the rower. Besides that, in a blaze of unrestricted inspiration, they suggested they might be inclined to improve the

rower’s working environment by a non-monetary reward and recognition scheme if there was improved perf ormance by the

rower.


FINISH

The next year, the Green team won by 2 miles.......


The Red team upper management immediately fi red the rower based on his unsatisf actory perf ormance.


Af ter several months the consultants came to the conclusion that the ratio of captains to rowers was the problem in the Red team.

Based on this analysis a solution was proposed: the structure of the Red team has to be

changed!


A bonus was paid to the Captains, Directors, and Managers f or the strong leadership and motivation they showed during the preparation

phase and as an incentive f or them to find a better rower f or the next race.


The consulting company prepared a new analysis of the restructuring activity, which showed that the strategy was good, the motivation was great, the restructuring

was executed correctly, but the tool used (which was not included in the original data) was sub-standard and had to

be improved.


Currently the Red team management is having a new boat designed;

and to demostrate fiscal and HR dexterity for stockholders they also outsourced the rowing to India.


Summary - Project Management

‘Port and Cigars’ analogy PM lecture concludes the main meal

of a series of lectures on experimental techniques.

An opportunity to hear the words and to reflect on what is needed to achieve project.

“…and now you’ve heard it before”

I hope that1. it gives you confidence to learn by

tryingor

2. It encourages you to take a real course

e.g. http://www.prince2.com/whatisp2.html#processmodel

http://www.prince2.com/whatisp2.html#processmodel


..but it is more fun to look at Wikiquotes

"The more you plan the luckier you get. "

"If it can go wrong it will - Murphy's law. “

"Anything that can be changed will be changed until there is no time left to change anything.“

"Work expands to fill the time available for its completion - Parkinson's law.“

"A minute saved at the start is just as effective as one saved at the end."

"A little risk management saves a lot of fan cleaning." "Activity is not achievement."

"The sooner you get behind schedule, the more time you have to make it up.“

"Any project can be estimated accurately (once it's completed)."

"There's never enough time to do it right first time but there's always enough time to go back and do it again."

M. Tyndel, Graduate Lecture – Project managementFeb 21 st, RAL Project Management Final student...

Documents

Transcript of M. Tyndel, Graduate Lecture – Project managementFeb 21 st, RAL Project Management Final student...