klastorin (traduccion)

7/29/2019 klastorin (TRADUCCION)

http://slidepdf.com/reader/full/klastorin-traduccion 1/205



Importance of Project Management

• Projects represent change and allow organizations toeffectively introduce new products, new

process, new programs

• Project management offers a means for dealing withdramatically reduced product cycle times

• Projects are becoming globalized making them moredifficult to manage without a formal methodology

• Project management helps cross-functional teams to be more effective



Management of IT Projects

• More than $250 billion is spent in the US each year onapproximately 175,000 information technology projects.

• Only 26 percent of these projects are completed on time andwithin budget.

• The average cost for a development project for a large companyis more than $2 million.

• Project management is an $850 million industry and is expectedto grow by as much as 20 percent per year.

Bounds, Gene. “The Last Word on Project

Management” IIE Solutions, November, 1998.



What Defines a Project?

•

•

• •

•

•

•

How does a project

differ from a

program?



Project Management versus Process Management

“Ultimately, the parallels between process and projectmanagement give way to a fundamental difference: process management seeks to eliminate variabilitywhereas project management must accept variability

because each project is unique.”

Elton, J. & J. Roe. “Bringing Discipline to Project

Management” Harvard Business Review, March-April,

1998.



Measures of Project Success

•

•

• •

•

•

•

Was the movie

“Titanic”

a success?



Delayed Openings are a Fact of Life in the Foodservice,

Hospitality Industry

Disney's shipbuilder was six months late in delivering its new cruise ships,and thousands of customers who had purchased tickets were stranded.

Even with that experience, their second ship was also delivered well after the

published schedules. Universal Studios in Orlando, Fla. had been building a

new restaurant and entertainment complex for more than two years. They

advertised a December opening, only to announce in late November that it

would be two or three months late.

Even when facilities do open close to schedule, they are rarely finished

completely and are often missing key components. Why do those things

happen? With all of the sophisticated computers and project management

software, why aren't projects completed on schedule?

Frable, F. Nation's Restaurant News (April 12, 1999)



IT Project Outcomes

More than 200%

late

Cancelled

On-Time

Less than 20%

late

21-50% late

51-100% late

101-200% late

26%

29%

6%

16%

9%

8%

6%

Source: Standish Group Survey, 1999 (from a

survey of 800 business systems projects)



Why do Projects Fail?

Studies have shown that the following factorscontribute significantly to project failure:

• Improper focus of the project management system

• Fixation on first estimates

• Wrong level of detail

• Lack of understanding about project management tools; too much

reliance on project management software

• Too many people

• Poor communication

• Rewarding the wrong actions



Why do IT Projects Fail?

• Ill-defined or changing requirements

• Poor project planning/management

• Uncontrolled quality problems

• Unrealistic expectations/inaccurate estimates

• Naive adoption of new technology

Source: S. McConnell, Construx Software Builders, Inc.



QuickTime™ and aPhoto - JPEG decompressor

are needed to see this picture.

Have You Ever Lost Sight of theProject Goals?



Not all Projects Are Alike…

“[in IT projects], if you ask people what‟s done and what remains to be

done there is nothing to see. In an IT project, you go from zero to 100

percent in the last second--unlike building a brick wall where you can see

when you‟re halfway done. We‟ve moved from physical to non-physical

deliverables….”

J. Vowler (March, 2001)

Engineering projects = task-centric

IT projects = resource-centric



Shenhar’s Taxonomy of Project Types

De g ree o f U n cert ain t y/ R isk

Sys tem C o m pl e xi t y/S c op e

H ig h

Low -

T e c h

A s s em b l y P ro ject s

A rr ay P ro ject s

Sy st e m P ro ject s

M e d i um- T e c h

H ig h- T e c h

S u p e r Hi gh - T e c h

Co n s tr u ct io n

N e w c e ll pho n e

N e w s h r i n k - w r ap p e d so ftw ar e

ER P im p l e me n t a ti on i n m u lti- na ti ona l

fi r m

A ut o r e pa ir

A d van c e d r ada r sy st e m



Project Life Cycle

Time Phase 1 Phase 2 Phase 3 Phase 4Formation & Planning Scheduling & Evaluation &Selection Control Termination

R e q u i r e d R e s o u r

c e s



Life Cycle Models: Pure Waterfall

ConceptDesign

Requirements

Analysis

Architecture

Design

Detailed

Design

Coding &

Debugging

System

Testing

Source: S. McConnell

Rapid Development (Microsoft Press, 1996)



Life Cycle Models: Code & Fix



Design, Cost, Time Trade-offs

Target

COST

D E S I G N

Due Date

Budget

Constraint

Optimal Time-Cost

Trade-off

Required

Performance



Optional Scope Contracts

Fixed Scope Contract specifies SCHEDULE, COST, SCOPE

Optional Scope Contract specifies SCHEDULE, COST, QUALITY

(general design guidelines may be indicated)

Since it is widely accepted that you can select

three of the four dimensions (or perhaps only

two), what to do?



Importance of Project Selection

“There are two ways for a business to succeed

at new products: doing projects right, and

doing the right projects.”

Cooper, R.G., S. Edgett, & E. Kleinschmidt.

Research • Technology Management , March-April, 2000.



Project Initiation & Selection

• Critical factors 1) Competitive necessity

2) Market expansion

3) Operating requirement

• Numerical Methods 1) Payback period

2) Net present value (NPV) or Discounted Cash Flow (DCF)

3) Internal rate of return (IRR)

4) Expected commercial value (ECV)

• Project Portfolio 1) Diversify portfolio to minimize risk

2) Cash flow considerations

3) Resource constraints



Payback Period

Number of years needed for project to

repay its initial fixed investment

Example: Project costs $100,000 and is expected

to save company $20,000 per year

Payback Period = $100,000 / $20,000 = 5 years



Net Present Value (NPV)

Discounted Cash Flow (DCF)

Let Ft = net cash flow in period t (t = 0, 1,..., T)

F 0 = ini tial cash investment in time t = 0

r = discount rate of return (hurdle rate)

NPV =

Ft

1 + r t•t = 0

T



Internal Rate of Return (IRR)

Find value of r such that NPV is equal to 0

F0 + F1

1 + r + F2

1 + r 2= 0

Example (with T = 2):

Find r such that



DCF Project Example*

*Hodder, J. and H.E. Riggs. “Pitfalls in Evaluating Risky Projects”, Harvard

Business Review, Jan-Feb, 1985, pp. 128-136.

Product

Demand Product Life

Annual Net

Cash Inflow Probability

High 20 years $24 million 0.3

Medium 10 years $12 million 0.5

Low Abandon Project None 0.2

Phase I Research and Product De velopment

$18 million annual research cost for 2 years60% probability of succ ess

Phase II Market Development

Undertaken only if product development is succes sful

$10 million annual expenditure for 2 years to develop marketing and

distribution c hannels (net of any revenues earned in tes t marketi

Phase III SalesProceeds only if Phase I and I I verify opportunity.

Production is subcontracted and all cash flows are after-tax and occu

at year's end.

The results of Phase II (available at the end of year 4) identify the

product's mar ket po tential as indic ated below:



DCF Project Example (cont’d)

Year Expe cte d Cash Flow (in $ million)

1 -18

2 -18

3 0.6 (-10) = -

4 0.6 (-10) = -5 - 14 .6 (0.3 x 24 + 0.5 x 12) = 7.92

15 - 24 .6 (0.3 x 24) = 4.32

What is the internal rate of return for this project?



DCF Example Continued

What if you can sell the product (assuming that both Research and

Product Development AND Market Development are successful) to a

third party? What are the risks AT THAT POINT IN TIME?

Assume that discount rate r 2 is 5%

Probability

What is 20 years of cash inflow at $24M/year? $299.09 0.3

What is 10 years of cash inflow at $12M/year? $92.66 0.5

Expected value of product at Year 4: $136.06



DCF Example Continued

Expected cash flows (with sale of product at end of year 4) are now:

Outflow Inflow Net Probability

Expe cted

Cash Flow

Year 1 18.00$ (18.00)$ 1 (18.00)$

Year 2 18.00$ (18.00)$ 1 (18.00)$Year 3 10.00$ (10.00)$ 0.6 (6.00)$

Year 4 10.00$ 136.06$ 126.06$ 0.6 75.63$

What is the internal rate of return for this project?



Criticisms of NPV/DCF

1) Assumes that cash flow forecasts are accurate; ignoresthe “human bias” effect

2) Fails to include effects of inflation in long term

projects

3) Ignores interaction with other proposed and ongoing

projects (minimize risk through diversification)

4) Use of a single discount rate for the entire project (risk

is typically reduced as the project evolves)



Expected Commercial Value (ECV)

Develop New

Product

Technical Failure

Technical Success

Probability = pt

Probability = 1 - pt

Launch New

Product

Commercial

Failure (with net

benefit = 0)

Commercial Success

(with net benefit =

NPV)

Probability = pc

Probability = 1 - pc

Risk class 1 Risk class 2



DCF Example Revisited

Discount rate r 1 Discount rate r 2

Research &

Product

Development

Development

Succeeds

Probability = pt

Development Fails

Probability = 1 - pt

Market

Development

Product Demand

High 0.3

Product Demand

Medium

Product Demand

Low

0.5

0.2

Drop project



Ranking/Scoring ModelsProfitability/value

1) Increase in profitability?2) Increase in market share?3) Will add knowledge to organization that can be leveraged by other projects?4) Estimated NPV, ECV, etc.

Organization's Strategy1) Consistent with organization's mission statement?2) Impact on customers?

Risk 1) Probability of research being successful?2) Probability of development being successful?3) Probability of process success?4) Probability of commercial success?5) Overall r isk of project6) Adequ ate market demand?7) Competitors in market

Organizat ion Costs

1)

Is new facilit y needed?2) Can use current personnel?3) External consultants needed?4) New hires needed?

Miscellaneous Factors1) Impact on environmental standards?2) Impact on workforce safety?3) Impact on quality?4) Social/political implications



Scoring Attributes

vi xi =1 - exp L - xi

1 - exp L - U.

To convert various measurement scales to a (0, 1) range….

LINEAR SCALE: value of attribute i is

EXPONENTIAL SCALE: value of attribute i is

vi xi =xi - L

U - L

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1 2 3 4 5 6 7

Res ponse

Linear Scale

Exponential Scale



Ranking/Scoring Example

Attribute Measurement ScaleAttribute

Weight (wi)

1) Does project increase market share? unlikely likely 30%

2) Is new facilit y needed? yes no 15%

3) Are there safety concerns? likely unsure no 10%

4) Likeli hood o f successful t echnical development? unlikely likely 20%

5) Likelihood of successful commercial development? unlikely likely 25%

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5



Attribute #1 #2 #3 #4 #5

Project

Score (V j)

Project A 4 yes likely 4 1

Project B 2 no unsure 3 4

Linear ScaleProject A 0.75 0.25 0 0.75 0 0.413

Project B 0.25 0.75 0.5 0.5 0.75 0.525

Exponential ScaleProject A 0.97 0.64 0.00 0.97 0.00 0.581

Project B 0.64 0.97 0.88 0.88 0.97 0.845

Ranking/Scoring Example (cont’d)



Analyzing Project Portfolios: Bubble Diagram

Expected NPV

Prob of Commercial Success

High Zero

Low

High



Analyzing Project Portfolios: Product vs Process

Extent of Process Change

Source: Clark and Wheelwright, 1992



Key Elements of Project Portfolio Selection Problem

1. Multi-period investment problem

2. Top management typically allocates funds to different

product lines (e.g., compact cars, high-end sedans)

3. Product lines sell in separate (but not necessarilyindependent) market segments

4. Product line allocations are changed frequently

5. Conditions in each market segment are uncertain from

period to period due to competition and changing

customer preferences



“Stage-Gate” Approach

Installation Plan

Facility Prep

Training Plan

Implementation

Detail Design

Schedule & Budget

Contingency Plan

Product &

Performance Reviews

Initiation Define Design ControlImprove

Work Statement

Risk Assessment

Purchasing Plan

Change Mgt

Initiation

Project Review

Charter

Source: PACCAR Information Technology Division

Renton, WA

Production close-out

Lessons learned

Post-project audit



Project Selection Example

Y e a r (t)

1 2 3 4

Project A ($40) $10 $20 $20

Project B ($65) ($25) $50 $50Budget

Limit (B t ) $120 $20 $40 $55



Phases of Project Management

n Project formulation and selection

n Project planningu Summary statement

u Work breakdown structure

u Organization plan

u risk management

u

Subcontracting and bidding processn Project scheduling

u Time and schedule

u Project budget

u Resource allocation

u Equipment and material purchases

n Monitoring and controlu Cost control metrics

u Change orders

u Milestone reports



Project Planning

n

Summary Statementu Executive summary: mission and goals, constraints

u Description and specifications of deliverables

u Quality standards used (e.g., ISO)

u Role of main contractor and subcontractors

u Composition and responsibilities of project team

n Organization Planu Managerial responsibilities assigned; signature authority

u Cross impact matrix (who works on what)

u Relationship with functional departments

u Project administration

u Role of consultants

u Communication procedures with organization, client, etc.



Importance of Project Planning

The 6P Rule of Project Management:

Prior Planning Prevents Poor Project

Performance

“If you fail to plan, you will plan to fail”

Anonymous



Work Breakdown Structure (WBS)

1) Specify the end-item “deliverables”

2) Subdivide the work, reducing the dollars and

complexity with each additional subdivision

3) Stop dividing when the tasks are manageable “work packages” based on the following:

• Skill group(s) involved

• Managerial responsibility

• Length of time

• Value of task



Work Packages/Task Definition

The work packages (tasks or activities) that are defined

by the WBS must be:

• Manageable

• Independent

• Integratable

• Measurable



Design of a WBS

“The usual mistake PMs make is to lay out too many tasks;subdividing the major achievements into smaller and

smaller subtasks until the work breakdown structure

(WBS) is a „to do‟ list of one-hour chores. It‟s easy to get

caught up in the idea that a project plan should detaileverything everybody is going to do on the project. This

springs from the screwy logic that a project manager‟s job

is to walk around with a checklist of 17,432 items and tick

each item off as people complete them….”

The Hampton Group (1996)



Two-Level WBS

1. Charity Auction

1.1 Event

Planning

1.2 Item

Procurement

1.3 Marketing 1.4. Corporate

Sponsorships

WBS level 1

WBS level 2



Three-Level WBS

1.1 Event

Planning

1.2 Item

Procurement

1.3 Marketing

1. Charity Auction

1.4 Corporate

Sponsorships

1.1.1 Hire Auctioneer

1.1.2. Rent space

1.1.3 Arrange for decorations

1.2.1 Silent

auction items

1.2.2 Live auction

items

1.2.3 Raffle items

1.3.1 Individual

ticket sales

1.3.2 Advertising

1.1.4 Print catalog

WBS level 1

WBS level 2

WBS level 3



Estimating Task Durations (cont’d)

• Benchmarking

• Modular approach

• Parametric techniques

• Learning effects



Beta Distribution

Completion time of task j

Optimistic Timet jo

Pessimistic Time t j p

Time

Probability density

function

Expected duration =Most Likely Time = tm



Beta Distribution

For each task j, we must make three estimates:

most optimistic time

most pessimistic time

most likely time

t jo

t j p

t jm

Expected duration j =t jo + t j

p + 4t jm

6

Variance of task j = j2 = t j

p

- t jo 2

36



Estimating Task Durations: Painting a Room

Task : Paint 4 rooms, each is approximately 10‟ x 20‟. Use flat paint on walls,

semi-gloss paint on trim and woodwork. Each room has two doors and four

windows. You must apply masking tape before painting woodwork around thedoors and windows. Preparation consists of washing all walls and woodwork

(some sanding and other prep work will be needed). Only one coat of paint is

necessary to cover existing paint. All supplies will be provided at the start of the

task. Previous times on similar painting jobs are indicated in the table below.

hours min hours min

27 25 31 52

38 25 19 1533 12 26 27

17 44 30 27

26 7 25 21

22 1 24 28

14 2 32 58

30 27 32 1

28 30 13 43

21 13 42 45

23 59 22 57

27 44 32 1523 15 32 31

37 6 27 15

17 54 26 11

17 13 21 52

What is your estimate of the average time you will

need? What is your estimate of the variance?



Estimating Task Durations with Incentives

Task: Consider the painting job that you have just estimated. Now, however, there are

explicit incentives for meeting your estimated

times. If you finish painting the room before

your specified time, you will receive a $10

bonus payment. HOWEVER, if you finish

the painting job after your specified time, you

will be fined $1000. Revised estimated time =



Estimating Task Durations with Incentives

Task: Consider the painting job that you have just estimated. Now, however, there are

explicit incentives for meeting your estimated

times. If you finish painting the room before

your specified time, you will receive a $10

bonus payment. If you finish the painting job

after your specified time, there is no penalty.

Revised estimated time =



Role of Project Manager/Team

Project Manager

Client

Subcontractors

Regulating

Organizations

Project Team

Functional

Managers

Top

Management



Responsibilities of a Project Manager

To the organization and top management

• Meet budget and resource constraints

• Engage functional managers

To the project team • Provide timely and accurate feedback

• Keep focus on project goals • Manage personnel changes

To the client• Communicate in timely and accurate manner

• Provide information and control on changes/modifications

• Maintain quality standards

To the subcontractors • Provide information on overall project status



Project Team

What is a project team?

A group of people committed to achieve a

common set of goals for which they hold

themselves mutually accountable

Characteristics of a project team

• Diverse backgrounds/skills

• Able to work together effectively/develop synergy

• Usually small number of people • Have sense of accountability as a unit



“I design user interfaces to please an audience of one.

I write them for me. If I’m happy, I know some cool

people will like it. Designing user interfaces by

committee does not work very well; they need to becoherent. As for schedule, I’m not interested in

schedules; did anyone care when War and Peace came

out?”

Developer, Microsoft Corporation As reported by MacCormack and Herman, HBR Case 9-600-097:

Microsoft Office 2000



Intra-team Communication

M = Number of project team membersL = Number of links between pairs of team members

If M =2, then L = 1

If M =3, then L = 3



Number of Intra-team Links

L = Number of Intra-team Links = N

2=

N(N-1)

2



Importance of Communication

On the occasion of a migration from the east, men discovered aplain in the land of Shinar, and … said to one another, “Come, let

us build ourselves a city with a tower whose top shall reach the

heavens….” The Lord said, …“Come, let us go down, and there

make such a babble of their language that they will not

understand one another’s speech.” Thus, the Lord dispersedthem from there all over the earth, so that they had to stop

building the city.

Genesis 11: 1-8



Project Performance and Group Harmony

Two schools of thought:

1) “Humanistic school” -- groups that have positivecharacteristics will perform well

2) “Task oriented” school -- positive group

characteristics detract from group performance

What is the relationship between the design of multidisciplinary project teams and project success?



Project Performance and Group Harmony (cont’d)

Experiment conducted using MBA students at UW and

Seattle U using computer based simulation of pre-operationaltesting phase of nuclear power plant*

Total of 14 project teams (2 - 4 person project teams) with atotal of 44 team members; compared high performance (low

cost) teams vs low performance (high cost) teams

Measured: Group Harmony

Group Decision Making Effectiveness

Extent of Individual‟s Contributions to Group

Individual Attributes

*Brown, K., T.D. Klastorin, & J. Valluzzi. “Project Management

Performance: A Comparison of Team Characteristics”, IEEE Transactions on

Engineering Management , Vol 37, No. 2 (May, 1990), pp. 117-125.



Group Harmony: High vs Low Performing Groups



Extent of Individual Contribution: High vs LowPerforming Groups



Decision Making Effectiveness: High vs LowPerforming Groups



Project Organization Types

• Functional: Project is divided and assigned to appropriate functional

entities with the coordination of the project being carried out by

functional and high-level managers

• Functional matrix: Person is designated to oversee the project

across different functional areas

• Balanced matrix: Person is assigned to oversee the project and

interacts on equal basis with functional managers

• Project matrix: A manager is assigned to oversee the project and is

responsible for the completion of the project

• Project team: A manager is put in charge of a core group of

personnel from several functional areas who are assigned to the

project on a full-time basis



Project Organization Continuum

Project Team

Organization

Project Matrix

Project fully managed

by functional managers Project fully managed by

project team manager

FunctionalOrganization

Functional Matrix

Balanced Matrix



A Business School as a Matrix Organization

Dean

Associate Dean for

Undergraduate

Program

Associate Dean for

MBA Programs

Director of

Doctoral Program

Accounting

Department Chair

Marketing

Department Chair

Finance Department

Chair

Gloria

Diane

Bob

Zelda Larry

Curly

Moe

Barby

Leslie



Matrix Organizations & Project Success

• Matrix organizations emerged in 1960‟s as analternative to traditional means of project

teams

• Became popular in 1970‟s and early 1980‟s

• Still in use but have evolved into many different

forms

• Basic question: Does organizational structure

impact probability of project success?



Organizational Structure & Project Success• Studies by Larson and Gobeli (1988, 1989)

• Sent questionnaires to 855 randomly selected PMI members

• Asked about organizational structure (which one best describes the primary

structure used to complete the project)

• Perceptual measures of project success: successful, marginal, unsuccessful

with respect to :

1) Meeting schedule2) Controlling cost

3) Technical performance

4) Overall performance

• Respondents were asked to indicate the extent to which they agreed with

each of the following statements:1) Project objectives were clearly defined

2) Project was complex

3) Project required no new technologies

4) Project had high priority within organization



• Classification of 547 respondents (64% response rate)

30% project managers or directors of project mgt programs16% top management (president, vice president, etc.)

26% managers in functional areas (e.g., marketing)

18% specialists working on projects

• Industries included in studies

14% pharmaceutical products10% aerospace

10% computer and data processing products

others: telecommunications, medical instruments, glass products,software development, petrochemical products, houseware goods

• Organizational structures:

13% (71): Functional organizations

26% (142): Functional matrix

16.5% (90): Balanced matrix

28.5% (156): Project matrix

16% (87): Project team

Study Data



ANOVA Results by Organizational Structure

Controlling

Cost

Meeting

Schedule

Technical

Performance

Overall

Results

Organizational Structure N Ave (SD) Ave (SD) Ave (SD) Ave (SD)

A

Functional

Organization 71 1.76 (.83) 1.77 (.83) 2.30 (.77) 1.96 (.84)

B Functional Matrix 142 1.91 (.77) 2.00 (.85) 2.37 (.73) 2.21 (.75)

C Balanced Matrix 90 2.39 (.73) 2.15 (.82) 2.64 (.61) 2.52 (.61)

D Project Matrix 156 2.64 (.76) 2.30 (.79) 2.67 (.57) 2.54 (.66)

E Project Team 87 2.22 (.82) 2.32 (.80) 2.64 (.61) 2.52 (.70)

Total Sample 546 2.12 (.79) 2.14 (.83) 2.53 (.66) 2.38 (.70)

F-statistic 10.38* 6.94* 7.42* 11.45*

Scheffe Results

A,B < C,D,E

E < D A, B < C < D,E A, B < C,D, E A, B < C,D, E

*Statisticall y signif icant at a p<0.01 level



Summary of Results

• Project structure significantly related to project success

• New development projects that used traditional functional organization

had lowest level of success in controlling cost, meeting schedule,

achieving technical performance, and overall results

• Projects using either a functional organization or a functional matrix had

a significantly lower success rate than the other three structures

• Projects using either a project matrix or a project team were more

successful in meeting their schedules than the balanced matrix

• Project matrix was better able to control costs than project team

• Overall, the most successful projects used a balanced matrix, project

team, or--especially--project matrix



Subcontracting = Business Alliance

n When you subcontract part (or all) of a

project, you are forming a business

alliance....

Intelligent Business Alliances: “A business relationship for

mutual benefit between two or more parties with compatible

or complementary business interests and/or goals”

Larraine Segil, Lared Presentations



Communication and Subcontractors

How is knowledge

transferred?

What types of communication mechanism(s) will be

used between company and subcontractor(s)?

WHAT a companycommunicates.....

HOW a companycommunicates.....



Personality Compatibility

Corporate

Personality

Subcontractor

Personality

Individual

Personality

Project



Subcontracting Issues

n• What part of project will be subcontracted? n• What type of bidding process will be used? What type of

contract?

n• Should you use a separate RFB (Request for Bids) for

each task or use one RFB for all tasks?n• What is the impact on expected duration of project?

n• Use a pre-qualification list?

n• Incentives? Bonus for finishing early? Penalties for

finishing after stated due date?

• What is impact of risk on expected project cost?



Basic Contract Types

n Fixed Price Contract u Client pays a fixed price to the contractor irrespective of actual audited

cost of project

n Cost Plus Contract

u Client reimburses contractor for all audited costs of project (labor, plant,& materials) plus additional fee (that may be fixed sum or percent of costs

incurred)

n Units Contract

u Client commits to a fixed price for a pre-specified unit of work; final

payment is based on number of units produced



Incentive (Risk Sharing) Contracts

General Form:Payment to Subcontractor = Fixed Fee + (1 - B) (Project Cost)

where B = cost sharing rate

Cost Plus Contract

B = 0 B = 1

Fixed Price Contract

Linear & Signalling

Contracts



Why Use Incentive Contracts?

Expected Cost of Project = $100MTwo firms bid on subcontract

Firm 1 Firm 2

Fixed Fee (bid) $5 M $7 M

Project Cost $105 M $95 M

(inefficient producer)

What is result if Cost Plus Contract (B = 0) used?



Washington State Bid Code (WAC 236-48-093)

n WAC 236-48-093: A contract shall be awarded to the lowest responsible and responsive

bidder based upon, but not limited to, the following criteria where applicable and onlythat which can be reasonably determined:

n 1) The price and effect of term discounts...price may be determined by life cycle costing

if so indicated in the invitation to bid

n 2) The conformity of the goods and/or services bid with invitation for bid or request for

quotation specifications depicting the quality and the purposes for which they are

required.n 3) The ability, capacity, and skill of the bidder to perform the contract or provide the

services required.

n 4) The character, integrity, reputation, judgement, experience, and efficiency of the

bidder.

n 5) Whether the bidder can perform the contract with the time specified.

n 6) The quality of performance on previous contracts for purchased goods or services.n 7) The previous and existing compliance by the bidder with the laws relating to the

contract for goods and services.

n 8) Servicing resources, capability, and capacity.



Competitive Bidding: Low-Bid System

n “In the low-bid system, the owner wants the most building for the least money, while the contractor wants the least building for the most money. Thetwo sides are in basic conflict.”

Steven Goldblatt

Department of Building Construction

University of Washington

The Seattle Times, Nov 1, 1987



Precedence Networks

Networks represent immediate precedence relationshipsamong tasks (also known as work packages or activities)

and milestones identified by the WBS

Milestones (tasks that take no time and cost $0 but indicate

significant events in the life of the project) Two types of networks: Activity-on-Node (AON)

Activity-on-Arc (AOA)

All networks: must have only one (1) starting and one (1)ending point



Precedence Networks: Activity-on-Node (AON)

A

B

C

D

Start End



Precedence Diagramming

Standard precedence network (either AOA or AON) assumes that a successor

task cannot start until the predecessor(s) task(s) have been completed.Alternative relationships can be specified in many software packages:

Finish-to-start (FS = a): Job B cannot start until a days after Job A is

finished

Start-to-start (SS = a): Job B cannot start until a days after Job A hasstarted

Finish-to-finish (FF = a): Job B cannot finish until a days after Job A

is finished

Start-to-finish (SF = a): Job B cannot finish until a days after Job A

has started



Critical Path Method (CPM): Basic Concepts

Task A

7 months

Task B

3 months

End

Task C

11 months

Start



Critical Path Method (CPM): Basic Concepts

Start

Task A

7 months

Task B

3 months

Task C

11 months

End

ESStart = 0

LFStart = 0

ESA = 0

LFA = 8

ESB = 7

LFB = 11

ESC = 0

LFC = 11

ESEnd = 11

LFEnd = 11

ES j = Earliest starting time for task (milestone) j

LF j = Latest finish time for task (milestone) j

AON P d N t k Mi ft P j t



AON Precedence Network: Microsoft Project

Start

1 0d

Wed 12/20/00 Wed 12/20/00

Ta sk A

2 7d

Wed 12/20/00 Thu 12/28/00

Task C

4 11d

Wed 12/20/00 Wed 1/3/01

End

5 0d

Wed 1/3 /0 1 Wed 1/3 /0 1

Ta sk B

3 3d

Fri 12/29 /00 Tue 1 /2 /01



Critical Path Method (CPM): Example 2

Ta sk A 14 w k s

Ta sk D

12 w k s

Ta sk E 6 w k s

Ta sk B 9 w k s

Ta sk C 20 w k s

Ta sk F 9 w k s

START END

ES F = LF F =

ES D = LF D =

ES ST ART = 0LFST ART = 0

ES A = LF A =

ES B = LF B =

ES END = LF END =

ES C = LF C =

ES E = LF E =



Example 2: Network Paths

Path Tasks

Expected

Duration (wks)

1 START-A-D-F-END 35

2 START-A-D-E-END 323 START-B-D-F-END 30

4 START-B-D-E-END 27

5 START-C-E-END 26



Example 2: CPM Calculations

E A R L I E S T L A T E S T

Task or

Milestone

Duration

( )

Start Time

(ESi) Finish Time Start Time

Finish Time

(LFi)

ST ART 0 0 0 0 0

A 14 0 14 0 14

B 9 0 9 5 14C 20 0 20 9 29

D 12 14 26 14 26

E 6 26 32 29 35

F 9 26 35 26 35

END 0 35 35 35 35

ti



Example 2: Calculating Total Slack (TSi)

Task or

Milestone

Duration

( )

Earliest

Start Time

(ESi)

Lastest

Finish Time

(LFi)

Total Slack

(TSi)

Critical

Task?

ST ART 0 0 0 0 YesA 14 0 14 0 Yes

B 9 0 14 5 No

C 20 0 29 9 No

D 12 14 26 0 Yes

E 6 26 35 3 No

F 9 26 35 0 Yes

END 0 35 35 0 Yes

ti

Total Slack for task i = TSi = LFi - ESi - ti



Slack (Float) Definitions (for task i)

Total Slack (TSi) = LF

i

- ESi

- ti

Free Slack (FSi) = ESi,min - ESi - ti

where ESi,min = minimum early start time of all tasks that

immediately follow task i

= min (ES j for all task j Si)

Safety Slack (SSi) = LFi - LFi,max - ti

where LFi,max = maximum late finish time of all tasks that

immediately precede task i

= min (LF j for all task j Pi)

Independent Slack (ISi) = max (0, ESi,min - LFi,max - ti)

E l #2 LP M d l



Example #2: LP Model

Decision variables: START j = start time for task j

END = ending time of project (END milestone)

Minimize END

subject to

STARTj ≥ FINISHi for all tasks i that immediately precede task j

STARTj ≥ 0 for all tasks j in project

where FINISHi = STARTi + ti = STARTi + duration of task i

E l #2 E l S l M d l



Example #2: Excel Solver Model

Gantt Chart



Gantt Chart

Microsoft Project 4.0

P j t B d ti



Project Budgeting• The budget is the link between the functional units and the project

• Should be presented in terms of measurable outputs

• Budgeted tasks should relate to work packages in WBS and

organizational units responsible for their execution

• Should clearly indicate project milestones

• Establishes goals, schedules, and assigns resources (workers,

organizational units, etc.)

• Should be viewed as a communication device

• Serves as a baseline for progress monitoring & control

• Update on rolling horizon basis

• May be prepared for different levels of aggregation (strategic,

tactical, short-range)

P j t B d ti ( t’d)



Project Budgeting (cont’d)

• Top-down Budgeting: Aggregate measures (cost,

time) given by top management based on

strategic goals and constraints

• Bottom-up Budgeting: Specific measures aggregated

up from WBS tasks/costs and subcontractors

I i P j t B d t



Issues in Project Budgets

• How to include risk and uncertainty factors?

• How to measure the quality of a project budget?

• How often to update budget?

• Other issues?

C iti l P th M th d (CPM) E l 2



Critical Path Method (CPM): Example 2

Ta sk A 14 w k s

Ta sk D

12 w k s

Ta sk E 6 w k s

Ta sk B 9 w k s

Ta sk C 20 w k s

Ta sk F 9 w k s

START END

ES F = 26 LF F = 35

ES D = 14 LF D = 26

ES ST ART = 0LFST ART = 0

ES A = 0

LF A = 14

ES B = 0 LF B = 14

ES END = 35 LF END = 35

ES C = 0 LF C = 29

ES E = 26 LF E = 35

P j t B d t E l



Project Budget Example

Task or

Milestone

Duration

(tj)

Early Start

Time (ESj)

Latest Start

Time (LSj)

No. of

Resource A

workers

No. of

Resource B

workers

Material

Costs

Direct Labor

Cost/wk

Labor +

Materials

ST ART 0 0 0 - - - - -

A 14 0 0 2 0 340$ 800$ 1,140$

B 9 0 5 4 12 125$ 8,800$ 8,925$

C 20 0 9 3 14 -$ 9,600$ 9,600$

D 1214 14 0 8 200$ 4,800$ 5,000$

E 6 26 29 1 0 560$ 400$ 960$

F 9 26 26 4 10 90$ 7,600$ 7,690$

END 0 35 35 - - - - -

Cost for Resource A worker = $400/week

Cost for Resource B worker = $600/week

P j t B d t E l ( t’d)



Project Budget Example (cont’d) Early Start Times

Task 1 2 3 4 5 6 7 8 9 10 11 12

A 1140 800 800 800 800 800 800 800 800 800 800 800

B 8925 8800 8800 8800 8800 8800 8800 8800 8800

C 9600 9600 9600 9600 9600 9600 9600 9600 9600 9600 9600 9600

D

E

F

Weekly Subtotals 19665 19200 19200 19200 19200 19200 19200 19200 19200 10400 10400 10400

Cumulative 19665 38865 58065 77265 96465 115665 134865 154065 173265 183665 194065 204465

Late Start Times

Task 1 2 3 4 5 6 7 8 9 10 11 12

A 1140 800 800 800 800 800 800 800 800 800 800 800

B 8925 8800 8800 8800 8800 8800 8800 8800

C 9600 9600 9600 9600

D

E

F

Weekly Subtotals 1140 800 800 800 9725 9600 9600 9600 19200 19200 19200 19200

Cumulative 1140 1940 2740 3540 13265 22865 32465 42065 61265 80465 99665 118865

W e e k

W e e k

C l ti C t



Cumulative Costs

Range of

feasible budgets

Weekly Costs (Cash Flows)



Weekly Costs (Cash Flows)

M i C h Fl



Managing Cash Flows

• Want to manage payments and receipts

• Must deal with budget constraints on

project and organization requirements (e.g.,

payback period)• Organization profitability

C h Fl E l



Cash Flow Example

M1

END

START

Task B

8 mos

Receive payment

of $3000

Receive payment

of $3000

Make paymentof $5000

Task C

4 mos

Task A

2 mos

M2

Task D8 mos

Task E

3 mos

C h Fl E l S l M d l



Cash Flow Example: Solver Model

Material Management Issues



Material Management Issues

When to order materials? How much to order?

Example:

• Single material needed for Task B (2 units) and Task E (30 units)

• Fixed cost to place order = S

• Cost of holding raw materials proportional to number of unit-weeks in

stock

• Cost of holding finished product greater than the cost of holding raw

materials

• Project can be delayed (beyond 17 weeks) at cost of $P per week

Material Management Example



Material Management Example

Task A

4 wks Task B

8 wks Task C

5 wks

Task D

6 wks Task E

2 wksTask F

3 wks

EndStart 2 units

30 units

LSA = 0 LSB = 4 LSC = 12

LSD = 6 LSE = 12 LSF = 14

Lot-Sizing Decisions in Projects



Lot Sizing Decisions in Projects

• To minimize holding costs, only place orders at Late Starting Times

• Can never reduce holding costs by delaying project

Time

1 2 3 4 5 6 7 8 9 10 11 12

Demand: 2 30

Order option #1: 32

Order option #2: 2 30

Choose the option that minimizes inventory cost = order cost + holding

cost of raw materials



Time-Cost Tradeoffs

Ti C t T d ff E l



Time-Cost Tradeoff Example

Task

Normal

Duration Normal Cost

Marginal Cost

to Crash One

Week A 7 $60 $8

B 6 $85 $5

C 15 $55 $10

D 10 $120 $4

A

B

C

D

Start End

Ti C t T d ff E l ( t’d)



Time-Cost Tradeoff Example (cont’d)

Project

Duration

(weeks) Critical Path(s) Task(s) Reduced

Total Direct

Cost

22 Start-A-C-End - $320

21 Start-A-C-End A $328

Start-A-B-End

20 Start-A-C-End C $338

Start-A-B-End

19 Start-A-C-End C $348

Start-A-B-End

18 Start-A-C-End A, B $361

Start-A-B-End

Linear Time-Cost Tradeoff



Linear Time-Cost Tradeoff In theory, the normal or expected duration of a task can be reduced by

assigning additional resources to the task

Time

Cost

Crash

Point

Normal

Point

Slope (b j) = Increase in cost by

reducing task by one time unit

Normal time =Crash time =

Normal

cost =

Crash

cost =

t j Nt j

c

C j

c

C j N

Balancing Overhead & Direct Costs



Balancing Overhead & Direct Costs

Project

Duration

Cost

Indirect

(overhead)

Costs

Direct

Costs

Total Cost

Crash

Time

Normal Time Minimum Cost

Solution

Time-Cost Tradeoff (Direct Costs Only)



( y)

Given Normal point with cost and time

and Crash point with cost and time

Assume constant marginal cost of crashing task j =

Decision Variables: S j = Starting time of task j

END = End time of project

t j = Duration of task j

Minimize Total Direct Cost =

S j ≥ Si + ti for all tasks i P j

for all tasks in project

END = Tmax

t j, S j ≥ 0

C j N

C jc

t jc

t j N

b j =C j

c - C j N

t jc - t j

N

b j t j• j

t jc t j t j

N

General Time-Cost Tradeoffs



where

I = indirect (overhead) cost/time period

P = penalty cost/time period if END is delayed beyond

deadline Tmax

L = number of time periods project is delayed beyond

deadline Tmax

Minimize Total Costs = + I (END) + P L b j t j• j

QUESTION: HOW TO DEFINE L?

Software Project Schedules



j

“Observe that for the programmer, as for the chef, the urgency of the patron may govern the scheduled completion of the task, but itcannot govern the actual completion. An omelet, promised in tenminutes, may appear to be progressing nicely. But when it has notset in ten minutes, the customer has two choices--wait or eat itraw. Software customers have the same choices.

The cook has another choice; he can turn up the heat. Theresult is often an omelet nothing can save--burned in one part, rawin another.”

F.P. Brooks, “The Mythical Man-Month”, Datamation, Vol 20, No 12 (Dec, 1974), pp.44-52.

Coordination Costs (Software Development Project)



Coordination Costs (Software Development Project)

n Assume you want to develop program that will require (approximately) 50,000 lines of

PERL code

n A typical programmer can write approximately 1500 lines of code per week

n Coordination time is M (M-1)/2 weeks

No. of Programmers

No. of

WeeksCoding

No. of

CoordinationWeeks

Total

Number of Weeks

1 33.33 0 33.33

2 16.67 1 17.67

3 11.11 3 14.11

4 8.33 6 14.33

5 6.67 10 16.67

6 5.56 15 20.56

7 4.76 21 25.76

8 4.17 28 32.17

9 3.70 36 39.7010 3.33 45 48.33

11 3.03 55 58.03

Brook’s Law



Brook s Law

“Adding manpower to a latesoftware project makes it later.”

n F.P. Brooks, “The Mythical Man-Month”, Datamation, Vol 20, No 12 (Dec, 1974), pp. 44-52.

Compressing New Product Development



p g pProjects

Traditional Method

Design follows a sequential pattern where

information about the new product is slowly

accumulated in consecutive stages

Stage 0 Stage 1 Stage N

New Product Development Process



New Product Development Process

Overlapped Product Design

Allows downstream design stages to start before

preceding upstream stages have finalized their specifications….

Stage 0

Stage 1

Stage N

Issues and Tradeoffs



Issues and Tradeoffs

What are the tradeoffs when moving from a

traditional sequential product design process

to an overlapped product design process?

• Increased uncertainty (that leads to additional

work)

• Can add additional resources to tasks to reduce

duration--but costs are increased

Classic PERT Model Defined



• Since task durations are now random variables, time of any

milestone (e.g., end of project) is now RV

• Assume all tasks are statistically independent

• Use values of j to identify expected critical path

• Since time of event (e.g., ESk ) is now sum of independent RV‟s,

central limit theorem specifies that ESk is approximatelynormally distributed with mean E[ESk ] and variance Var[ESk ]

where there exists s paths to task k

Expected early start time of task k = E ESk =max

s j•

tasks j on path s

Classic PERT Model (cont’d)



Expect Project Duration = E[E SEND] = j•tasks j on CP

Variance of Project Duration = Var[E SEND] = j2•

tasks j on CP

Thus, expected project duration is defined as:

Using central limit theorem and standard normal distribution:

P ESENDŠ Tmax = P z ŠTmax - E ESEND

Var ESEND

PERT Example #1



Duration Estimates Expected

Task Description Predecessors Optimistic Pessimistic Likely Duration Variance

A Requi rements Analysis none 2 14 6 6.67 4.00

B Programming A 4 12 7 7.33 1.78

C Hardware acquisi tion A 2 13 8 7.83 3.36D User training A 12 18 14 14.33 1.00

E Implementation B, C 3 7 5 5.00 0.44

F Testing E 3 7 4 4.33 0.44

END End of project D, F 0 0 0 0.00 0.00

StartTask A

RequirementsAnalysis

Task C Hardware

Acquisition

Task B Programming

Task F Testing

Task D User

Training

Task E Implementation

End

PERT Example #1 (cont’d)



Expected

Task Path Early Start Variance Due Date Zi Pr(zi)

B,C,D Start-A 6.67 4.00 6 -0.33 0.37

E Start-A-C 14.50 7.36 15 0.18 0.57

F Start-A-C-E 19.50 7.81 20 0.18 0.57

End Start-A-C-E-F-End 23.83 8.25 25 0.41 0.66

PERT Expe cted Dur at i on = 23 .8 3 Ex pected CP = {Star t , A, C , E, F , En d}

PERT Var ia nce = 8.25 0

StartTask A

RequirementsAnalysis

Task C Hardware

Acquisition

Task B Programming

Task F Testing

Task D User

Training

Task E Implementation

End

PERT Example #2



Task B

B = 12

B 2 = 4

Task D

D = 3

D 2 = 1

Task A

A = 4

A 2 = 2

Task C

C = 10

C 2 = 5

END START

Example #3: Discrete Probabilities



Task A Task B Task C Task D

Value Prob Value Prob Value Prob Value Prob

7 0.333 2 0.2 5 0.2 3 0.3

8 0.333 12 0.8 15 0.2 12 0.7

9 0.333 25 0.6

START END

Task A

(8.0)

Task B

(10.0)

Task C

(19.0)

Task D

(9.3)

Example #3 (cont’d)



Task A Task B Task C Task D Critical Pr ob of Len gth PATHS

Combination Value Prob Value Prob Value Prob Value Prob Path CP of CP A,D B, D C

1 7 0.333 2 0.2 5 0.2 3 0.3 A, D 0.004 10 0.004 0.000 0.000

2 7 0.333 2 0.2 5 0.2 12 0.7 A, D 0.009 19 0.009 0.000 0.000

3 7 0.333 2 0.2 15 0.2 3 0.3 C 0.004 15 0.000 0.000 0.004

4 7 0.333 2 0.2 15 0.2 12 0.7 A, D 0.009 19 0.009 0.000 0.000

5 7 0.333 2 0.2 25 0.6 3 0.3 C 0.012 25 0.000 0.000 0.012

6 7 0.333 2 0.2 25 0.6 12 0.7 C 0.028 25 0.000 0.000 0.028

7 7 0.333 12 0.8 5 0.2 3 0.3 B, D 0.016 15 0.000 0.016 0.000

8 7 0.333 12 0.8 5 0.2 12 0.7 B, D 0.037 24 0.000 0.037 0.000

9 7 0.333 12 0.8 15 0.2 3 0.3 B, D 0.016 15 0.000 0.016 0.000

10 7 0.333 12 0.8 15 0.2 12 0.7 B, D 0.037 24 0.000 0.037 0.000

11 7 0.333 12 0.8 25 0.6 3 0.3 C 0.048 25 0.000 0.000 0.048

12 7 0.333 12 0.8 25 0.6 12 0.7 C 0.112 25 0.000 0.000 0.112

13 8 0.333 2 0.2 5 0.2 3 0.3 A, D 0.004 11 0.004 0.000 0.000

14 8 0.333 2 0.2 5 0.2 12 0.7 A, D 0.009 20 0.009 0.000 0.00015 8 0.333 2 0.2 15 0.2 3 0.3 C 0.004 15 0.000 0.000 0.004

16 8 0.333 2 0.2 15 0.2 12 0.7 A, D 0.009 20 0.009 0.000 0.000

17 8 0.333 2 0.2 25 0.6 3 0.3 C 0.012 25 0.000 0.000 0.012

18 8 0.333 2 0.2 25 0.6 12 0.7 C 0.028 25 0.000 0.000 0.028

19 8 0.333 12 0.8 5 0.2 3 0.3 B, D 0.016 15 0.000 0.016 0.000

20 8 0.333 12 0.8 5 0.2 12 0.7 B, D 0.037 24 0.000 0.037 0.000

21 8 0.333 12 0.8 15 0.2 3 0.3 B, D 0.016 15 0.000 0.016 0.000

22 8 0.333 12 0.8 15 0.2 12 0.7 B, D 0.037 24 0.000 0.037 0.000

23 8 0.333 12 0.8 25 0.6 3 0.3 C 0.048 25 0.000 0.000 0.048

24 8 0.333 12 0.8 25 0.6 12 0.7 C 0.112 25 0.000 0.000 0.112

25 9 0.333 2 0.2 5 0.2 3 0.3 A, D 0.004 12 0.004 0.000 0.000

26 9 0.333 2 0.2 5 0.2 12 0.7 A, D 0.009 21 0.009 0.000 0.00027 9 0.333 2 0.2 15 0.2 3 0.3 C 0.004 15 0.000 0.000 0.004

28 9 0.333 2 0.2 15 0.2 12 0.7 A, D 0.009 21 0.009 0.000 0.000

29 9 0.333 2 0.2 25 0.6 3 0.3 C 0.012 25 0.000 0.000 0.012

30 9 0.333 2 0.2 25 0.6 12 0.7 C 0.028 25 0.000 0.000 0.028

31 9 0.333 12 0.8 5 0.2 3 0.3 B, D 0.016 15 0.000 0.016 0.000

32 9 0.333 12 0.8 5 0.2 12 0.7 B, D 0.037 24 0.000 0.037 0.000

33 9 0.333 12 0.8 15 0.2 3 0.3 B, D 0.016 15 0.000 0.016 0.000

34 9 0.333 12 0.8 15 0.2 12 0.7 B, D 0.037 24 0.000 0.037 0.000

35 9 0.333 12 0.8 25 0.6 3 0.3 C 0.048 25 0.000 0.000 0.048

36 9 0.333 12 0.8 25 0.6 12 0.7 C 0.112 25 0.000 0.000 0.112

6.8% 32.0% 61.1%




Length of Cumulative

CP's Prob Prob

10 0.004 0.00

11 0.004 0.01

12 0.004 0.01

15 0.108 0.12

19 0.019 0.14

20 0.019 0.16

21 0.019 0.18

24 0.224 0.4025 0.599 1.00

Task A Task B Task C Task D

6.8% 32.0% 61.1% 38.8%

Criticality Indices

Expected Project Duration = 23.22

Monte-Carlo Simulation (PERT Example 1)



Task Duration Early Latest Total Expected

Task (Uniform Dist) Start Finish Slack Duration Variance

A 4.99 0 4.99 0.00 6.67 4.00

B 4.75 4.99 9.74 0.00 7.33 1.78C 3.38 4.99 9.74 1.36 7.83 3.36

D 12.20 4.99 21.02 3.83 14.33 1.00

E 5.94 9.74 15.68 0.00 5.00 0.44

F 5.34 15.68 21.02 0.00 4.33 0.44

END 0.00 21.02 21.02 0.00 0.00 0.00

Run Proj ect Duration t(B) t(C) t(D) t(E) t(F)

1 31.07 1 0 0 1 1

2 27.41 0 1 0 1 1

3 23.97 1 0 0 1 1

4 28.93 0 1 0 1 1

5 26.85 1 0 0 1 1

6 28.82 0 0 1 0 0

7 28.77 0 1 0 1 1

197 30.37 0 1 0 1 1

198 29.78 1 0 0 1 1

199 25.33 1 0 0 1 1200 29.70 0 1 0 1 1

Ave 27.13 48.5% 42.0% 9.5% 90.5% 90.5%

Var 16.777

Project Makespan Lower Limit Upper Limit

95% Confidence interval 26.56 27.72

99% Confidence interval 26.37 27.90

Calculating Confidence Intervals



For a confidence interval, we can use the sample meanand the estimated standard error of the mean

where s is the sample standard deviation and n is the

number of trials

Using a normal approximation, a (1- a) two-

sided confidence interval is given by

sX = snX

X -+ za/2 s

New Product Development Projects



START

Lease

Mfg/Office

Space

Identify/hirestaff

Design of

physical unit

Electronics

design Software

Assemble prototype Beta test

prototype

END

Beta test fails (with

probability of 0.25)

and rework is needed

Beta test fails (withprobability of 0.25)

and rework is needed

New Product Development Projects (cont’d)



START

Lease

Mfg/Office

Space

Identify/hire

staff

Design of

physical unit

Electronics

design Software

Assemble prototype Beta test

prototype

END

Beta test fails andrework is needed

Prob = .25

Prob = .75

Critical Chain and the Theory of Constraints (TOC)



• Use deterministic CPM model with buffers to deal with any

uncertainties,

• Place project buffer after last task to protect the customer‟s

completion schedule,

• Exploit constraining resources (make certain that resources are

fully utilized),

• Avoid wasting time slack time by encouraging early task

completions,

• Carefully monitor the status of the buffer(s) and communicate

this status to other project team members on a regular

basis, and• Make certain that the project team is 100 percent focused on

critical chain tasks

Project “Goal” (according to Goldratt): Meet Project Due Date

Project Buffer Defined



• Project Buffer is placed at the end of the project to protect the

customer‟s promised due date

PERT Example #1 Revisited with Project Buffer

Start

Task B

Programming

User Task D User

training

Task E

Implementation

End Project

Buffer

Task A requirements

analysis

Task C Hardware acquisition

Task F

Testing

Calculating Project Buffer Size



For tasks k on critical chain, we can calculate project buffer

using following formula that project will be completed

within worst-case duration estimates around 90 percent of

the time:

For those “who want a scientific approach to sizingbuffers....”

Buffer = •tasks k on cr itic al chain

tk p - k

2

Implications of Project Uncertainty



Assume that the duration of both tasks A and B are described by a

normal distribution with a mean of 30 days

START END

Task A

Task B

What is the probability that the project will be completed within 30

days?

Uncertainty and Worker Behavior



Consider a project with two tasks that must be completed serially

The duration of each task is described by a RV with values Ti (i = 1, 2)

Values of T1 Prob Values of T2 Prob

7 0.3 14 0.5

8 0.4 18 0.5

9 0.38.0 16

Start Task 1 Task 2 End

Parkinson’s Law (Expanding Work)



“Work expands so as to fill the time available for its

completion”

Professor C.N. Parkinson (1957)

Set a deadline D = 24 days

So T(D) = project makespan (function of D) where

E[T(D)] = E(T1) + E(T2) + E[max(0, D - T1 - T2)]

Values of T1 Prob Values of T2 Prob

Project

Makespan Prob

7 0.3 14 0.5 24 0.15

7 0.3 18 0.5 25 0.158 0.4 14 0.5 24 0.2

8 0.4 18 0.5 26 0.2

9 0.3 14 0.5 24 0.15

9 0.3 18 0.5 27 0.15

E[T(D)] = 25 days

Procrastinating Worker



Set a deadline D = 24 daysE’[T(D)] = E(T1) + E(T2) + E{max[0, D - T1 - E(T2)]}

Can show that E[T(D)] ≥ E’[T(D)] ≥ D

What are the implications for project managers?

Values of T1 Prob

E[De lay] =

max[0, D - T1 - E(T2)] E[Makespan]

7 0.3 1 24

8 0.4 0 24

9 0.3 0 25

8 0.3 24.30

Schoenberger’s Hypothesis



An increase in the variability of task durations willincrease the expected project duration….

Schoenberger’s Hypothesis Illustrated



START END

Task A

Task B

Duration of

Task A Probability

Duration of

Task B Probability

12 0.1 10 0.5

14 0.8 15 0.5

16 0.1

14.0 12.5

Schoenberger’s Hypothesis Illustrated



Realization

Task A

Duration

Task B

Duration Probability Max (A, B)

1 12 10 0.05 12

2 14 10 0.4 14

3 16 10 0.05 16

4 12 15 0.05 15

5 14 15 0.4 15

6 16 15 0.05 16

Duration of

Task A Probability

Duration of

Task B Probability

12 0.3 10 0.5

14 0.4 15 0.5

16 0.314.0 12.5

Increasing the variance of Task A:

Results in an increased expected duration = 14.65 days

Expected duration equals 14.55 days

Risk Management



• All projects involve some degree of risk

• Need to identify all possible risks and outcomes

• Need to identify person(s) responsible for managing

project risks

• Identify actions to reduce likelihood that adverse

events will occur

Risk Analysis



Risk Exposure (RE) or Risk Impact =

(Probability of unexpected loss) x (size of loss)

Example: Additional features required by client

Loss: 3 weeksProbability: 20 percent

Risk Exposure = (.20) (3 weeks) = .6 week

How to Manage Project Risks?



Preventive Actions

• Actions taken in anticipation of adverse events

• May require action before project actually begins

• Examples?

Contingency Planning • What will you do if an adverse event does occur?

• “Trigger point” invokes contingency plan

• Frequently requires additional costs

Risk and Contracts



H igh Low Low H igh

Degree of Risk

Contractor Client

Fixed Price Contract Cost Plus Contract

Firm price

Elements

can be

renegotiated Incentives

T&M

with limits

Cost Plus

with

Incentives

Time &

materials

Tornado Diagram



Wage Rate

Direct Labor Hours

Material Units Needed

Early Completion Bonus

Material Unit Cost

Interest rates

Energy costs

Overhead

Project Cost ($000's)

$1290

$1265

$1260

$1310

$1350

$1350

$1380

$1400

$1700

$1720

$1680

$1690

$1640

$1620

$1625

$1760

$1500 $1600 $1800$1700$1400$1300$1200

Sensitivity Chart



Wage Rate 0.85

Direct Labor Hours 0.73

Material Units Needed 0.62

Early Completion Bonus -0.45

Material Un it Cos t 0.42

Inter es t rates 0.28

Energy cos ts 0.19

Overhead 0.10

0 0.5 1.0-0.5

Rank Order Correlation with Total Project Cost

Van Allen Company



Strike Expected

(wks) Prob Duration

3 0.45 1.354 0.3 1.20

5 0.25 1.25

E[Strike Duration] 3.80

Resource Allocation & Leveling



Resource Leveling: Reschedule the noncriticaltasks to smooth resource requirements

Resource Allocation: Minimize project

duration to meet resource availability constraints

Resource Allocation & Leveling



Three types of resources: 1) Renewable resources: “renew” themselves

at the beginning of each time period (e.g.,

workers)

2) Non-Renewable resources: can be used at

any rate but constraint on total number

available

3) Doubly constrained resources: bothrenewable and non-renewable

Resource Leveling



Task B

2 wks

Task E

3 wks

Task C

9 wks

Task D

5 wks

Task F

2 wks

Task A

3 wks

START Task G

5 wksEND

Task Workers Duration ( t j) Early Start Late Start

A 7 3 0 0

B 3 2 0 3C 2 9 3 4

D 10 5 3 3

E 4 3 2 5

F 5 2 2 11

G 6 5 8 8

Resource Leveling: Early Start Schedule



Resource Leveling: Late Start Schedule



Resource Leveling: Microsoft Project



5

10

15

20

25

Workers Ov era ll ocat ed: Al loca te d:

T W T F S S M T W T F S S M T W T F S S M T W T F S

Dec 17, '00 Dec 24, '00 Dec 31, '00 Jan 7, '01

10 10 10 10 10 10 10 10 10 10 16 16 16 16 16 21 21 21

Renewable Resource Allocation Example(Single Resource Type)



Ta sk B 3 w k s

Ta sk D 5 w ks

Ta sk A 4 w k s

Ta sk E

4 w k s

START

END

Ta sk C 1 w k

3 workers

5 workers

6 workers

8 workers

7 workers

Maximum number of workers available = R = 9 workers

Resource Allocation Example: Early Start Schedule




Start

End

Week 1 2 3 4 5 6 7 8 9 10 11 12

No. of Worke rs/wk 8 8 8 11 14 8 8 8 7 7 7 7

Cumulative Workers 8 16 24 35 49 57 65 73 80 87 94 101

"Waste d" worker-wks 1 1 1 - - - - - - - - -

Task B:

5 workers

Task A:

3 workers

Task C:

6 workers

Task E:

7 workersTask D:

8 workers

Resource Allocation Example: Late Start Schedule




Start

End

Week 1 2 3 4 5 6 7 8 9 10 11 12

No. of Workers/wk 5 5 5 11 11 11 11 14 7 7 7 7

Cumulative Workers 5 10 15 26 37 48 59 73 80 87 94 101

"Waste d" worker-wks - - - - - - - - 2 2 2 2

Task B:

5 workers

Task A:

3 workers

Task C:

6 workers

Task E:

7 workersTask D:

8 workers

Resource Allocation HeuristicsRk



n Some heuristics for assigning priorities to available tasks j, where denotes the

number of units of resource k used by task j

n 1) FCFS: Choose first available task

n 2) GRU: (Greatest) resource utilization =

n 3) GRD: (Greatest) resource utilization x task duration =

n 4) ROT: (Greatest) resource utilization/task duration =

n 5) MTS: (Greatest) number of total successors

n 6) SPT: Shortest processing time = min {t j}

n 7) MINSLK: Minimum (total) slack

n 8) LFS: Minimum (total) slack per successor

n 9) ACTIM j: (Greatest) time from start of task j to end of project = CP - LS j

n 10) ACTRES j: (max) (ACTIM j)

n 11) GENRES j: w ACTIM j + (1-w) ACTRES j where 0 ≤ w ≤ 1

R jk

R jk

•

k

R jk / t j•

k

R jk

•

k

t j

Resource Allocation Problem #2



Task A1

6 da sTask A2

4 da s

EndStart Task B1

3 da s

Task C1

2 da s

Task B2

5 da s

Task C2

5 da s

Purple CrewGold Crew

How to schedule tasks to minimize project makespan?



Priority scheme: schedule tasks using total slack (i.e., tasks with

smaller total slack have higher priority)

Task A1 Task B1 Task C1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Task A2 Task B2 Task C2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Gold Crew

Purple

Resource Allocation Example (cont’d)



But, can we do better? Is there a better priority scheme?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Gold Crew

Purple

Microsoft Project Solution (Resource Leveling Option)



Soluti on by: M icrosoft Project 2000

Critical Chain Project Management

Id if h i i l h i f k h d i h ll



• Identify the critical chain: set of tasks that determine the overall

duration of the project

• Use deterministic CPM model with buffers to deal with uncertainty

• Remove padding from activity estimates (otherwise, slack will be

wasted). Estimate task durations at median.

• Place project buffer after last task to protect customer‟s completion

schedule

• Exploit constraining resource(s)

• Avoid wasting slack times by encouraging early task completions

• Have project team focus 100% effort on critical tasks • Work to your plan and avoid tampering

• Carefully monitor and communicate buffer status

Critical Chain Buffers



Project Buffer : placed after last task in project to protect schedule

Feeding Buffers: placed between a noncritical task and a critical task

when the noncritical task is an immediate predecessor of the critical task

Resource Buffers: placed just before a critical task that uses a new

resource type

Critical Chain Illustrated



Task A1

6 da s

Task A2

4 da s

EndStart

Task B1

3 da s

Task C1

2 da s

Task B2

5 da s

Task C2

5 da s

Resource Buffers

Feeding Buffers

Non-Renewable Resources

12 units



Task B5 wks

Task D2 wks

Task C3 wks

Task A6 wksST ART END

6 units

10 units

8 units

Task Duration

No. of Nonrene wable

Resource s Units

Needed Early Start Late Start

A 6 6 0 0

B 5 12 6 6

C 3 10 6 8

D 2 8 11 11

Non-Renewable Resources: Graphical Solution



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

40

36

32

28

24

20

16

12

8

4

Cumulative ResourcesSupplied

C u m u l a t i v e R e s o u r c e s

Cumulative Resources

Required

Resource Allocation Problem #3

Issue: When is it better to “team” two or more



Issue: When is it better to team two or more

workers versus letting them work separately?

• Have 2 workers, Bob and Barb, and 4 tasks: A, B, C, D

• Bob and Barb can work as a team, or they can work separately

• When should workers be assigned to tasks? Which configurationdo you prefer?

How to Assign Project Teams?



Configuration #1

Bob and Barb work jointly on all four tasks; assume that they can complete each

task in one-half the time needed if either did the tasks individually

A C

B D

Start End

Configuration #2 Bob and Barb work independently. Bob is assigned to tasks A and C; Barb is

assigned to tasks B and D

Bob and Barb: Configuration #1



TASK A TASK B TASK C TASK D

Duration Prob Duration Prob Duration Prob Duration Prob

6 0.33 9 0.667 12 0.6 10 0.255 0.33 6 0.333 7 0.4 6 0.75

4 0.33

Expected

duration 5.0 8.0 10.0 7.0

Configuration #1

Bob and Barb work jointly on all four tasks.

What is the expected project makespan?


Bob and Barb work independently. Bob is assigned to tasks A and C; Barb is





Realization # A B C D

Bob

A + C

Barb

B + D

max

(A+C,

B+D) Prob

1 6 9 12 10 18 19 19 0.03

2 6 9 12 6 18 15 18 0.10

3 6 9 7 10 13 19 19 0.02

4 6 9 7 6 13 15 15 0.07

5 6 6 12 10 18 16 18 0.02

6 6 6 12 6 18 12 18 0.05

7 6 6 7 10 13 16 16 0.01

8 6 6 7 6 13 12 13 0.039 5 9 12 10 17 19 19 0.03

10 5 9 12 6 17 15 17 0.10

11 5 9 7 10 12 19 19 0.02

12 5 9 7 6 12 15 15 0.07

13 5 6 12 10 17 16 17 0.02

14 5 6 12 6 17 12 17 0.05

15 5 6 7 10 12 16 16 0.01

16 5 6 7 6 12 12 12 0.03

17 4 9 12 10 16 19 19 0.03

18 4 9 12 6 16 15 16 0.10

19 4 9 7 10 11 19 19 0.02

20 4 9 7 6 11 15 15 0.07

21 4 6 12 10 16 16 16 0.02

22 4 6 12 6 16 12 16 0.05

23 4 6 7 10 11 16 16 0.01

24 4 6 7 6 11 12 12 0.03




Bob and Barb work independently. Bob is assigned to tasks A and C; Barb is


max (A+C,

B+D) Prob

Cumulative

Prob

12 0.07 0.07

13 0.03 0.10

15 0.20 0.30

16 0.20 0.5017 0.17 0.67

18 0.17 0.83

19 0.17 1.00

Expected Project Makespan: 16.42

Parallel Tasks with Random Durations



START END

Task B

Task A

• Assume that both Tasks A and B have possible durations:

8 days with probability = 0.5

10 days with probability = 0.5

• What is expected duration of project? (Is it 9 days?)

Project Monitoring and Control



n “It is of the highest importance inthe art of detection to be able to

recognize, out of a number of acts,

which are incidental and which are

vital. Otherwise your energy and

attention must be dissipated instead

of being concentrated.”

Sherlock Holmes

Status Reporting?



One day my Boss asked me to submit a statusreport to him concerning a project I was working

on. I asked him if tomorrow would be soon enough.

He said, "If I wanted it tomorrow, I would have

waited until tomorrow to ask for it!"

New business manager, Hallmark Greeting Cards

Control System Issues



n What are appropriate performance metrics?

n What data should be used to estimate the value of each

performance metric?

n

How should data be collected? From which sources? Atwhat frequency?

n How should data be analyzed to detect current and future

deviations?

n How should results of the analysis be reported? To whom?

How often?

Controlling Project Risks

K i t t l i k d i j t



Key issues to control risk during projecct:

(1) what is optimal review frequency, and

(2) what are appropriate review acceptance levels

at each stage?

“Both over -managed and under-managed

development processes result in lengthy design

lead time and high development costs.”

Ahmadi & Wang. “Managing Development Risk in

Product Design Processes”, 1999

Project Control & System Variation



Common cause variation: “in-control” or normal

variation

Special cause variation: variation caused by forces

that are outside of the system

According to Deming:

• Treating common cause variation as if it were special cause variation

is called “tampering”

• Tampering always degrades the performance of a system

Control System Example #1

Project plan: We estimate that a task



n Project plan: We estimate that a task

will take 4 weeks and require

n 1600 worker-hours

At the end of Week 1, 420 worker-hours

have been used

Is the task “out of control”?

Control System Example (cont’d)

Week 2: Task expenses = 460 worker-hours



Week 2: Task expenses 460 worker hours


370380

390

400

410

420

430

440

450

460

470

1 2 3 4

Week

Week

Planned Cost

(BCWS) Actual Cost

CumulativeActual Cost

(ACWP)

1 400 420 420

2 400 460 880

Control System Example (cont’d)

Week 3: Task expenses = 500 worker-hrs




Week

Planned cost

(worker-hours)

Actual cost

(worker-hours)

Cumulative cost

(worker-hours)

1 400 420 420

2 400 460 880

3 400 500 1380

0

100

200

300

400

500

600

1 2 3 4

Week

Earned Value Analysis

I t t t h d l d k f d



• Integrates cost, schedule, and work performed

• Based on three metrics that are used as the basic

building blocks:

BCWS: Budgeted cost of work scheduled

ACWP: Actual cost of work performed

BCWP: Budgeted cost of work performed

Schedule Variance (SV)



Schedule Variance (SV) = difference between value of

work completed and value of scheduled work

Schedule Variance (SV) = Earned Value - Planned Value

= BCWP - BCWS

Cost Variance (CV)



Cost Variance (CV) = difference between value of work completed and actual

expenditures

Cost Variance (CV) = Earned Value - Actual Cost

= BCWP - ACWP

Earned Values Metrics Illustrated



W o r k e r - H o u r s

Week 1 Week 2 Week 3 Week 4 Week 5 Week 6

Present time BAC

Actual Cost(ACWP)

Earned Value

(BCWP)

Planned Value

(BCWS)

Schedule Variance

(SV)

Cost Variance

(CV)

Relative Measure: Schedule Index

C



Schedule Index (SI) =

BCWP

BCWS

If SI = 1, then task is on schedule

If SI > 1, then task is ahead of schedule

If SI < 1, then task is behind schedule

Relative Measure: Cost Index



Cost Index (CI) = BCWPACWP

If CI = 1, then work completed equals

payments (actual expenditures)

If CI > 1, then work completed is ahead

of payments

If CI < 1, then work completed is behindpayments (cost overrun)

Example #2

W E E K 1 2 3 4 5 6 7 8 9 10



1 2 3 4 5 6 7 8 9 10

6 6 6 8 10

12 12 12

10 10 12 12 12

Scheduled

Worker-Hrs 6 6 6 20 22 22 10 12 12 12

Scheduled

Worker-Hrs

(BCWS) 6 12 18 38 60 82 92 104 116 128

Task A (36 worke r-hrs)

Task B (36 worker-hrs)

Task C (56 worker-hr


Progress report at the end of week #5:



Week 1 2 3 4 5

Task A 15% 30% 40% 60% 80%

Task B 25% 65%

Task C Not started yet

Cumulative Percent of Work Completed:

Worker-Hours Charged to Project:

Week 1 2 3 4 5

Task A 5 6 8 10 10

Task B 15 10

Task C Not started yet


Progress report at the end of week #5:



W E E K 1 2 3 4 5 6 7 8 9 10

Cumulative

Scheduled

Worker-Hrs

(BCWS) 6 12 18 38 60 82 92 104 116 128Actual Worker-

Hrs Used

(ACWP) 5 11 19 44 64

Earned Value

(BCWP) 5.4 10.8 14.4 30.6 52.2

Schedule

Variance (SV) -0.6 -1.2 -3.6 -7.4 -7.8

Cost Variance

(CV) 0.4 -0.2 -4.6 -13.4 -11.8


140

BAC



0

20

40

60

80

100

120

1 2 3 4 5 6 7 8 9 10

Week

ACWP

BCWP

BCWS

Schedule

Variance

Cost

Variance

Using a Fixed 20/80 Rule

Cumulative Percent of Work Completed:



W E E K 1 2 3 4 5 6 7 8 9 10

Cumulative

Scheduled

Worker-Hrs

(BCWS) 6 12 18 38 60 82 92 104 116 128Actual Worker-

Hrs Used

(ACWP) 5 11 19 44 64

Earned Value(BCWP) 7.2 7.2 7.2 14.4 14.4Schedule

Variance ( SV) 1.2 -4.8 -10.8 -23.6 -45.6Cost Variance

(CV) 2.2 -3.8 -11.8 -29.6 -49.6

Week 1 2 3 4 5Task A 20% 20% 20% 20% 20%

Task B 20% 20%

Task C Not s tar ted yet

Using a Fixed 20/80 Rule

140



0

20

40

60

80

100

120

140

1 2 3 4 5 6 7 8 9 10

Week

BCWP

ACWP

BCWS

Updating Forecasts: Pessimistic Viewpoint

Assumes that rate of cost overrun will continue



= (64/52.2) 128 = 1.23 x 128 = 156.94 worker-hrs

Estimate at Completion (EAC) = ACWP

BCWPBAC = 1

CIBAC .

Assumes that rate of cost overrun will continue

for life of project….

Updating Forecasts: Optimistic Viewpoint

Assumes that cost overrun experienced to date



Estimate at Completion (EAC) = BAC - CV = 128 + 11.8 = 139.8 worker-hrs .

Assumes that cost overrun experienced to date

will cease and no further cost overruns will beexperienced for remainder of project life…

Multi-tasking with Multiple Projects



Project A Project B

A-1 B-1 A-2 B-2 A-3 A-4B-3 B-4

Consider two projects with and without multi-tasking

How to prioritize your work when you have multipleprojects and goals?

Due-Date Assignment with Dynamic Multiple Projects



• Projects arrive dynamically (common situation for bothmanufacturing and service organizations)

• How to set completion (promise) date for new projects?

• Firms may have complete control over due-dates or only partial

control (i.e., some due dates are set by external sources)

• How to allocate resources among competing projects and tasks (so

that due dates can be realized)?

• What are appropriate metrics for evaluating various rules?

What Does the Research Tell Us?

• Study by Dumond and Mabert* investigated four due date assignment



rules and five scheduling heuristics• Simulated 250 projects that randomly arrive over 2000 days

• average interarrival time = 8 days

• 6 - 49 tasks per project (average = 24); 1 - 3 resource types

• average critical path = 31.4 days (range from 8 to 78 days)

• Performance criteria: 1) mean completion time

2) mean project lateness

3) standard deviation of lateness

4) total tardiness of all projects

• Partial and complete control on setting due dates

* Dumond, J. and V. Mabert. “Evaluating Project Scheduling and Due Date Assignment Procedures:

An Experimental Analysis” Management Science, Vol 34, No 1 (1988), pp 101-118.

Experimental Results



• No one scheduling heuristic performs best across all due datesetting combinations

• Mean completion times for all scheduling and due date rules not

significantly different

• FCFS scheduling rules increase total tardiness • SPT-related rules do not work well in PM (SASP)

• Best to use more detailed information to establish due dates

Project Management Maturity Models

M h d l i i i ’ l l f



• Methodologies to assess your organization’s current level of

PM capabilities

• Based on extensive empirical research that defines “best

practice” database as well as plan for improving PM process

• Process of improvement describes the PM process from

“ineffective” to “optimized”

• Also known as “Capability Maturity Models”

PM Maturity Model Example*

1) Ad-Hoc The project management process is described as disorganized, and occasionally even



chaotic. Systems and processes are not defined. Project success depends on individual effort.

Chronic cost and schedule problems.

2) Abbreviated: Some project management processes are established to track cost, schedule,

and performance. Underlying disciplines, however, are not well understood or consistently

followed. Project success is largely unpredictable and cost and schedule problems are the norm.

3) Organized: Project management processes and systems are documented, standardized, and

integrated into an end-to-end process for the company. Project success is more predictable. Cost

and schedule performance is improved.

4) Managed: Detailed measures of the effectiveness of project management are collected and used

by management. The process is understood and controlled. Project success is more uniform.

Cost and schedule performance conforms to plan.

klastorin (traduccion)

Documents