esd.36j system & project managementdspace.mit.edu/bitstream/handle/1721.1/80702/esd... ·...

+

-

ESD.36J System & Project Management

Dynamics of Project Performance

System Dynamics and Project Management

Lecture #20, SD Class Eight (11/13/00)

Copyright © 2003 James M. Lyneis

Gathering Data+

-

Progress

Rework But, the splitdiscovery can be cannot be measured and determined used to estimate as it is split after the fact happening

Quality Measured

Progress

Rework Discovery

Productivity

Work To Be Done

Undiscovered Rework

Known Rework

Work Really Done

Can be

Copyright © 2003 9/23/03 - ESD.36J SPM James M. Lyneis 2

Topics+

-

• A System Dynamics View of Execution and Adaptation

• Project dynamics summary

• Strategic project management & learning

• Lessons Learned

• Broader issues – multi-project and market interactions


+

-

Selected Issues in Execution & Adaptation

� Managing Risks and Changes: � Schedule adjustments � Staffing strategies

A Strategic View – Deciding in advance the best way to handle problems if

they arise


+

-

In the face of a projected schedule overrun, is it better to slip as soon as you know it or wait and see if corrective actions will help?

Copyright © 20039/23/03 - ESD.36J SPM James M. Lyneis 5

+

- Graph for Staff Level 20

15

10

5

0

No Slip

Slip Early (and all along)

Slip Late

0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 Time (Month)

Staff Level : Class7x Base People Staff Level : Class7x Slip Late People Staff Level : Class7x Slip Early People


+

-Graph for Cumulative Effort Expended

400

300

200

100

0

Slip Early (and all along)

No Slip

Slip Late

0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 Time (Month)

Cumulative Effort Expended : Class7x Base Month*Person Cumulative Effort Expended : Class7x Slip Late Month*Person Cumulative Effort Expended : Class7x Slip Early Month*Person


+

-Conclusions

� If project priorities favor cost over schedule, it’s better to slip the schedule early rather than wait until the damage is done. [Note: if rework creation is high enough, not slipping early can make the project finish later!]

� Use buffers early.


+

-Selected Staffing Issues

� Should staff be added to a late project (Homework 4 question)? � Unless new staff are 100% productive from

day 1, adding staff increases cost; � And, can make project later if the quality hit

is large enough � Should you start with the expected full

team, or staff up gradually?


Topics+

-




• Lessons Learned



+

-

Examples of Behavior Modes on a Project

Project Staffing

Typical Plan

TimeCopyright © 2003

9/23/03 - ESD.36J SPM James M. Lyneis 11

+

-Examples (continued)

Fraction Complete

Time

.5

1

Typical Plan


+

-Examples (continued)

Productivity

Time

1

2 Typical

Plan

(Normalized)


+

-

Project problems persist in spite of numerous advances in the last 30 years

� PERT and CPM (Critical Path Method)� Waterfall, Spiral, …

� Emphasis on “soft,” people factors

� Microsoft Project

.. and Learning is not happening Why???


+

-Why??

Project performance problems are fundamentally dynamic problems,

and Managers mental models and typical tools (computer models): � take a partial view � view a project statically � treat projects as if they were unique


+

-

Systems thinking and dynamic modeling provide the means of ...

... understanding the structure of projects, and how that structure creates behavior; ... designing robust projects

... learning across projects


+

-Dynamics of Project Performance

� The “rework cycle” � Quality � Undiscovered rework

� Feedback effects on productivity and work quality � Positive, re-enforcing, often “vicious

circles” � Negative, controlling

� Knock-on effects between work phases Copyright © 2003


+

- The Rework CycleKey to Project Dynamics

le

WORK

BEING DONE

Peop Productivity Quality

WORK TO BE DONE

KNOWN REWORK

UNDISCOVERED REWORK

WORK REALLY DONE

Rework Discovery


+

-

Work Quality to Date

CompletionTime

Expected Completion

Time Availability

of Prerequisites

Progress

Schedule Pressure

Out-of-Sequence Work

Morale

Hours at Completion

Hours

to Date

Skill & Experience

Hiring

EquivalentStaff on Project

StaffingRequested

TurnoverOrganizational

Changes

Staff

Quality

Added Work

Work

Overtime

Time Remaining

Be Done

Project Dynamics

Scheduled

Perceived

Expected ExpendedProgress

Rework Discovery

Size

Productivity

Obsoleted

Work To

Undiscovered Rework

Known Rework

Work Really Done


+

-

A system dynamics model usually represents several phases of work, but is more aggregate than a CPM model

System Engineering

Software Code and Test

Hardware Design

Hardware Build and Test

Integrate and Test

Software Design


Topics+

-




• Lessons Learned



+

-Corporate Strategy for the Project

Determining the fit of the project to business objectives (the “mission”)� features / scope � schedule milestones (time to market) � delivered quality (defects) � resources & budget And the mix/timing of “projects” necessary to

achieve corporate strategy

Operationally, “projects” implement corporate strategy.


+

-Strategic Project Management

Understanding how project “design” decisions affect project performance …� Organization, process, ...� Buffers, phase overlap, ...� Staffing strategies, schedule slip, ...� ...

… and how they affect other current projects (portfolio issues), and future projects.

Operationally, “day-to-day project decisions” implement project strategy.


+

-

An important part of strategic project management is learning from past

projects ...


+

-

Why Do Organizations Seem So Poor at Learning Lessons From Prior Projects?


+

-Action is a feedback process

RealWorld

Decisions Information Feedback

Goals


+

-

For example, controlling shower temperature

Real World

Reduce hotReduce hot water flowwater flow TemperatureTemperature

too hottoo hotInformationDecisions Feedback

Goals


+

-Or a project ...

Real World

Add peopleAdd people Project behindProject behind scheduleschedule

InformationDecisions Feedback

Goals


+

-

Action is based on mental models …

RealWorld


Strategy, Mental Structure, Models Decision of Real

Rules World


+

-Model for shower control

RealWorld



too hottoo hot

Strategy, Mental Structure, Models

of Real World

Decision Rules

““Reduce hot water flowReduce hot water flow Temperature =Temperature = when temp. too hot”when temp. too hot” hot flow + coldhot flow + cold

flowflowCopyright © 2003


+

-And project control

Real World

Project behindProject behindAdd peopleAdd people scheduleschedule

InformationDecisions Feedback

Strategy,Structure, Decision

Rules

““Add people when weAdd people when we fall behind.”fall behind.”

Mental Models of Real World

Progress =Progress = People *People *

ProductivityProductivityCopyright © 2003


+

-

Learning also results from feedback (“experience”)

RealWorld


Strategy, Mental Structure, Models Decision of Real

Rules World


+

-

Action, and learning, become difficult when there are significant time delays

RealWorld



too hottoo hot

Delay

Strategy,Structure,

Mental Models

““Reduce hot water flowReduce hot water flow

of Real World Temperature =Temperature =

hot flow + coldhot flow + cold when temp. too hot”when temp. too hot” flowflow

9/23/03 - ESD.36J SPM Copyright © 2003 James M. Lyneis 33

Decision Rules

+

-

… and where the cause effect chains are many (“side effects”)

Real World


Add peopleAdd people

Delay Delay Experience dilution, supervision time, Congestion, … Project behindProject behind

scheduleschedule

Learning from experience is difficult

Strategy, Mental Structure, Models

of Real World

Decision Rules

Progress =Progress =““Add people when weAdd people when we People *People *

fall behind.”fall behind.” Productivity * …Productivity * …Copyright © 2003


-

+

-

© 2003

Learning is difficult in a complex world

Mental ModelsStrategy, Structure,Decision Rules

dynamics from cognitive maps


controlled experiments

Adapted from

, etc.

ESD.36J SPM Copyright James M. Lyneis

• Misperceptions of feedback • Unscientific reasoning • Judgmental biases • Defensive routines

• Inability to infer

• Implementation failure • Game playing • Inconsistency • Performance is goal

• Selective perception • Missing feedback • Delay • Bias, distortion, error • Ambiguity

Real World • Unknown structure • Dynamic complexity • Time delays • Inability to conduct

Sterman, Business Dynamics

-

+

-

© 2003

• failure

• • •

goal

• l• •

• Learni l

Decisions

• • • i• •

i i

Mental Models • •

•

• i

• l l• i

Real World • • • •Models

can play a significant role

Adapted from

, etc.


Real World Implementation

Game playing Inconsistency Performance is

Virtual World Perfect imp ementation Consistent incentives Consistent application of decision rules

ng can be goa

Virtual World Complete, accurate, immediate feedback

Real World Selective perception Miss ng feedback Delay Bias, distortion, error

• Amb gu ty

Information Feedback

Mapping of feedback structure Disciplined application of scientific reasoning Discussability of group process, defensive behaviour

Strategy, Structure, Decision Rules

Simulation used to nfer dynamics of cognitive maps correctly

Virtual Worlds • Known structure

Variable evel of comp exity Controlled exper ments

Unknown structure Dynamic complexity Time delays Inability to conduct controlled experiments Sterman, Business

Dynamics

Steps in Learning – Scientific Method

+

-

1. Gather data 2. Develop model(s)3. Validate/Calibrate4. Evaluate lessons: what would have happened if …

� Unplanned events (risks), changes, … did not occur � Management changes had not been made

5. Systematize via data base, benchmarks, rules ofthumb, training, game?

This is not easy and takes effort over multipleprojects on an on-going basis to be successful!


+

-“But my project is different!”

� The rework cycle, with dynamic productivity and work quality, is one generic model (or framework).

� Benchmarking, lessons, and learning can be transferred across projects.


Topics+

-




• Lessons Learned



+

-Lessons -- 1

� Getting a feasible project design (consistent mission) is the first step to avoiding adverse project dynamics

� Prior projects are the best source of information with which to design a robust project

� There is an optimal tradeoff among scope, budget, schedule, and delivered defects

Bid & plan project right


+

-Lessons -- 2

� Recognize the rework cycle and plan to minimize its consequences:� Quality -- do it right the first time � Undiscovered rework -- avoid quality on

quality surprises (estimate real progress) � Rework discovery time -- prioritize rework

detection and correction over new work

It’s the undiscovered rework!


(and minimize mistakes)+

-

Actions to Reduce Rework Discovery Delay

1. Recognizing the existence of the rework cycle and the costs of undiscovered rework is half the battle � Professor Eppinger’s DSM example � F-18


-

+

-

© 2003

Semiconductor Development Example 1 2 3 4 5 6 7 8 9

1 Set customer target • x x • 2 Estimate sales volumes x • x x • 3 Establish pricing direction x • x • 4 Schedule project timeline • x 5 Development methods x • x x x x 6 Macro targets/constraints x x • x x x x 7 Financial analysis x x x x x • 8 Develop program map x • x 9 Create initial QFD matrix x x x x •

10 Set technical requirements x x x x • x 11 Write customer specification x x x x x • O O O O O O O O 12 High-level modeling x x x x • x x x 13 Write target specification x x x x x x x x x • x x 14 Develop test plan x x x x x • x 15 Develop validation plan x x x x • 16 Build base prototype x x x x x x • 17 Functional modeling x x x x x • x x x x x x x x O O O O O O O O O O 18 Develop product modules x x x x x x x x x • O 19 Lay out integration x x x x x x x x x • 20 Integration modeling x x x x x x x • x x x 21 Random testing x x • x x x 22 Develop test parameters x x x x x x x • x x x 23 Finalize schematics x x x x x • x x O O O O O 24 Validation simulation x x x x x x x • x x 25 Reliability modeling x x x x x • x 26 Complete product layout x x x x x • x x 27 Continuity verification x x x x x x • 28 Design rule check x x x • 29 Design package x x x x x • O O O O O O O 30 Generate masks x x x x • x O 31 Verify masks in fab x x x • 32 Run wafers x • x O 33 Sort wafers x • 34 Create test programs x • 35 Debug products x x x x x • O O O O O O O 36 Package products x x x • 37 Functionality testing x x x • 38 Send samples to customers x x x x • 39 Feedback from customers x • 40 Verify sample functionality x • 41 Approve packaged products x x x x • 42 Environmental validation x x x x • 43 Complete product validation x x x x x • 44 Develop tech. publications x x • x x 45 Develop service courses x x • x 46 Determine marketing name x x x x x • x 47 Licensing strategy x x x • 48 Create demonstration x x x x x x • 49 Confirm quality goals x x x x x • 50 Life testing x x x • x x 51 Infant mortality testing x x x x • x 52 Mfg. process stabilization x x x • O O 53 Develop field support plan x x • 54 Thermal testing x x x • 55 Confirm process standards x • x x 56 Confirm package standards x x x x x • x 57 Final certification x x x x x x x x x x x • 58 Volume production x x x • x 59 Prepare distribution network x x x x x x x x • 60 Deliver product to customers x x x x x x x x x •

x = Information Flows = Planned Iterations O = Unplanned Iterations • = Generational Learning

Concurrent Activity Blocks

Potential Iterations

Generational Learning

Sequential Activities


10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

+

-

F-18 Program (on budget and ahead of schedule)

� “I used your work and advice from the start. You made us understand the importance of rework, of finding it early, of reducing it, as well as how we could go about doing it. As a result, we cut our rework effort to one-quarter of what we expected. Not many may know how important your work was, but I do…. Every time someone asked me how we managed to succeed, I take them to the model diagram I keep on my wall, and use that to explain it to them.” Lou Carrier, VP Northrop-Grumman and former F-18 project manager.


+

-F-18 Actions

� Model diagram – clear picture, highlighted importance of rework (minimize “kill the messenger”)

� Integrated product teams, including internal HR consultants to facilitate design-mfg. dialogue

� Real time reporting of staff, costs, progress


+

-

Reducing Rework Discovery Delay (and minimizing mistakes)

1. Recognizing the existence of the rework cycle andthe costs of undiscovered rework is half the battle

2. Make maximum use of integrated product teams,including customers and all functions (PeaceShield, F-18)

3. Focus on quality even if it means delaying interimmilestones or start of downstream work

� Peace Shield (delayed staff roll off; reduced phase overlap)

� Auto (design reviews, prototypes, CAE, physical tests) Corollary: don’t charge ahead on a design phase if upstream

(or supplier) information is not available or of high quality


+

-Lessons -- 3

� Minimize the vicious circles; avoid, within a phase, ...� aggressive staffing

� sustained overtime

� schedule pressure (measuring and rewarding the wrong behaviors)

� ... and between phases � rigidly sticking to scheduled start without

necessary prerequisites (use buffers early)


+

-Lessons -- 4

� Identify key risks and ... � put effort into avoidance and early

detection � alternative processes (e.g. spiral), staffing

strategies (e.g., integrated teams), tools, etc. can improve performance in the right situation, but involve both short-term (implementation) and long-term (greater scope, lower productivity) costs and are not right for every project


+

-Lessons -- 5

� Manage changes and problems wisely � Replan and price changes immediately

(don’t hope you can make it up, then discovery trouble 6 months later)

� Negotiate to reduce scope if schedule is critical


+

-Lessons -- 6

� Learn ...� Identify leverage points and policies � Share across projects (individual managers

learn, but few share)� Reuse what works


+

-

Selling System Dynamics (Modeling)

� Must be a persistent and costly dynamic problem � Illustrate causes (use rework cycle and feedback

examples) � Provide an example of use relevant to your

organization “SDM students are hereby granted permission

to use any of my lecture slides for internalcompany presentations, with appropriateattribution.” James M. Lyneis, 11/13/2003


Topics+

-




• Lessons Learned



+

-Broader Issues

� Is what’s best for the project best for the company?

� Issues in product portfolios � Market and Customer Dynamics -

setting the mission dimension as a part of corporate strategy


+

-

Is What’s Best for the Project Best for the Company?

Examples of conflicting practices -�

�

�


+

-Projects Rarely Operate in Isolation

� While some projects are nearly self-contained with dedicated staff and technical independence, most often projects: � Share staff resources with other projects � Are impacted by technical progress on other

projects, and impact the technical progress on other projects

� And, ...


+

-Projects Rarely Operate in Isolation

� … Sometimes projects are part of “programs” -- groups of projects managed in a co-ordinated way to obtain benefits not available from managing them individually, e.g.: � large development effort broken into pieces

for easier management � build off common platforms � phased upgrades


+

-Issues in Product Portfolios

Portfolio interactions -� staffing and other resources � technical interdependencies

What happens on one project hassignificant knock-on effects to otherprojects. Aggressive projectassumptions (“inconsistentmission”) adversely affect more thanthe one project.


+

-Portfolio Resourcing Issues

� Constraints on Shared Resources � Late and over-budget projects delay ramp-

up of downstream projects � Shared resources (e.g. test facilities) can

also create bottlenecks � Staff working simultaneously on multiple

projects create inefficiencies and delays


+

-

Staff working simultaneously on multiple projects ...

� … may increase total “output” per person as more time is productive ...


+

-

Staff working simultaneously on multiple projects ...

� … but often slows progress on the portfolio of projects: � Productivity reduced by need to refamiliarize

with a project with every shift � Average lead time for an individual’s tasks =

Workload Total / Total Time Available � Even if a person is not overloaded in the

aggregate, there will be some times whenmultiple projects need simultaneous input


+

-Time per project with more projects

j

j

0 1 2 3 4 5

j

j

0 1 2 3 4 5

A. Engineer's value-adding time B. Engineer's time per pro ect

Number of pro ects per engineer

Tota

l tim

e sp

ent o

n va

lue-

addi

ng ta

sks

0%

20%

40%

60%

80%

100%

Number of pro ects per engineer

Tim

e sp

ent p

er p

roec

t on

valu

e-ad

ding

task

s

0%

20%

40%

60%

80%

100%

Adapted from: Smith, Preston G., and Reinersten, Donald G., Developing Products in Half the Time, 2nd Edition, John Wiley & Sons, 1998. Copyright © 2003


Figure by MIT OCW.

+

- Nelson Repenning’s Tipping Point

� Not only are resources constrained, but � Because of technical

interdependencies, failure to adequately complete the first project causes more work and rework on the second project


+

- Repenning’s Model

n l i i

l

i i

i i

Product n+1

Product n+2

ProductConceptDeve opment Product Des gn and Test ng

ConceptDevelopment

ConceptDeve opment

Product Des gn and Test ng

Product Des gn and Test ng

launch dat e for product n-1 launch dat e for product n launch dat e for product n+1

Model Year s Model Year s+1 Model Year s+ 2 Model Year s+ 3

� Two kinds of development work 1 concept development work - take place two years prior to launch

2 detailed design work- takes place one year prior to launch

� Launch date is fixed (this is relaxed in subsequent work).


Model Feedback Structure cont.+

-

-

DELAY

Goal for Number of DesignConcept Development Design Problems in Problems

Activities on Next Year's Product This Year's Product

-+

+ -R B Problem Gap

Tipping Loop Rework Loop

+ResourcesResources Dedicated to ThisDedicated to Next Year's ProductYear's Product

-


+

-

Response to Unplanned Increases in Workload

% of Concept Development Work Quality of Finished Completed Design (% Defective)

Model Year Model Year


+

-

Linking SD Project Models: A Multi-Project Portfolio Model, Or Exogenously-Linked Single Project Models?

� If the number of projects active at one time is small, exogenous (time-dependent) inputs could connect single project models � staff available given needs of other projects � precedence and other constraints

� With a larger number of projects and complex technical interdependencies, a multi-project model is warranted


+

-Upstream - Downstream Projects

etion

Progress

li

l i

l i Availabiliisi

i

le

f

le

l i

ill & i

i

l

ional ff

l

ime

ini

i

l

li

led l i

l i Availabiliisi

i

le

l

i

i

Equi lff on

i if

lity

l

i

l

Upstream Progress & Quality

Start DownstreamEffect of

Upstream P&Q

Downstream Progress

Upstream Upstream Progress

Upstream Project

?

Work Qua ty to Date

Scheduled Comp et on

Time

Expected Comp et on

Time ty

of Prerequ tes

Perce ved Progress

ScheduPressure

Out-o -Sequence Work

Mora

Expected Hours at

Comp et on

Hours Expended

to Date

SkExper ence

Hir ng

Equiva ent Staff on Project

Staffing Requested

Progress

Rework Discovery

Turnover OrganizatSize

Changes

Sta

Productivity Quality

Added Work Obso eted

Work

Overt

Time Rema ng

Work To Be Done

Und scovered Rework

Known Rework

Work Real y Done

Work Qua ty to Date

ScheduComp et on

Time

Expected Comp et on

Time ty

of Prerequ tes

Perce ved

Schedule Pressure


Mora

Expected Hours at

Comp

Hours Expended

to Date

Skill & Exper ence

Hir ng

va ent StaProject

Staffing Requested

Progress

Rework Discovery

Turnover Organ zat onal Size

Changes

Staf

Productivity Qua


Work

Overt me

Time Remaining

Work To Be Done

Undiscovered Rework

Known Rework

Work Real y Done

on Downstream

Rework Discovery

Required

Staff Available

Downstream Project Copyright © 2003


+

-

Parallel Projects (Resource Competition not shown)

Project 1 Phase 1 Progress & Quality

li

l i

l i Availabiliisi

i

le

f

le

l i

ill & i

i

l

ional ff

l

ime

ini

i

l

Effect of Phase 2

Work Qua ty to Date

Scheduled et on

Time

Expected Comp et on

Time ty

of Prerequ tes

Perce ved Progress

ScheduPressure

Out-o -Sequence Work

Mora

Expected Hours at

Comp et on

Hours Expended

to Date

SkExper ence

Hir ng

Equiva ent Staff on Project

Staffing Requested

Progress

Rework Discovery

Turnover OrganizatSize

Changes

Sta

Productivity Quality


Work

Overt

Time Rema ng

Work To Be Done

Und scovered Rework

Known Rework

Work Real y Done

Phase 2 Rework

Req’d. Effect of Phase Discovery Comp

Phase 2 Progress

ProgressPhase 1 Rework P&Q On

etion Discovery Phase 1

Phase 2 Progress & Progress

li

led l i

l i Availabiliisi

i

le

l

i

i

Equi lff on

i if

lity

l

i

l

1 P&Q on

Req’d. Phase 1

Work Qua ty to Date

ScheduComp et on

Time

Expected Comp et on

Time ty

of Prerequ tes

Perce ved

Schedule Pressure


Mora

Expected Hours at

Comp

Hours Expended

to Date

Skill & Exper ence

Hir ng

va ent StaProject

Staffing Requested

Progress

Rework Discovery

Turnover Organ zat onal Size

Changes

Staf

Productivity Qua


Work

Overt me

Time Remaining

Work To Be Done

Undiscovered Rework

Known Rework

Work Real y Done

Phase 2

Quality Project 2 Copyright © 2003


+

-Possible Solutions

� Realistic planning and budgeting ! � Maintaining a buffer of surge capacity � Reducing the number of projects and/or

the scope per project


+

-

Company Strategy as it Affects the Portfolio

� Mix of projects that allow company to remain competitive over time

� Developing new technologies and skills � How much diversification is needed to

reduce risk (and is it worth it if short of resources?) Addressing these issues requires a

different model.


+

-Mission Dimensions

Priority & Specific Objectives

High, Stretch Medium Low, Slack

Features / Scope

Defects/ Undiscovered Rework

Resources / Cost

Time-to-Market/ Schedule

What should the “objectives” be? How many can be “high” priority?

Copyright © 2003 9/23/03 - ESD.36J SPM James M. Lyneis

+

-Revenue and Profit Drivers …

Sales

Market Demand

+ + Revenue

+ +

ProfitMarket Share -

Costs

Price +


+

-

“Mission” Elements Affect Attractiveness

Market Demand

+ Sales + Revenue +

+

ProfitMarket Share Price -

+ +

Costs

Product Attractiveness

Product Newness

Quality

Scope

+

+

+

Time to Market

-


+

-

“Mission” elements have negative impacts as well …

Revenue

Profit

Sales

Market Demand

Market Share Price

+ +

+ +

Product Attractiveness

Product Newness

Quality

Scope Developement Cost

CostsWarranty Costs

+

+

+

+

Time to Market

+ +

+

-

-

-

Mission Tradeoffs


+

-

Market Model with feedbacks through profit and budget

Market Demand

+ Sales + Revenue +

+

Market Share - Profit Price

+ +

Warranty Costs + Costs +

Product Attractiveness +

Quality+

Scope Developement Cost

+Product +

Newness Development - Budget

Time to Market

Mission Tradeoffs


+

-

Why Do Organizations Seem So Poor at Learning Lessons From Prior Projects?

� Belief that every project is different� Projects are transient phenomena -

companies often do not have organizations, money, systems, etc. that span projects, esp. to learn the lessons

� Limited span and career path of good project managers

� Lack of a framework (model) for comparing


+

-

Is What’s Best for the Project Best for the Company?

� Examples of conflicting practices -� Using the best people only � Using only experienced people � Introducing new processes (Sterman and

Repenning work at Sloan School)


esd.36j system & project managementdspace.mit.edu/bitstream/handle/1721.1/80702/esd... ·...

Documents