n ajakjaodolffoofdodosoksd

Importance of Project Management

Projects represent change and allow organizations to
effectively introduce new products, new
process, new programs

Project management offers a means for dealing with
dramatically reduced product cycle times

Projects are becoming globalized making them more
difficult 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 on
approximately 175,000 information technology projects.

Only 26 percent of these projects are completed on time and
within budget.

The average cost for a development project for a large company
is more than $2 million.

Project management is an $850 million industry and is expected
to 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 project
management give way to a fundamental difference:
process management seeks to eliminate variability
whereas 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 factors
contribute 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 a
Photo - JPEG decompressor
are needed to see this picture.
Have You Ever Lost Sight of the
Project Goals?
Not all Projects Are Alike
[in IT projects], if you ask people whats 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 youre halfway done. Weve moved from physical to non-physical
deliverables.
J. Vowler (March, 2001)
Engineering projects = task-centric
IT projects = resource-centric
Shenhars 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 4
Formation & 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
Concept
Design
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 F
t
= net cash flow in period t (t = 0, 1,..., T)
F
0
=initial cash investment in time t =0
r = discount rate of return (hurdle rate)

NPV =
F
t
1 + r
t

t = 0
T
Internal Rate of Return (IRR)

Find value of r such that NPV is equal to 0

F
0
+
F
1
1 + r
+
F
2
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 Ne t
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 Development
$18 million annual research cost for 2 years
60% probability of succ ess
Phase II Market Development
Undertaken only if product development is successful
$10 million annual expenditure for 2 years to develop marketing and
distribution c hannels (net of any revenues earned in test marketing)
Phase III Sales
Proceeds only if Phase I and II verify opportunity.
Production is subcontracted and all cash flows are after-tax and occur
at year' s end.
The results of Phase II (available at the end of year 4) identify the
product's market potential as indic ated below:
DCF Project Example (contd)
Year Expected Cash Flow (in $ million)
1 -18
2 -18
3 0. 6 (-10) = - 6
4 0. 6 (-10) = - 6
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 cte d
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; ignores
the 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 = p
t

Probability = 1 - p
t

Launch New
Product
Commercial
Failure (with net
benefit = 0)
Commercial Success
(with net benefit =
NPV)
Probability = p
c

Probability = 1 - p
c

Risk class 1 Risk class 2
DCF Example Revisited
Discount rate r
1
Discount rate r
2

Research &
Product
Development
Development
Succeeds
Probability = p
t

Development Fails
Probability = 1 - p
t

Market
Development
Product Demand
High
0.3
Product Demand
Medium
Product Demand
Low
0.5
0.2
Drop project
Ranking/Scoring Models
Profit abilit y/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.
Organizat ion's Strat egy
1) 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 risk of project
6) Adequate 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 Factors
1) Impact on environmental standards?
2) Impact on workforce safety?
3) Impact on qualit y?
4) Social/polit ical implications
Scoring Attributes
v
i
x
i
=
1 - exp L - x
i
1 - expL - 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
v
i
x
i
=
x
i
- 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
Response
A
t
t
r
i
b
u
t
e

V
a
l
u
e
Linear Scale
Exponential Scale
Ranking/Scoring Example
Attribute Measurement Scale
Attribute
Weight (w
i
)
1) Does project increase market share? unlikely likely 30%
2) Is new facility needed? yes no 15%
3) Are there safety concerns? likely unsure no 10%
4) Likelihood of successful technical 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 Scale
Project 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 Scale
Project 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 (contd)
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 necessarily
independent) 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 Control Improve
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 $50
Budget
Limit (B
t
)
$120 $20 $40 $55
Phases of Project Management
n Project formulation and selection
n Project planning
u Summary statement
u Work breakdown structure
u Organization plan
u risk management
u Subcontracting and bidding process
n Project scheduling
u Time and schedule
u Project budget
u Resource allocation
u Equipment and material purchases
n Monitoring and control
u Cost control metrics
u Change orders
u Milestone reports
Project Planning
n Summary Statement
u 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 Plan
u 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. Its easy to get
caught up in the idea that a project plan should detail
everything everybody is going to do on the project. This
springs from the screwy logic that a project managers 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 (contd)
Benchmarking
Modular approach
Parametric techniques
Learning effects

Beta Distribution
Completion time of task j
Optimistic Time t
j
o
Pessimistic Time t
j
p
Time
Probability density
function
Expected duration =
Most Likely Time = t
m
Beta Distribution
For each task j, we must make three estimates:
most optimistic time
most pessimistic time
most likely time
t
j
o
t
j
p
t
j
m
Expected duration
j
=
t
j
o
+ t
j
p
+ 4t
j
m
6
Variance of task j =
j
2
=
t
j
p
- t
j
o 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 the
doors 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 15
33 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 15
23 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 Im happy, I know some cool
people will like it. Designing user interfaces by
committee does not work very well; they need to be
coherent. As for schedule, Im 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 members
L = 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 a
plain 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 anothers speech. Thus, the Lord dispersed
them 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 positive
characteristics 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 (contd)
Experiment conducted using MBA students at UW and
Seattle U using computer based simulation of pre-operational
testing phase of nuclear power plant*
Total of 14 project teams (2 - 4 person project teams) with a
total 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 Individuals 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 Low
Performing Groups
Decision Making Effectiveness: High vs Low
Performing 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
Functional
Organization
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 1960s as an
alternative to traditional means of project
teams
Became popular in 1970s and early 1980s
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 schedule
2) 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 programs
16% 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 products
10% 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
Sche dule
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 Balance d 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
*Statistically significant 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 company
communicates.....
HOW a company
communicates.....
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 = $100M
Two 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 only
that 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. The
two 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 relationships
among 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 has
started
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
ES
Start
= 0
LF
Start
= 0
ES
A
= 0
LF
A
= 8
ES
B
= 7
LF
B
= 11
ES
C
= 0
LF
C
= 11
ES
End
= 11
LF
End
= 11
ES
j
= Earliest starting time for task (milestone) j
LF
j
= Latest finish time for task (milestone) j
AON Precedence Network: Microsoft Project
Start
1 0d
Wed 12/20/00 Wed 12/20/00
Task 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 1/3/01 Wed 1/3/01
Task 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
START
= 0
LF
START
= 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 32
3 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 14
C 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
t
i
Example 2: Calculating Total Slack (TS
i
)
Task or
Milestone
Duration
( )
Earliest
Start Time
(ESi)
Lastest
Finish Time
(LFi)
Total Slack
(TSi)
Critical
Task?
START 0 0 0 0 Yes
A 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
t
i
Total Slack for task i = TS
i
= LF
i
- ES
i
- t
i
Slack (Float) Definitions (for task i)
Total Slack (TS
i
) = LF
i
- ES
i
- t
i

Free Slack (FS
i
) = ES
i,min
- ES
i
- t
i

where ES
i,min
= minimum early start time of all tasks that
immediately follow task i
= min (ES
j
for all task j S
i
)
Safety Slack (SS
i
) = LF
i
- LF
i,max
- t
i

where LF
i,max
= maximum late finish time of all tasks that
immediately precede task i
= min (LF
j
for all task j P
i
)
Independent Slack (IS
i
) = max (0, ES
i,min


- LF
i,max
- t
i
)

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 + t
i
= STARTi + duration of task i
Example #2: Excel Solver Model
Gantt Chart
Microsoft Project 4.0
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)
Project Budgeting (contd)
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
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?
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
START
= 0
LF
START
= 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
Project Budget Example
Task or
Milestone
Duration
(tj)
Early Start
Time (ESj)
Latest Start
Time (LSj)
No. of
Resource A
worke rs
No. of
Resource B
worke rs
Material
Costs
Direct Labor
Cost/wk
Labor +
Materials
START 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 12 14 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
Project Budget Example (contd)
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
Weekl y Subtotals 19665 19200 19200 19200 19200 19200 19200 19200 19200 10400 10400 10400
Cumulati ve 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
Weekl y Subtotals 1140 800 800 800 9725 9600 9600 9600 19200 19200 19200 19200
Cumulati ve 1140 1940 2740 3540 13265 22865 32465 42065 61265 80465 99665 118865
W e e k
W e e k
Cumulative Costs
Range of
feasible budgets
Weekly Costs (Cash Flows)
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
Cash Flow Example
M1
END
START
Task B
8 mos
Receive payment
of $3000
Receive payment
of $3000
Make payment
of $5000
Task C
4 mos
Task A
2 mos
M2
Task D
8 mos
Task E
3 mos
Cash Flow Example: Solver Model
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
Task A
4 wks
Task B
8 wks
Task C
5 wks
Task D
6 wks
Task E
2 wks
Task F
3 wks
End
Start
2 units
30 units
LS
A
= 0 LS
B
= 4 LS
C
= 12
LS
D
= 6 LS
E
= 12 LS
F
= 14
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
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
Time-Cost Tradeoff Example (contd)
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
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 (bj) = Increase in cost by
reducing task by one time unit
Normal time = Crash time =
Normal
cost =
Crash
cost =
t
j
N
t
j
c
C
j
c
C
j
N
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)
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
S
i
+ t
i
for all tasks i P
j

for all tasks in project
END = T
max

t
j
, S
j
0
C
j
N
C
j
c
t
j
c
t
j
N
b
j
=
C
j
c
- C
j
N
t
j
c
- t
j
N
b
j
t
j
j

t
j
c
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 T
max

L = number of time periods project is delayed beyond
deadline T
max


Minimize Total Costs = + I (END) + P L

b
j
t
j
j

QUESTION: HOW TO DEFINE L?
Software Project Schedules
Observe that for the programmer, as for the chef, the urgency of
the patron may govern the scheduled completion of the task, but it
cannot govern the actual completion. An omelet, promised in ten
minutes, may appear to be progressing nicely. But when it has not
set in ten minutes, the customer has two choices--wait or eat it
raw. Software customers have the same choices.
The cook has another choice; he can turn up the heat. The
result is often an omelet nothing can save--burned in one part, raw
in another.


F.P. Brooks, The Mythical Man-Month,
Datamation, Vol 20, No 12 (Dec, 1974), pp.
44-52.
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
Weeks
Coding
No. of
Coordination
Weeks
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. 70
10 3. 33 45 48. 33
11 3. 03 55 58. 03
Brooks Law
Adding manpower to a late
software 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
Projects
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
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
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., ES
k
) is now sum of independent RVs,
central limit theorem specifies that ESk is approximately
normally distributed with mean E[ES
k
] and variance Var[ES
k
]
where there exists s paths to task k
Expected early start time of task k = E ES
k
=
max
s

j
tasks j on path s
Classic PERT Model (contd)
Expect Project Duration = E[E S
END
] =
j
tasks j on CP
Variance of Project Duration = Var[E S
END
] =
j
2

tasks j on CP
Thus, expected project duration is defined as:
Using central limit theorem and standard normal distribution:
P ES
END
T
max
= P

z
T
max
- E ES
END
Var

ES
END
PERT Example #1
Duration Estimates Expected
Task Description Predecessors Optimistic Pessimistic Likely Duration Variance
A Requi rements Anal ysis none 2 14 6 6.67 4.00
B Programmi ng A 4 12 7 7.33 1.78
C Hardware acqui si tion A 2 13 8 7.83 3.36
D User trai ni ng A 12 18 14 14.33 1.00
E Impl ementati on B, C 3 7 5 5.00 0.44
F Testing E 3 7 4 4.33 0.44
END End of proj ect D, F 0 0 0 0.00 0.00
Start
Task A
Requirements
Analysis
Task C
Hardware
Acquisit ion
Task B
Programming
Task F
Test ing
Task D
User
Training
Task E
Implementation
End
PERT Example #1 (contd)
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 Expected Duration = 23.83 Expected CP ={Start, A, C, E, F, End}
PERT Variance = 8.250
Start
Task A
Requirements
Analysis
Task C
Hardware
Acquisit ion
Task B
Programming
Task F
Test ing
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 (contd)
Task A Task B Task C Task D Critical Prob of Length 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 0.000
15 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 0.000
27 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%
Example #3 (contd)
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 0.40
25 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 1.78
C 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 1
200 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 mean
and 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


s
X
=
s
n
X
X
-
+
z
a/2
s
X
New Product Development Projects
START
Lease
Mfg/Office
Space
Identify/hire
staff
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 (with
probability of 0.25)
and rework is needed
New Product Development Projects (contd)
START
Lease
Mfg/Office
Space
Identify/hire
staff
Design of
physical unit
Electronics
design
Software
Assemble prototype
Beta test
prototype
END
Beta test fails and
rework 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 customers
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
customers 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 sizing
buffers....
Buffer =

tasks k on critical chain

t
k
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 T
i
(i = 1, 2)
Values of T
1 Prob
Values of T
2 Prob
7 0.3 14 0.5
8 0.4 18 0.5
9 0.3
8.0 16
Start Task 1 Task 2 End
Parkinsons 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(T
1
) + E(T
2
) + E[max(0, D - T
1
- T
2
)]
Values of T
1 Prob
Values of T
2 Prob
Project
Makespan
Prob
7 0. 3 14 0. 5 24 0. 15
7 0. 3 18 0. 5 25 0. 15
8 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 days
E[T(D)] = E(T
1
) + E(T
2
) + E{max[0, D - T
1
- E(T
2
)]}
Can show that E[T(D)] E[T(D)] D
What are the implications for project managers?
Values of T
1 Prob
E[Delay] =
max[0, D - T
1
- E(T2)] E[Makespan]
7 0. 3 1 24
8 0. 4 0 24
9 0. 3 0 25
8 0. 3 24. 30
Schoenbergers Hypothesis
An increase in the variability of task durations will
increase the expected project duration.
Schoenbergers 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
Schoenbergers 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 0.3
14.0 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 weeks
Probability: 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
High Low Low High
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
1310
$1350
$1350
$1380
$1400
$1700
$1720
$1680
$1690
$1640
$1620
$1625
$1760
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 Unit Cost 0. 42
Interest rates 0. 28
Energy costs 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 1.35
4 0.3 1.20
5 0.25 1.25
E[Strike Duration] 3.80
Resource Allocation & Leveling
Resource Leveling: Reschedule the noncritical
tasks 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: both
renewable 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 wks
END
Task Workers Duration (tj) Early Start Late Start
A 7 3 0 0
B 3 2 0 3
C 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 Overall ocated: Al l ocated:
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
Maximum number of workers available = R = 9 workers
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 workers
Task D:
8 workers
Resource Allocation Example: Late Start Schedule
Maximum number of workers available = R = 9 workers
Start
End
Week 1 2 3 4 5 6 7 8 9 10 11 12
No. of Worke rs/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 workers
Task D:
8 workers
Resource Allocation Heuristics
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
j
k
R
j
k

k
R
j
k
/ t
j
k
R
j
k

k
t
j
Resource Allocation Problem #2
Task A1
6 days
Task A2
4 days
End
Start
Task B1
3 days
Task C1
2 days
Task B2
5 days
Task C2
5 days
Purple Crew Gold 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
Crew
Resource Allocation Example (contd)
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
Crew
Microsoft Project Solution (Resource Leveling Option)
Solution by: Microsoft Project 2000
Critical Chain Project Management
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 customers 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 days
Task A2
4 days
End
Start
Task B1
3 days
Task C1
2 days
Task B2
5 days
Task C2
5 days
Resource Buffers
Feeding Buffers
Non-Renewable Resources
Task B
5 wks
Task D
2 wks
Task C
3 wks
Task A
6 wks
ST ART END
6 units
12 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 Resources
Supplied
C
u
m
u
l
a
t
i
v
e

R
e
s
o
u
r
c
e
s
Weeks
Cumulative Resources
Required
Resource Allocation Problem #3
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 configuration
do 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 0.25
5 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: Configuration #2
Bob and Barb work independently. Bob is assigned to tasks A and C; Barb is
assigned to tasks B and D
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 0.03
9 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: Configuration #2
Bob and Barb work independently. Bob is assigned to tasks A and C; Barb is
assigned to tasks B and D
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 0.50
17 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 in
the 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 status
report 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? At
what 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
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
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 (contd)
Week 2: Task expenses = 460 worker-hours






Is the task out of control?
370
380
390
400
410
420
430
440
450
460
470
1 2 3 4
We ek
C
o
s
t

(
i
n

w
o
r
k
e
r
-
h
o
u
r
s
)
Week
Planned Cost
(BCWS) Actual Cost
Cumulative
Actual Cost
(ACWP)
1 400 420 420
2 400 460 880
Control System Example (contd)
Week 3: Task expenses = 500 worker-hrs






Is the task out of control?
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
We ek
W
o
r
k
e
r
-
h
o
u
r
s
Earned Value Analysis
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
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) =
BCWP
ACWP
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 behind
payments (cost overrun)
Example #2
W E E K
1 2 3 4 5 6 7 8 9 10
6 6 6 8 10
12 12 12
10 10 12 12 12
Weekly
Scheduled
Worker-Hrs 6 6 6 20 22 22 10 12 12 12
Cumulative
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 worke r-hrs)
Example #2 (contd)
Week 1 2 3 4 5
Task A 15% 30% 40% 60% 80%
Task B 25% 65%
Task C Not started yet
Progress report at the end of week #5:
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
Example #2 (contd)
Progress report at the end of week #5:
W E E K
1 2 3 4 5 6 7 8 9 10
Cumulative
Sche duled
Worker-Hrs
(BCWS) 6 12 18 38 60 82 92 104 116 128
Actual Worker-
Hrs Use d
(ACWP) 5 11 19 44 64
Earned Value
(BCWP) 5.4 10.8 14.4 30.6 52.2
Sche dule
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
Example #2 (contd)
0
20
40
60
80
100
120
140
1 2 3 4 5 6 7 8 9 10
We ek
P
e
r
f
o
r
m
a
n
c
e

M
e
t
r
i
c
ACWP
BCWP
BCWS
Schedule
Variance
Cost
Variance
BAC
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
Sche duled
Worker-Hrs
(BCWS) 6 12 18 38 60 82 92 104 116 128
Actual Worker-
Hrs Use d
(ACWP) 5 11 19 44 64
Earned Value
(BCWP) 7.2 7.2 7.2 14.4 14.4
Sche dule
Variance (SV) 1.2 -4.8 -10.8 -23.6 -45.6
Cost Variance
(CV) 2.2 -3.8 -11.8 -29.6 -49.6
Week 1 2 3 4 5
Task A 20% 20% 20% 20% 20%
Task B 20% 20%
Task C Not started yet
Using a Fixed 20/80 Rule
0
20
40
60
80
100
120
140
1 2 3 4 5 6 7 8 9 10
Week
C
o
s
t

(
i
n

W
o
r
k
e
r
-
h
o
u
r
s
)
BCWP
ACWP
BCWS
Updating Forecasts: Pessimistic Viewpoint



= (64/52.2) 128 = 1.23 x 128 = 156.94 worker-hrs
Estimate at Completion (EAC) =
ACWP
BCWP
BAC =
1
CI
BAC .
Assumes that rate of cost overrun will continue
for life of project.
Updating Forecasts: Optimistic Viewpoint
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 be
experienced 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-4 B-3 B-4
Consider two projects with and without multi-tasking
How to prioritize your work when you have multiple
projects and goals?
Due-Date Assignment with Dynamic Multiple Projects
Projects arrive dynamically (common situation for both
manufacturing 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 date
setting 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
Methodologies to assess your organizations 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.
5) Adaptive: Continuous improvement of the project management process is enabled by feedback
from the process and from piloting innovative ideas and technologies. Project success is the
norm. Cost and schedule performance is continuously improving.
* source: The Project Management Institute PM Network (July, 1997), Micro Frame Technologies, Inc. and
Project Management Technologies, Inc. (http://pm32.hypermart.net/)