Project

Management
Journal
Volume 35, Number 3 The Professional Research Journal
SEPTEMBER 2004 of the Project Management Institute

PAPERS

5 PERT SCHEDULING WITH RESOURCE CONSTRAINTS USING

QUALITATIVE SIMULATION GRAPHS
Ricki G. Ingalls and Douglas J. Morrice

15 THROUGH THE LOOKING GLASS: EXAMINING THEORY

DEVELOPMENT IN PROJECT MANAGEMENT WITH THE
RESOURCE-BASED VIEW LENS
Kam Jugdev

27 THE IMPACT OF OWNER REPRESENTATIVES IN A

DESIGN-BUILD CONSTRUCTION ENVIRONMENT
Denis R. Petersen and E. Lile Murphree Jr.

39 UNDERSTANDING THE ROLE OF “VISION”

IN PROJECT SUCCESS
Dale Christenson and Derek H.T. Walker

53 WHY MONTE CARLO SIMULATIONS OF PROJECT NETWORKS

CAN MISLEAD
Terry Williams
PMBOK® Guide–Third Edition
ISBN: 193069945X, paperback,
$39.95 Member $49.95 Nonmember
ISBN: 1930699506, CD-ROM,
$39.95 Member $49.95 Nonmember
Due for release in October
Project management is a relatively young competency, still
fighting for full recognition as a profession. Yet the role of proj-
ect management is fast becoming an indispensable one within
organizations across all industries. Organizations, and their
senior executives, can no longer ignore the significance of
effective project management; it bridges the gap between
strategy and results. This shift in thinking and elevation of
project management in importance to the “C-Level” has
further heightened the need for and availability of best prac-
tice and methodology resources.
The first edition of the PMBOK® Guide was a
collaborative effort by practitioners and academics
trying to define project management and its role in
business. Was project management a formulaic practice
of applying what is done on most projects? Or was
it a creative enterprise designed to assist learning
and action in organizations?
When the PMBOK® Guide – 2000 Edition, with more than one million copies in circulation,
was published four years ago, PMI received thousands of valuable recommendations for improvements.
Those recommendations have been reviewed and, where appropriate, incorporated into the new Third Edition.
The result is an updated version that offers increased clarity and emphasis on processes and the treatment of processes.
Process Groups are now mapped to show process integration, and various processes have been added, including:
• Develop Project Charter • Create Work Breakdown Structure
• Develop Preliminary Project Scope Statement • Activity Resource Estimating
• Monitor and Control Project Work • Manage Project Team.
• Close Project

All of the process inputs, tools and techniques, and outputs have been revised to support the improved integration and map-
ping of the processes. Process flow diagrams have also been added to eight chapters to provide enhanced support to the inte-
gration of processes. An introduction also has been added that describes the process flow diagrams and provides a legend of
the symbols.
The Project Team also made an effort to ensure better consistency of material within and between chapters and to make sure
all figures and tables have been updated for clarity and completeness. An expanded index and glossary reflect the changes the
project management profession has experienced during the three years since the previous edition.
In addition, many chapters have been rewritten and expanded to improve descriptions of processes and Knowledge Areas.
The PMBOK® Guide – Third Edition defines the project management office’s role and responsibility in an organization and
introduces the concept of a project management system.
Finally, those familiar with the PMBOK® Guide will notice improvements and changes that are occurring, literally, in real
time. The PMBOK® Guide is changing as quickly as the profession itself and as project management continues to grow and
evolve, so, too, will this reference that serves as the world’s de facto standard.
Project
Management
Journal
The Professional Research Journal of the Project Management Institute
Volume 35, Number 3
SEPTEMBER 2004
3 From the Editor
Christophe N. Bredillet, PhD, MBA, Ingénieur EC Lille

PAPERS

5 PERT SCHEDULING WITH RESOURCE CONSTRAINTS USING

QUALITATIVE SIMULATION GRAPHS
Ricki G. Ingalls and Douglas J. Morrice

15 THROUGH THE LOOKING GLASS: EXAMINING THEORY

DEVELOPMENT IN PROJECT MANAGEMENT WITH THE
RESOURCE-BASED VIEW LENS
Kam Jugdev

27 THE IMPACT OF OWNER REPRESENTATIVES IN A

DESIGN-BUILD CONSTRUCTION ENVIRONMENT
Denis R. Petersen and E. Lile Murphree Jr.

39 UNDERSTANDING THE ROLE OF “VISION” IN PROJECT SUCCESS

Dale Christenson and Derek H.T. Walker

53 WHY MONTE CARLO SIMULATIONS OF PROJECT NETWORKS CAN MISLEAD

Terry Williams

62 Cover to Cover—Book Reviews

Kenneth H. Rose, PMP

66 Guidelines for Project Management Journal Book Reviews

67 Notes for Authors

 PROJECT MANAGEMENT JOURNAL
EDITORIAL REVIEW BOARD
John R. Adams, Western Carolina University; Frank
T. Anbari, PMP, The George Washington University;
Bud Baker, Wright State University; Rick Bilbro, The
Innova Group, Inc.; David Christensen, Cedar City,
UT; David Cleland, University of Pittsburgh; Helen S.
Cooke, Cooke & Cooke; Dan H. Cooprider, Creative
People Management; Steven V. DelGrosso, IBM;
Deborah Fisher, University of New Mexico; Vipul
Gupta, Saint Joseph’s University; Francis T.
Hartman, The University of Calgary; Gary C.
Humphreys, Humphreys & Associates; Lewis
Ireland, PMP, Project Technologies Corp.; Peter
Kapsales, Bellcore; Young Hoon Kwak, The
George Washington University; Lee R. Lambert,
PMP, Lambert Consulting Group, Inc.; Alexander
Laufer, Technion-Israel Inst. of Technology; William
Leban, PMP, DeVry University; Robert Loo, The
University of Lethbridge; Kenneth G. Merkel, PMP,
University of Nebraska; Michael D. Okrent, Agilent
Technologies Inc.; Peggy C. Poindexter, Great Falls,
VA; Tzvi Raz, Tel Aviv Unversity; Paul B. Robinson,
The Eastern Management Group; Arnold M. Ruskin,
PMP, Claremont Consulting Group; Richard L.
Sinatra, PMP, Potomac, MD; Dwight Smith-
Daniels, Arizona State University; James Snyder,
Springfield, PA; Paul Solomon, PMP, B-2 Earned
Value Management Systems; Robert G. Staples,
Monroe, VA; Charles J. Teplitz, University of San
Diego; Zeljko M. Torbica, PMP, Florida International
University; Walter A. Wawruck, Vancouver, BC,
Canada; Itzhak Wirth, Saint John’s University; Janet
K. Yates, San Jose State University.
PUBLICATION & MEMBERSHIP
The Project Management Journal (ISSN 8756-9728/03)
is published quarterly (March, June, September, December)
by the Project Management Institute. The Journal is printed
in the USA by Cadmus Magazine, Richmond, VA.
Periodical postage paid at Newtown Square, PA 19073
USA and at additional mailing offices. Canadian
agreement #40030957. Postmaster: Send address
changes to Project Management Institute, Headquarters,
Four Campus Boulevard, Newtown Square, PA 19073-
3299 USA; tel: +1-610-356-4600, fax: +1-610-356-
4647.
The mission of the Journal is to provide information
advancing the state of the art of the knowledge of project
management. The Journal is devoted to both theory and
practice in the field of project management. Authors are
encouraged to submit original manuscripts that are
derived from research-oriented studies as well as
practitioner case studies. All articles in the Journal are the
views of the authors and are not necessarily those of PMI.
Subscription rate for members is $14 U.S. per year and
is included in the annual dues.
“PMI” is a trade and service mark registered in the United States and other nations; “PMP” and the PMP logo are registered certification marks in the United States and other nations;
“PMBOK” is a trademark registered in the United States and other nations; and the PMI logo, “PM Network”, “Project Management Journal”, “PMI Today”, and
“Building professionalism in project management.” are trademarks of the Project Management Institute, Inc.
Editor
Christophe N. Bredillet, PhD, MBA,
Ingénieur EC Lille
PMI PUBLISHING STAFF
Publisher
David Parker; david.parker@pmi.org
Product Editor
Richard Schwartz; richard.schwartz@pmi.org
Contributing Editor
Dan Goldfischer; dan.goldfischer@pmi.org
Project Management Journal Associate
Natasha Pollard; natasha.pollard@pmi.org
Bookstore Administrator
Regina Madonna; regina.madonna@pmi.org
Publications Planner
Barbara Walsh; barbara.walsh@pmi.org
Publications Processor
Dotti Bobst; dotti.bobst@pmi.org
General e-mail:
pmipub@pmi.org
Journal in print:
pmjinprint@pmi.org
Book Review Editor
Kenneth H. Rose, PMP
Claims for undelivered copies must be made no later
than two months following month of publication. The
publisher will supply missing copies when losses have been
sustained in transit and when the reserve stock will permit.
PMI is a nonprofit professional organization whose
mission is to serve the professional interests of its collective
membership by: advancing the state of the art in the
leadership and practice of managing projects and
programs; fostering professionalism in the management
of projects; and advocating acceptance of project
management as a profession and discipline.
Membership in PMI is open to all at an annual dues of
$119 U.S. For information on PMI programs and
membership, or to report change of address or
problems with your subscription, contact:
Project Management Institute, Headquarters,
Four Campus Boulevard, Newtown Square, PA 19073-
3299 USA; tel: +1-610-356-4600; fax: +1-610-356-
4647; e-mail: pmihq@pmi.org
© 2004 Project Management Institute, Inc. All rights reserved.
PMI Regional Service Centre: Europe - Middle
East - Africa (EMEA), Avenue des Gaulois 7, B-1040
Brussels, Belgium; tel: +32-2-743-15-73; fax: +32-2-743-
15-50; e-mail: emea-servicecentre@pmi.org
EDITORIAL & ADVERTISING SERVICES
Address manuscripts and other editorial submissions,
advertising and mailing list rental inquiries, and requests
for reprints/bulk copies/reprint permission to: Project
Management Institute, Publishing Department,
Four Campus Boulevard, Newtown Square, PA 19073-
3299 USA. tel: +1-610-356-4600, fax: +1-610-356-
4647; e-mail: pmipub@pmi.org
Unless otherwise specified, all letters and articles
sent to the PMI Publishing Department are assumed for
publication and become the copyright property of PMI if
published. PMI is not responsible for loss, damage, or
injury to unsolicited manuscripts or other material.
READER SERVICES
Photocopies. Project Management Journal has been
registered with the Copyright Clearance Center, Inc.
Consent is given for copying of articles for personal or
internal use, or for the personal use of specific clients. This
consent is given on the condition that the copier pays
through the Center the per copy fee stated in the code
on the first page of each article for copying beyond that
permitted by Sections 107 or 108 of the U.S. Copyright
Law. The appropriate fee should be forwarded with a
copy of the first page of the article to the Copyright
Clearance Center, Inc., 222 Rosewood Drive, Danvers,
MA 01923 USA (tel: +1-508-750-8400; fax: +1-508-
750-4744). The fee indicated on the first page of an
article in this issue will apply retroactively to all articles
published in the journal, regardless of the year of
publication. This consent does not extend to other kinds
of copying, such as for general distribution, resale,
advertising and promotion purposes, or for creating new
collective works. Special written permission must be
obtained from the publisher for such copying.
Permissions. Requests to reprint articles published
in Project Management Journal must be made in writing
to the publisher.
Reprints. Individual articles may be purchased
through the Knowledge & Wisdom Center
(www.pmi.org/k&wc)
Glossy Reprints. Requests for glossy reprints of
individual articles in quantities of 100 or more can be
sent to pmipub@pmi.org.
Bulk Copies of Current Issues. Copies of the
current Project Management Journal can be obtained in
quantities of 25 or more. Orders must be placed 40
days prior to date of issue. The cost is $5.50 per copy
plus shipping.
Back Issues. Back issues are available on request
at $20 each. Call +1-610-356-4600 for availability.
 F ROM THE E DITOR
Christophe N. Bredillet, PhD, MBA, Ingénieur EC Lille
Dear Colleagues and Friends,
During the recent PMI Research Conference held in
London (11–14 July), I had the chance to talk to
some “practitioners” about the presentations, and
more generally, about the role of Research and how
theories were supporting practices. The question
about the balance between “theoretical” papers and
more “practically” oriented papers was raised, with
some practitioners complaining about “theoretical”
papers being too far from possible applications.
(For the record, the ratio between researchers and
practitioners attending the PMI Research
Conferences was 2:1 in Paris (2000), 1:2 in Seattle
(2002) and 1:1 in London.)
Interestingly, most of the papers presented by
well-known academics in the field were emphasizing
the failure of “classical” project management in
addressing the so-called “new business environ-
ment” and the new challenges with which organiza-
tions have to cope. This situation lead them to call
for the development of new theories, not only for
project management, but for program and portfolio
management, as well.
The day following the PMI Research Conference,
I had the opportunity to attend a meeting organized
at University College London, as part of the UK
EPSRC Network-named “Rethinking Project
Management.” During this meeting, the same assess-
ment led to the same recommendations.
This gives me the opportunity to clarify Journal
editorial policy: Project Management Journal aims to
attract papers that contain some theory rather
than no theory. Having written this, I would like to
clarify the underlying rational and provide some
indication of what is/is not theory.
Let me summarize
my position: sound prac- Research
tice needs supporting the-
ories that result from
Theory Practice
grounded research on
practice and that chal-
lenge these theories,
either explicitly or tacitly.
We have systemic links between practice, theory and
research, and breaking any one of them leads to
impoverishment of the whole system. You can sure-
ly find examples of this in your own environment!
In the words of Poincaré in La science et l’hypothèse
(1903), Paris, Flammarion, “Science is facts, just as
houses are made of stone… But a pile of stones is not nec-
essarily a house, and a collection of facts is not necessar-
ily science.”
How can one attempt to gain and express
knowledge of the word “theory” and of what con-
stitutes a “theoretical contribution”? Rather than
writing a treatise on this subject, I prefer to
encourage readers to refer to some excellent con-
tributions. David A. Whetten, in Academy of
Management Review’s special issue on theory
(1989, 14(4), pp. 490-495) addressed the ques-
tion of what constitutes a theoretical contribution
and provided a list of seven key questions consti-
tuting a summary answer: 1) What’s new? 2) So
what? 3) Why so? 4) Well done? 5) Done well? 6)
Why now? and 7) Who cares? More recently, a
famous debate took place in Administrative Science
Quarterly (ASQ) (1995, 40(3)) between Sutton &
Staw (“What Theory is not?,” pp. 371-384), Weick
(“What Theory is not, Theorizing is”, pp. 385-
390) and DiMaggio (“Comments on “What
Theory is Not”, pp. 391-397).
I support a balanced approach à la Weick: ref-
erences, data, lists of variables or constructs, stories,
diagrams, hypotheses, or predictions may consti-
tute to some extent the premises of theories, even if
not full-blown theories. As Weick stated in 1995,
“Perhaps the ultimate trade-off is the one between
process and product, between theorizing and theo-
ry, between doing it and freezing it” (ASQ, p. 390).
Before concluding this editorial, I cannot resist
sharing with you two more quotations:
“There is nothing so practical as a good theory.”
— Lewin, K. (1936) Principles of topological psy-
chology. New York: McGraw-Hill.
“Theory, when properly fashioned, ultimately
provides more simplicity and clarity than uninte-
grated and scattered information.”
— Millon, T. (1990). Toward a new personology:
An evolutionary model. New York: Wiley-Interscience.
September 2004 Project Management Journal • 3
 T O O UR R EADERS
In order to provide you better, timelier service,
Project Management Journal is now
“delivered” online!

Project Management Journal Upgrade Emphasis on Electronic Delivery

Among the hallmarks of any professional associ- To further enhance the Journal’s academic
ation are advancing a body of knowledge and sup- impact and provide members with more timely
porting the growth and use of new knowledge. Project access to its learning, PMI is emphasizing electronic
Management Journal, PMI’s academic and research distribution as the Journal’s primary delivery
quarterly, has supported those ends as the profession method.
has grown. Now is the appropriate time to pursue Effective the April issue, PMI began posting a
additional gains toward advancing the profession printable version of the Journal on the Members Area
and practice of project management and increasing of the Web site, concurrent with its limited print dis-
PMI’s global eminent influence. tribution. In conjunction with this evolution, PMI has
As part of its effort to promote the worldwide uti- discontinued automatic mailing of the Journal to mem-
lization of project management, PMI bers who do not request a printed copy.
has begun a project to elevate the Note that a request for con- (Institutional and academic sub-
quality and reach of the Journal.
tinued receipt of the Journal scribers, including libraries, will auto-
Electronic delivery is just one part of matically continue to receive their
the improvement plan, which focuses in print need only be made
printed versions.)
on improved content and better one time—not once per Although the emphasis is on
access to the Journal’s learning for issue. PMI members who timely Web access, PMI members
leading academicians and major aca-
received the June issue in have an option to continue receiving
demic libraries. In addition, Journal a printed copy, while still having
editorial staff will be seeking more print are already set to access to both current and archival
academically rigorous submissions, receive subsequent issues issues online. Members who notify
while reaching out to those in the in print, but may elect to opt PMI of their preference for this option
project management field with distin- will continue to receive the Journal in
out in the future.
guished credentials to staff its review print, along with their regularly
board and advisory committees. mailed copies of PM Network® and
Initiatives to reach out to other disciplines that PMI Today™. PM Network and PMI Today will contin-
are strongly aligned with project management will ue to be mailed each month to all members, regardless
attract individuals in those disciplines as potential of their Journal delivery option.
authors of Journal articles and thereby further expand Any current PMI member who wants to contin-
relevant knowledge. Increases in readership, applica- ue receiving the Journal in print (and has not already
tion and citation will result from this project. done so) should e-mail pmjinprint@pmi.org with
PMI’s goal is to have the Journal considered a “tier his or her name, member identification number and
one” quarterly. Tier one status means the publication mailing address. (If the printed option is selected,
is the leading journal in its field. In the words of the please use this e-mail address to ensure optimum
Institute for Scientific Information (ISI), PMI seeks to service.) The deadline to sign up for a printed ver-
make the Journal “implactful,” or having both impact sion of the Journal is 29 October for the December
and implications. 2004 issue.

4 • Project Management Journal September 2004

PERT SCHEDULING WITH RESOURCE CONSTRAINTS
USING QUALITATIVE SIMULATION GRAPHS
ABSTRACT
The Qualitative Simulation Graph
Methodology (QSGM) is well suited to
address the Program Evaluation and
Review Technique (PERT) scheduling-with-
resources problem. In QSGM, all possible
scenarios are characterized in the prob-
lem. This is called the coverage property.
The coverage property of QSGM has two
important implications for the PERT
scheduling problem. First, it means that
all possible schedules are represented.
Second, it means that the model can char-
acterize all possible outcomes of a deci-
sion that needs to be made in the
schedule. This gives rise to the possibility
of robust scheduling decisions without
the need to rerun the schedule because
actual activity durations differ from esti-
mated activity durations.
Keywords: PERT Scheduling; Resource-
Constrained Project Scheduling; Qualitative
Simulation Graphs; Qualitative Discrete-
Event Simulation
©2004 by the Project Management Institute
Vol. 35, No. 3, 5-14, ISSN 8756-9728/03
RICKI G. INGALLS, School of Industrial Engineering and Management
Oklahoma State University, Stillwater, OK 74078
DOUGLAS J. MORRICE, Management Science and Information Systems Department
The University of Texas at Austin, Austin, TX 78712
Introduction
he Qualitative Simulation Graph Methodology (QSGM) that has been
T developed by Ingalls et al. (2000) is a general purpose Qualitative
Discrete-Event Simulation (QDES) framework that can be used for any
type of Discrete-Event Simulation (DES) problem. However, because of the
scenario explosion problem, the application of QSGM is better suited for cer-
tain types of problems. In particular, QSGM is best suited for problems with
a fixed time horizon or problems that are being analyzed at a more strategic
(or aggregate) level. Activity-on-arc PERT networks fit the fixed time horizon
and more strategic class of problems. The QSGM brings a very important
property to the PERT class of problems, and that is the coverage property. The
coverage property means that all possible scenarios are characterized in the
problem. A scenario is defined by any possible sequence of events. A schedule
is a sequence of events with specific timing. This has two important implica-
tions for the scheduling problem. First, it means that all possible schedules
are represented. Second, it means that, as long as the delay time intervals are
not violated, the model can characterize all possible outcomes of a decision
that needs to be made in the schedule.
This paper covers the application of QSGM to PERT networks with
resources. Section 2 gives a short background discussion on QSGM. Section 3
outlines the process for converting a PERT network with resources to an event
graph network. Section 4 gives an example of a PERT network with resources.
In Section 4, an example of PERT networks with deterministic delay times and
another example of the same PERT network with interval delay times are
given. Additionally, two key topics are discussed: the meaning of the output
and a new method for decision-making for scheduling an ongoing project.
Section 5 gives concluding remarks and possible future research.
Background
Schruben (1983) introduced Event Graphs (EGs) as a modeling paradigm for
DES. The basis for an EG model is represented in Figure 1. The vertices labeled
A and B represent events. The edge specifies that there is a relationship
between the two events. More specifically, the construct, called a scheduling
edge, can be interpreted as follows: “whenever event A occurs, if condition (i)
September 2004 Project Management Journal • 5

t (i)
A B
Figure 1: Event graph (Schruben 1983)
is true, then event B will be scheduled
to occur t time units later” [Schruben,
1983]. The quantity t may assume the
value zero, in which case B happens at
the same instant as A. Note that it is
possible (and often necessary) to spec-
ify an edge with no condition.
t (i) k later.” The EG construct in Figure 3 is
A B(n)
Figure 2: Event graph with parameter passing
(Schruben 1995)
It is also possible to parameterize
the event vertices in the EG, and thus
extend the basic construct in Figure 1.
Parameterization is accomplished
through vertex parameters and edge
attributes. A vertex parameter list is a
string of state variables associated with
a particular vertex. An edge attribute
list is a string of expressions associated
with a particular edge. These lists are
used in scheduling or canceling specif-
ic instances of events. Using these
parameters is analogous to passing val-
ues in subroutines in high-level pro-
gramming languages. The construct in
Figure 2 is interpreted as follows: “if
condition (i) is true at the instant
event A occurs, then event B(n) will be
scheduled to occur t time units later
with parameter string n, equal to k.”
(Schruben, 1995).
Ingalls, Morrice, Yücesan, &
Whinston (2003) extended EGs using
simulation graph models by allowing
the canceling of events with the edge
execution condition. The basic construct
of the event graph with the edge execu-
tion condition is given in Figure 3. The
nodes labeled A and B represent
events, and the edge specifies that
there is a relationship between the two
events. The construct is interpreted as
6 • Project Management Journal September 2004
follows: “if condition (i) is true at the
instant event A occurs, then event B
will be scheduled to occur t time units
later. Event B will be executed t time
units later with the state variables in
array n set equal to the values in array
t (i) (j) k
A B(n)
Figure 3: Event graph with parameter passing and
edge execution conditions (Ingalls et al. 2003)
k if condition (j) is true t time units
used in the development of the QSGM.
Ingalls, Morrice, & Whinston
(2000) developed the QSGM. QSGM
is a specific implementation of QDES
that uses interval values for the simula-
tion time, activity delays, state vari-
ables, and the graph theoretic for EGs.
This extension of traditional DES
allows the modeler to model uncer-
tainty in the timing of events and the
value of state variables without assum-
ing the timing of events is random
according to some prespecified proba-
bility distribution. As a result, QDES is
a robust form of DES because it
requires fewer, possibly erroneous
modeling assumptions. Also, the out-
put of the simulation characterizes all
possible outcomes.
When two or more event execu-
tion intervals intersect during the exe-
cution of a QDES simulation, the
event execution order becomes uncer-
tain and multiple scenarios result.
QDES simulates and generates all pos-
sible scenarios and the corresponding
uncertainty in their event timings. In
this sense, QDES characterizes all pos-
sible outcomes of the simulation. The
characterization of all possible out-
comes is called coverage. Coverage is
an important advantage of the QDES
approach because it does not miss
outcomes that a sampling-based
approach such as DES might with a
finite sample size.
The coverage property implies that
all possible scenarios of execution are
characterized in the problem. This has
two important implications for the
scheduling problem. First, it means
that all possible schedules are repre-
sented. Second, it means that, as long
as the delay time intervals are not vio-
lated, the model can characterize all
possible outcomes of a decision that
needs to be made in the schedule.
CONVERTING A PERT NETWORK TO A
SIMULATION GRAPH
PERT Without Resources
Schruben (1995) outlines how to con-
vert an activity-on-arc PERT network
without resources to an EG. The logic
behind the conversion is as follows:
1. Each node of the PERT network cor-
relates directly to an equivalent
node in the EG.
2. Each arc in the PERT network corre-
lates directly to an equivalent arc in
the EG with the same delay.
3. Each node in the EG has code to
count the number of times that the
node has been hit.
4. The scheduling condition for the EG
arc evaluates TRUE if the number of
hits for the node at the head of the
arc is equal to the number of incom-
ing arcs to that node.
To show how a simple PERT net-
work would be translated, consider the
PERT network conversion in Figure 4.
This 5-node network is converted by
counting the hits on nodes 2, 3, and 4
using the state variables H2, H3, and
H4, respectively. Each time that the
nodes (events) are executed, the
respective hit variable is incremented.
When the scheduling condition is met,
then the activity is scheduled with the
appropriate delay. For example, when
the 2 node is hit once (i.e., H2 == 1),
then the scheduling condition of
H2==1 is satisfied and the activity 2-4
is scheduled with delay time d24. This
counting of hits and scheduling activi-
ties continues until there are no more
events scheduled on the calendar.
PERT With Resources Using EGs
PERT networks with resources are
much more complicated to model. In
a PERT network with resources, not
only do all of the previous activities
need to be complete, but all of the nec-
essary resources must be available as
 4. Each hit node is connected to its cor-
responding start nodes with a sched-
2
uling condition and an execution
d2 condition. The scheduling condition
2 4
d1 checks to see if the hit node has been
hit the proper number of times (i.e.
d45 all predecessor activities have been
1 4 5 completed). The execution condition
d1 checks to see if the resources are
3 4 available.
d3
5. Each finish node has arcs pointing
back to start nodes of activities that
3 could need released resources. Each
one of these arcs has a scheduling
condition to determine if the hit
node has been triggered and an exe-
2 d2 cution condition to determine if the
4
necessary resources are available.
H2
H2=H2+1 == As can be seen in the conversion
1 of the PERT with resources network in
2
d1 Figure 5 to the corresponding EG in
d45 H4==2
H2==1

d1 Figure 6, the conversion can be very

1 3 4 5
complicated in order to handle com-
2 peting resources that are available for a
3 == H4=H4+1
H given task.
4 In the example in Figure 5, a PERT
d3
network with 5 activities and two
3 resources is shown. With each activity,
the quantity of each type of resource
H3=H3+1 that is needed is known. In Figure 6,
the translation of the PERT network in
Figure 4: Converting a PERT network without resources Figure 5 to an EG is shown. In Figure 6,
one can see the pattern described
well, in order to start the new activity. activity. For our purposes, the start above. First, the 0 node is created so
In order to convert a PERT network node is labeled Sx.y and the finish that the two resources can be initial-
with resources to an EG, the following node is labeled Fx.y, where x is the ized to their capacities. Second, each
steps must be followed: number of the head node and y is node in the PERT network in Figure 5
1. The EG network adds an initializa- the number of the finish node for has a corresponding hit node. When
tion node, called node 0 for the pur- the given arc in the PERT network. each of the incoming arcs come into
pose of initializing the available
number of resources in the network.
2.For each node in the PERT network,
create a corresponding node in the 2
EG network. This node will be the hit
d2
2

node and count the number of previ-

d1

4
(1
)
,1

ous completed activities that are nec-

,2
(1

essary to execute the activities

d45
associated with the corresponding 1 4 5
PERT node. For our purposes, the (1, 1)
nodes will be labeled Hx, where x is
d1

4
d3
3

the number of the PERT node. LEGEND

(1

)
,0

,1

3. For each arc in the PERT network, delay time

)

(0

(resource 1 needed,
there are two more nodes that are 3 resource 2 needed)
created in the EG. The first is the
start event that signifies the start of
the activity and the finish event that
signifies the completion of the Figure 5: PERT with resources example

September 2004 Project Management Journal • 7

 ignored and taken off of the calendar.
& d12 At that point, the S1.3 event could only
> =1 =1 S1.2 F1.2 be executed when another activity
S1 S2>

Q1
completes and frees up resource 1. In
S1=S1-1 S1=S1+1

2=
=1 this particular case, the F1.2 or F2.4

=1
S2=S2-1 S2=S2+1
2= events would need to be executed.
Q1 Q12=0
If any of the start events are exe-

S1 2>=
0 H1 Q1

1
cuted, they schedule the activity delay

>=

>= 1
S
3=

S1
=1 in the PERT network. In a DES, the

1&
S1=1 H1=H1+1 activity delay will either be a deter-
S1

=1
S2=2 Q12=(H1>=1) >=

3=
Q13=(H1>=1) 1 ministic value or a sample from a ran-

Q1
d13 dom distribution. When the
S1.3 F1.3 corresponding finish event is executed,
Q1
3= two things occur. First, the succeeding
S1=S1-1 =1 S1=S1-1 hit events are scheduled to occur
Q13=0 S1
>= instantaneously. Second, the finish
1
event schedules any start events that
& d24 have not occurred and need some of
> =1 =2 S2.4 F2.4 the resources that have just been
S1 S2>
Q2

released by the finish event.

4=

S1=S1-1 S1=S1+1 Continuing with the scenario

=1
=1

= S2=S2-1 S2=S2+1
Q 24 Q24=0
described above where the event S1.2
is executed first, F1.2 is executed when
S1 2>=
Q24

H2 Q2 the activity is complete. At that time,

>= 2
S

3=
==1

1&

=1 the resource availability variables

H2=H2+1 would be set to S1 = 1 and S2 = 2.
S1> >=2
S2

Q24=(H1>=1) Because F1.2 has just released the

=1

Q23=(H1>=1)
resources that activity 1.3 needs to exe-
&

0
S2.3 F2.3 cute and the variable Q13 is set to 1,
the event S1.3 is scheduled to execute.
Q23=0 The execution condition, S1 > 1,
would be true and S1.3 would execute
and begin activity 1.3.
Q24==1 S2>=1 d34 The DES simulation continues
H3 H3.4 F3.4 until all of the activities have been
completed. The output of one run of
H3=H3+1 S2=S2-1 S2=S2=1 the DES would be a randomly generat-
Q34=(H2>=2) Q34=0 ed schedule for the PERT problem.

S2>=1 & PERT With Resources Using QSGM

Q45==1 S2>=1
H4
H4=H4+1
Q45=(H4>=2)
Figure 6: PERT with resources converted to EG
the hit node, then the scheduling con-
ditions on the outgoing arcs become
true. In the case of node H1, the two
outgoing arcs both occur in zero time
and one of the arcs is not prioritized
over the other. In a DES framework,
the choice of which activity would start
first would be implementation-specif-
d45 The conversion of the PERT network
S4.5 F4.5 with resources to a QSGM model is
identical to the conversion to an EG
S1=S1-1 S1=S1+1 model. The difference lies in how the
S2=S2-1 S2=S2+1
Q45=0 activity delay times are handled. In the
QSGM, the delay times would be spec-
ified as time intervals. The QSGM
would then use the interval informa-
tion to generate all possible scenarios
ic. For this example, let us assume that in the problem. Also, all of the possi-
the event S1.2 would be executed first. ble resource deployments would be
In this scenario, the resource avail- represented. When the model is run as
ability variables would be set to S1 = 0 a QSGM, there are some differences
and S2 = 1. When the S1.3 event comes that occur in the execution. For the
off of the future events calendar, the QSGM, when node H1 is executed, the
execution condition, S1 > 1, would be two outgoing arcs both occur in zero
false and the S1.3 event would be time and one of the arcs is not priori-

8 • Project Management Journal September 2004

tized over the other. The execution PERT With Resources Examples put of the simulation would be all of
order of these two pending events on Using QDES the possible sequences of events and
the future events calendar would be To illustrate the QSGM, we use the their timing. The model certainly
uncertain (either event could be exe- “repairman model” from Pritsker, would be able to find the minimum
cuted first), so two scenarios would be Sigal and Hammesfahr (1989). Both makespan time using this type of
created. One scenario would assume the deterministic delay time version of approach. Although the QSGM can be
that the event S1.2 would be executed the model and the interval delay time used in this way, it is not an efficient
first and the second would assume that version of the model will be devel- search algorithm. Algorithms built to
the event S1.3 would be executed first. oped. The repairman model has 2 find optimal schedules would be
In the scenario where S1.2 is exe- repairmen available to perform the more efficient than using QSGM.
cuted first, the resource availability following tasks: However, the model did find that that
variables would be set to S1 = 0 and S2
= 1. When the S1.3 event comes off of
the future events calendar, the execu- Activity Nodes Delay Time Resources
(Min,Expected,Max) Needed
tion condition, S1 > 1, would be false
and the S1.3 event would be ignored 1. Disassemble power units 1,2 (1,3,5) 2
and taken off of the calendar. At that and instrumentation
point, the S1.3 event could only be 2. Install new assembly 1,3 (3,6,9) 2
executed when another activity com-
pletes and frees up resource 1. In this 3. Prepare for repair check 1,4 (10,13,19) 1
particular case, the F1.2 or F2.4 events 4. Clean, inspect, and repair 2,5 (3,9,12) 1
would need to be executed. power units.
If any of the start events are exe-
5. Calibrate instrumentation 2,3 (1,3,8) 1
cuted, then they automatically sched-
ule the activity delay in the PERT 6. Check interfaces 3,6 (8,9,16) 1
network. Unlike the activity delay in
the DES example, which will either be 7. Check assembly 3,4 (4,7,13) 1
a deterministic value or a sample from 8. Assemble and test power units 5,6 (3,6,9) 1
a random distribution, the activity
delay is an interval. This interval repre- 9. Repair check 4,6 (1,3,8) 1
sents the possible delays that can occur
Figure 7: Data for the “repairman” model (Pritsker et al. 1989)
with this activity. When the correspon-
ding finish event is executed, two
things occur. First, the succeeding hit QDES Example Using Deterministic minimum makespan is 34 and Figure
events are scheduled to occur instanta- Delay Times 9 shows two schedules that meet that
neously. Second, the finish event sched- In the deterministic delay time version minimum makespan. The output of
ules any start events that have not of this problem, the model used the the model showed that 112 of the 360
occurred and need some of the expected delay time in Figure 7 for the scenarios had a completion time of
resources that have just been released activity delays. Figure 8 shows the 34 and qualified for minimum
by the finish event. PERT diagram for this model. The out- makespan.
Continuing with the scenario
described above where the event S1.2 is
executed first, F1.2 is executed when the t=9, r=1
2 5
activity is complete. At that time, the
2

resource availability variables would be

r=

r=1

r=1
t=3

t=6

set to S1 = 1 and S2 = 2. Because F1.2

3,
t=

has just released the resources that activ-

ity 1.3 needs to execute and the variable t=6, r=2 t=9, r=1 LEGEND
Q13 is set to 1, the event S1.3 is sched- 1 3 6 t = delay time
r = resources needed
uled to execute. The execution condi-
t=

1
r=1
t=7
13

r=

tion, S1 >≥ 1, would be true and S1.3

,

3,
r=

would execute and begin activity 1.3.

t=
1

The qualitative simulation contin-

ues to generate all possible scenarios (or 4
event sequences) until they are enumer-
ated. Each of these scenarios has time
windows in which they could execute. Figure 8: PERT diagram for the “repairman” model

September 2004 Project Management Journal • 9

 the simulation will be examined.
1.2 Second, it will be shown how the out-
1.3 put can be used in the scheduling
process where the scheduler is making
1.4
sequencing decisions without needing
2.3 to rerun the simulation.
2.5
3.4 PERT Model Output
As an example of the output generated
3.6
by the simulation on a given PERT
4.6 problem, let us look at a scenario gen-
5.6 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 erated for the Repairman problem in
TIME Figure 10. The output for one scenario
is shown in Figure 11. The four
columns are the Event #, the Event, the
1.2 Time in which that event could occur,
1.3 and the Input Delay for each start event.
1.4 If the Event has an S designation, then
2.3 it is the start of a given activity. If the
Event has an F designation, then it is
2.5 the finish or completion of that activi-
3.4 ty. The Input Delay means that the
3.6 delay for that activity lies somewhere
4.6 in the Input Delay interval.
What information can be gleaned
5.6 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36
from this scenario? First, if the start
TIME
and finish events happen in this
sequence, the schedule will finish
Figure 9: Two minimum makespan schedules somewhere between time 26 and time
69. Also, the time window for this and
Example Using Interval Delay Times assure that the schedule is still valid is any other sequence that has a common
Now let us take the same to rerun the schedule with the up-to- set of beginning events is known. For
Repairman Model and apply qualita- date information. Second, if the actual example, this scenario’s first four
tive time constructs and use those delay time changes so much as to events are (S1.2, F1.2, S1.3, F1.3).
qualitative time constructs as a mecha- invalidate the sequence of the optimal When the F1.3 event occurs, the time
nism for scheduling all these activities schedule, then the scheduler is forced window for that event is (4,14). That
“real-time”. Figure 10 shows the PERT to reschedule for the remaining activi- time window will be true for any other
representation of this model. Notice ties. Taking advantage of the coverage scenario that shares this same event
that the delay times are now intervals. property of the QSGM, no reschedul- sequence for the first four events. This
First of all, the QSGM gives a ing is necessary. First, the raw output of leads to another observation. When
new framework for scheduling. In
the example, there are two resources
that must be allocated to competing t=(3,12), r=1
2 5
activities. In a traditional scheduling 2
algorithm, the delay times would be r=
t=(1,8)

t=(3,9)

,
5)
r=1

r=1

deterministic and the sequence of ,

(3
activities would be an optimal t=
answer based on those deterministic t=(3,9), r=2 t=(8,16), r=1 LEGEND
delay times. 1 3 6 t = delay time
t= r = resources needed
A weakness of optimal scheduling (1 1
0, r=
t=(4,13)

19
r=1

algorithms is that the actual delay ), ),

,8
times often vary from the inputs. This r= (3
1 t=
causes a couple of problems. First, the
scheduler is not sure if the change in 4
the timing of the activity would affect
his or her overall optimal schedule in
any significant way. The only way to Figure 10: Repairman model with interval activity delay times

10 • Project Management Journal September 2004

 Event Event Time Input order to place the project in the best decide which activity (or activities)
# Delay position to succeed. would be allocated the constrained
A scenario is made up of the resources. In the QSGM output, there
1 S1.2 [0.0,0] (1,5)
sequence of start and finish events for will be scenarios that simulate the out-
2 F1.2 (1.0,5.0) each of the activities in the PERT dia- come of each of the possible decisions
3 S1.3 (1.0,5.0) (3,9) gram. Some of these events would that the scheduler could make.
4 F1.3 (4.0,14.0) occur based on a decision by the project The sequence of other events
scheduler. A decision would occur occur in an uncertain order, which is to
5 S2.3 (4.0,14.0) (1,8)
when there are multiple activities that say that one event could occur before
6 S2.5 (4.0,14.0) (3,12) could start their process, but there is a another based on the actual timing of
7 F2.3 (5.0,22.0) lack of available resources to start all of the delay. These uncertain events are
the activities. The scheduler would outside the control of the scheduler.
8 S3.4 (5.0,22.0) (4,13)
9 F2.5 (7.0,26.0) Alternative Decision Uncertain Fixed Complete Scenario
10 S3.6 (7.0,26.0) (8,16) Event
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Max Min Time
11 F3.4 (9.0,35.0)
S1.2 69 19
12 S5.6 (9.0,35.0) (3,9) S1.3 75 19
S1.4 75 24
13 F3.6 (15.0,42.0)
S1.2 F1.2 S1.3 69 19
14 S1.4 (15.0,42.0) (10,19) S1.2 F1.2 S1.4 64 22
S1.2 F1.2 S2.3 64 22
15 F5.6 (15.0,44.0) S1.2 F1.2 S2.5 63 22
16 F1.4 (25.0,61.0) S1.2 F1.2 S2.5 S1.4 63 22
S1.2 F1.2 S2.5 S2.3 62 23
17 S4.6 (25.0,61.0) (1,8) S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 56 23
18 F4.6 (26.0,69.0) S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S5.6 56 26
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 51 23
Figure 11: PERT scenario output example S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.6 56 23
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S5.6 56 23
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 51 23
choosing which activity gets a free
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S5.6 51 26
resource, such as happens at event 5, S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 51 23
the user can look at the scenarios with S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 F3.4 S5.6 50 23
a common starting sequence. In this S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 F3.4 S5.6 F3.6 S4.6 F4.6 F5.6 48 24 24,48
case, there are three scenarios with S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 F3.4 S5.6 F3.6 S4.6 F5.6 F4.6 50 24 24,50
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 F3.4 S5.6 F5.6 S4.6 F3.6 F4.6 50 23 23,50
common starting sequences, (S1.2,
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 F3.4 S5.6 F5.6 S4.6 F4.6 F3.6 42 23 23,42
F1.2, S1.3, F1.3, S2.3), (S1.2, F1.2, F1.2 F2.3 F1.4 S1.3 F3.4 F4.6 48 26 26,48
S1.2 S2.5 S2.3 S1.4 F2.5 F1.3 S3.4 S3.6 F3.6 S5.6 S4.6 F5.6
S1.3, F1.3, S1.4), and (S1.2, F1.2, S1.3, S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 F3.6 S5.6 F3.4 S4.6 F5.6 F4.6 47 26 26,47
F1.3, S2.5). Activities (2,3), (1,4), and S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F1.4 F2.5 S1.3 F1.3 S3.4 S3.6 F3.6 S5.6 F5.6 F3.4 S4.6 F4.6 47 27 27,47
(2,5) are all eligible to be started at this S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F1.4 F5.6 S1.3 F1.3 S3.4 56 23
time. Using the information already S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F1.4 F5.6 S1.3 F1.3 S3.6 56 23
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F1.4 F5.6 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 F3.6 F4.6 56 23 23,56
generated by the simulation, decisions
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F1.4 F5.6 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 F4.6 F3.6 51 23 23,51
can be made based on the future of S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F1.4 F5.6 S1.3 F1.3 S3.4 S3.6 F3.6 F3.4 S4.6 F4.6 56 24 24,56
any one of these three alternatives. S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F5.6 F1.4 S1.3 F1.3 S3.4 62 23
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F5.6 F1.4 S1.3 F1.3 S3.6 62 23
Scheduling Decisions With S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F5.6 F1.4 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 F3.6 F4.6 62 23 23,62
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F5.6 F1.4 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 F4.6 F3.6 57 23 23,57
More Information
S1.2 F1.2 S2.5 S2.3 F2.3 S1.4 F2.5 S5.6 F5.6 F1.4 S1.3 F1.3 S3.4 S3.6 F3.6 F3.4 S4.6 F4.6 62 24 24,62
In reality, project scheduling deci- S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 62 25
sions are often made based the cur- S1.2 F1.2 S2.5 S2.3 F2.5 S5.6 62 25
rent status of the schedule, the S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F1.4 F5.6 S1.3 F1.3 S3.4 52 25
available resources, and the instinct of S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F1.4 F5.6 S1.3 F1.3 S3.6 52 25
S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F1.4 F5.6 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 F3.6 F4.6 52 25 25,52
the project manager on how the rest
S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F1.4 F5.6 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 F4.6 F3.6 47 25 25,47
of the project will be executed. Using S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F1.4 F5.6 S1.3 F1.3 S3.4 S3.6 F3.6 F3.4 S4.6 F4.6 52 26 26,52
the qualitative information in the S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F5.6 F1.4 S1.3 F1.3 S3.4 62 25
output generated by the simulation, S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F5.6 F1.4 S1.3 F1.3 S3.6 62 25
the project scheduler has more infor- S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F5.6 F1.4 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 F3.6 F4.6 62 25 25,62
mation about possible future scenar- S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F5.6 F1.4 S1.3 F1.3 S3.4 S3.6 F3.4 S4.6 F4.6 F3.6 57 25 25,57
S1.2 F1.2 S2.5 S2.3 F2.5 S1.4 F2.3 S5.6 F5.6 F1.4 S1.3 F1.3 S3.4 S3.6 F3.6 F3.4 S4.6 F4.6 62 26 26,62
ios and can make a more robust
decision on which activity to start in Figure 12: Qualitative repairman output using smallest upper bound completion time criterion

September 2004 Project Management Journal • 11

Any finish events could fall into this Alternative Decision Uncertain Fixed Complete Scenario
category of events. Still, other events in Event
the sequence are fixed, meaning that in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 AM Time
the scenario under observation, there S1.2 37.54
S1.3 37.56
is no alternative for that particular
S1.4 45.32
event in the sequence. A simple exam- S1.2 F1.2 S1.3 37.09
ple of a fixed event would be when S1.2 F1.2 S1.4 39.57
only one activity is remaining and S1.2 F1.2 S2.3 39.68
there are available resources. The start S1.2 F1.2 S2.5 38.30
S1.2 F1.2 S1.3 F1.3 S1.4 37.12
event for the last activity is the only
S1.2 F1.2 S1.3 F1.3 S2.3 37.11
event that could occur at that time. S1.2 F1.2 F1.3 S2.5 37.05
S1.3
Any scenario is the sequence of the S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 37.00
decision, uncertain, and fixed events. S1.2 F1.2 S1.3 F1.3 S2.5 S2.3 37.06
With the output of one model run, S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 35.41
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.6 35.98
all possible scenarios are represented
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F2.5 S3.6 36.09
and that information can be used to S3.4 36.18
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 F2.5 S5.6
make decisions that take into account S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F2.5 S3.6 F3.4 S4.6 36.13
all the uncertainty that could happen in S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F2.5 S3.6 F3.4 S5.6 35.50
the future. For example, our first sched- S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F2.5 S3.6 F3.4 S5.6 F3.6 S4.6 F4.6 F5.6 36.00 24,48
uling decision is whether to start activi- S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F2.5 S3.6 F3.4 S5.6 F3.6 S4.6 F5.6 F4.6 37.00 24,50
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F2.5 S3.6 F3.6 S5.6 F3.4 S4.6 F4.6 F5.6 37.00 26,48
ty 1-2, activity 1-3, or activity 1-4. Based
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F2.5 S3.6 F3.6 S5.6 F3.4 S4.6 F5.6 F4.6 36.50 26,47
on a user-defined decision criteria, it is S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F2.5 S3.6 F3.6 S5.6 F5.6 F3.4 S4.6 F4.6 37.00 27,47
determined that activity 1-2 is the best S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S3.6 34.31
activity to start at this time based on the S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S4.6 34.70
possible outcomes in the future. The S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S4.6 F2.5 S3.6 34.50
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S4.6 F2.5 S5.6 35.63
scheduler would have no need to make
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S4.6 F2.5 S3.6 F3.6 S5.6 F4.6 F5.6 36.50 30,43
any other decisions on the sequencing S2.3 F2.3 S3.4 S4.6 F2.5 S3.6 F3.6 S5.6 F5.6 F4.6 32.00 30,34
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 F3.4
of other activities at this point in time. S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S4.6 F2.5 S3.6 F4.6 S5.6 F3.6 F5.6 35.00 27,43
As examples of user-defined deci- S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S4.6 F2.5 S3.6 F4.6 S5.6 F5.6 F3.6 34.50 27,42
sion criteria, two different decision cri- S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S4.6 F4.6 S3.6 F2.5 S5.6 F3.6 F5.6 31.50 28,35
teria are shown. The first is the smallest S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.3 S3.4 F3.4 S4.6 F4.6 S3.6 F2.5 S5.6 F5.6 F3.6 35.00 28,42
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 36.77
upper-bound completion time criterion.
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.6 37.50
At any decision point, the scheduler S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 F3.4 S3.6 35.25
would choose the event that guaran- S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 F3.4 S4.6 36.25
tees the smallest upper bound on the S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 F3.4 S3.6 F3.6 S4.6 F4.6 F5.6 31.50 28,35
completion time. The other scheduling S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 F3.4 S3.6 F3.6 S4.6 F5.6 F4.6 35.50 28,43
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 F3.4 S3.6 F5.6 S4.6 F3.6 F4.6 35.00 27,43
criterion is the average midpoint com-
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 F3.4 S3.6 F5.6 S4.6 F4.6 F3.6 39.00 27,51
pletion time criterion. For all of the S1.4 F1.4 S2.3 F2.5 S5.6 S3.4 F5.6 S3.6 F3.4 S4.6 F3.6 F4.6 40.00 25,55
S1.2 F1.2 S1.3 F1.3 S2.5 F2.3
scenarios that follow any given deci- S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 F5.6 S3.6 F3.4 S4.6 F4.6 F3.6 38.00 25,51
sion, calculate the average midpoint of S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F2.3 S3.4 F5.6 S3.6 F3.6 F3.4 S4.6 F4.6 40.50 26,55
all of those scenarios. The decision
Figure 13 (part one): Qualitative repairman output using average midpoint criterion
that yields the smallest average mid-
point is the best decision.
In our example, the scheduler any other decisions or unexpected smallest upper-bound completion
would have to make a decision if S1.2, event sequences, the worst completion time and the average midpoint com-
S1.3, or S1.4 is the first event in our time will be 69. pletion time, respectively, are shown.
schedule. The average midpoint scores It should be noted that the criteria There are 5 designations in the scenar-
for these three alternatives are 37.54, that have been chosen for generating ios. First are alternatives. Alternatives
37.56, and 45.32, respectively. So the the schedules are to demonstrate the require a decision as to which activity
scheduler would choose S1.2 to be the concept. They may or may not be the starts first. Alternatives are marked
first event in the sequence using the best way to make the scheduling deci- with black background and white type.
average midpoint criteria. The smallest sions that are necessary. “Optimal” After a decision is made, that decision
upper-bound completion time scores for decision criteria are a topic for further is designated in bold. Uncertain events
these three alternatives are 69, 75, and research. are italic. Fixed events are underlined.
75, respectively. Again, S1.2 would be In Figures 12 and 13, the scenarios A complete scenario, based on the
chosen as the first event in the sched- generated by the QSGM methodology decisions made, the uncertain and
ule. This guarantees that, regardless of on the Repairman problem using the fixed events, has a gray background.

12 • Project Management Journal September 2004

 Alternative Decision Uncertain Fixed Complete Scenario one can see that the smallest upper-
Event bound criterion does in fact choose a
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 AM Time schedule that guarantees the smallest
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F5.6 F2.3 S3.4 39.33 upper bound on the completion time
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F5.6 F2.3 S3.6 39.33 at each decision point. Also, the aver-
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F5.6 F2.3 S3.4 S3.6 F3.4 S4.6 F3.6 F4.6 40.00 25,55
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F5.6 F2.3 S3.6 F3.4 S4.6 F4.6 F3.6
age midpoint criterion chooses the deci-
S3.4 37.50 25,50
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S2.3 F2.5 S5.6 F5.6 F2.3 S3.4 S3.6 F3.6 F3.4 S4.6 F4.6 40.50 26,55
sion that has the lowest average
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 37.26 midpoint at each decision point. How
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S5.6 37.96 do these compare to each other?
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 39.27 Smallest upper bound, in Figure 12, has
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.6 39.50
F1.2 F2.5 S2.3 S3.6
an average maximum of 50.1 for the
S1.2 S1.3 F1.3 S2.5 S1.4 F1.4 S5.6 F2.3 S3.4 F3.4 38.25
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 F3.4 S4.6 39.25
completion time, and an average min-
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 F3.4 S3.6 F3.6 S4.6 F4.6 F5.6 34.50 27,42 imum of the completion time of 24.5.
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 F3.4 S3.6 F3.6 S4.6 F5.6 F4.6 38.50 27,50 Also, the latest possible completion
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 F3.4 S3.6 F5.6 S4.6 F3.6 F4.6 38.00 26,50 time is 62. Average midpoint, in Figure
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 F3.4 S3.6 F5.6 S4.6 F4.6 F3.6 42.00 26,58
13, has an average maximum of 48.9
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 F5.6 S3.6 F3.4 S4.6 F3.6 F4.6 40.00 25,55
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 F5.6 S3.6 F3.4 S4.6 F4.6 F3.6 41.50 25,58
for the completion time and an aver-
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F2.3 S3.4 F5.6 S3.6 F3.6 F3.4 S4.6 F4.6 40.50 26,55 age minimum of the completion time
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F5.6 F2.3 S3.4 39.33 of 24.9. The latest possible completion
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F5.6 F2.3 S3.6 39.33 time for this schedule is 58. It would
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F5.6 F2.3 S3.4 S3.6 F3.4 S4.6 F3.6 F4.6 40.00 25,55 seem from this output that the desired
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F5.6 F2.3 S3.4 S3.6 F3.4 S4.6 F4.6 F3.6 37.50 25,50
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F1.4 S5.6 F5.6 F2.3 S3.4 S3.6 F3.6 F3.4 S4.6 F4.6
effect of the smallest upper-bound com-
40.50 26,55
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.4 36.48 pletion time criteria was actually better
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 36.39 achieved using the average midpoint cri-
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S5.6 37.64 teria. A criterion that judges an
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 37.73 extreme, such as the smallest upper-
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S5.6 37.75
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 S4.6
bound completion time, takes into
F3.4 38.13
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 F3.4 S5.6 38.00 account only one scenario and ignores
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 F3.4 S5.6 F3.6 S4.6 F4.6 F5.6 38.00 21,55 all other relevant scenarios. However,
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 F3.4 S5.6 F3.6 S4.6 F5.6 F4.6 39.00 21,57 the average midpoint criterion takes into
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 F3.4 S5.6 F5.6 S4.6 F3.6 F4.6 39.50 22,57 account all of the scenarios effected in
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 F3.4 S5.6 F5.6 S4.6 F4.6 F3.6 35.50 22,49
F1.2 S1.3 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S5.6
some way (in this case by including
S1.2 F1.3 S2.5 S3.4 F3.6 F3.4 S4.6 F4.6 F5.6 37.00 19,55
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 F3.6 S5.6 F3.4 S4.6 F5.6 F4.6 36.50 19,54
them in the average). It would seem
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F1.4 S3.4 F3.6 S5.6 F5.6 F3.4 S4.6 F4.6 37.00 20,54 that using more information might be
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F3.6 S3.4 33.58 better in determining the decisions
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F3.6 S5.6 35.20 needs to make a good schedule. Again,
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F3.6 S3.4 F1.4 S5.6 F3.4 S4.6 F4.6 F5.6 31.50 21,42
the authors stress the heuristic nature
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F3.6 S3.4 F1.4 S5.6 F3.4 S4.6 F5.6 F4.6 35.50 21,50
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F3.6 S3.4 F1.4 S5.6 F5.6 F3.4 S4.6 F4.6 37.50 21,54
of the two scheduling criteria that have
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F3.6 S3.4 F3.4 S5.6 F1.4 S4.6 F4.6 F5.6 32.50 23,42 been chosen. More research should be
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F3.6 S3.4 F3.4 S5.6 F1.4 S4.6 F5.6 F4.6 32.00 23,41 done in order to determine what
S1.2 F1.2 S1.3 F1.3 S2.5 S1.4 F2.5 S2.3 F2.3 S3.6 F3.6 S3.4 F3.4 S5.6 F5.6 F1.4 S4.6 F4.6 32.50 24,41 scheduling criterion is best.
Figure 13 (part two): Qualitative repairman output using average midpoint criterion
Conclusion
The PERT-with-resources problem is a
The scoring used and the final interval Now let us look at how this might general scheduling problem that has
of a scenario are in the far right-hand be used. Because the output data is the ability to be converted to a simula-
columns of the tables. already generated, the decision-mak- tion graph without cycles. Taking
What can be said about these sce- ing process of the scheduler is now a advantage of the coverage property, the
narios? Certainly, each is feasible. Each database query instead of a running of QSGM characterizes all possible sce-
scenario characterizes any schedule that the algorithm over again. The decision narios in its output. This information
has the same order of events. Each can can be much higher, since it is taking can be used to monitor and control a
be followed in order to complete the into account the range of things that project schedule on an ongoing basis.
tasks of the project. All are robust, could happen to impact the schedule It can also be used to make good
because as long as no activity times fall in the future. Without imposing ran- scheduling decisions based on future
outside the activity time intervals desig- dom assumptions on the intervals, a event sequences, regardless of how ran-
nated, all of the possible scenarios have robust schedule can be created. domness might affect the event
been taken into account in the decision. In looking at these two outputs, sequence. This is a significant change

September 2004 Project Management Journal • 13

in the way that project scheduling can
be done and should be a topic of
future research. It can also give the
project manager insights into how to
improve the schedule by forcing cer-
tain event sequences. This assumes
that the project manager could con-
trol the actual activity delay times that
accomplish the more aggressive
schedule.
Using the qualitative simulation
in the area of PERT scheduling with
resources is clearly a new way to
approach this problem. It breaks the
assumption of deterministic or even
probabilistic activity delay times and
creates a database of schedules that
take into account uncertainty. Since
this information is valuable to the
project manager or scheduler, PERT
scheduling is a particular application
area that seems rich for future research
using the QDES methodology.
Future Research
In the area of PERT scheduling with
resources using Qualitative Simulation
Graphs, there is much research that
can be done in scaling this technology
to industrial-sized problems. This scal-
ing would benefit from the research in
making the QSGM more efficient. An
obvious approach to this issue is paral-
lel processing the simulation. To run
this algorithm on a parallel processor
would be very straightforward.
However, there is also research that can
be done in problem representation,
database storage of simulation output,
improved scheduling criteria, and
packaging this technology to be usable
in a business setting.
Another area of research involves
the monitoring the results of the
schedule and then making adjust-
ments to the simulation output.
Referencing an earlier example, if the
scheduler knew that events (S1.2, F1.2,
S1.3, F1.3) had occurred at time 10, he
would know that the event timing of
(4,14) could be replaced with an event
timing of [10,10]. The future research
would involve understanding the
implications of this information. The
authors know that two things would
result from the work. First, the new
information would adjust the upper
and/or lower bounds of the remaining
events in all of the scenarios that have
this starting sequence. Second, the new
information would invalidate some
other scenarios with the same starting
sequence. There would be obvious
impact to the decision criteria and the
management of the information that
had been generated by the simulation.
References
Ingalls, R. G., Morrice, D. J., &
Whinston, A. B. (2000). The
Implementation of Temporal Intervals
in Qualitative Simulation Graphs, ACM
Transactions on Modeling and Computer
Simulation, 10(3), 215–240.
Ingalls, R. G., Morrice, D. J.,
Yücesan, E., & Whinston, A. B. (2003).
Execution Conditions: a Formalization
of Event Cancellation in Simulation
Graphs. INFORMS Journal on
Computing, 15(3).
Pritsker, A.A.B., Sigal, C.E.,
Hammesfahr, R.D.J. (1989). SLAM II:
Network Models For Decision Support
(p.192). Englewood Cliffs, NJ:
Prentice-Hall.
Schruben, L.W. (1983).
Simulation Modeling with Event
Graphs. Communications of the ACM,
26(11), 957–963.
Schruben, L. W. (1995). Graphical
Simulation Modeling and Analysis Using
SIGMA for Windows. Danvers, MA:
Boyd & Fraser Publishing Company.

RICKI G. INGALLS is Associate Professor and Director of the Center for Engineering Logistics and Distribution
(CELDi) in the School of Industrial Engineering and Management at Oklahoma State University. He joined
OSU in 2000 after 16 years in industry with Compaq, SEMATECH, General Electric, and Motorola. He has a
BS in Mathematics from East Texas Baptist College (1982), an MS in Industrial Engineering from Texas A&M
University (1984) and a PhD in Management Science from the University of Texas at Austin (1999). He is
serving as co-editor for the 2004 Winter Simulation Conference. His research interests include the supply
chain design issues and the development and application of qualitative discrete-event simulation. He is a
member of IIE. His e-mail address is ingalls@okstate.edu.

DOUGLAS MORRICE is Professor of Operations Management in the Department of Management in the McCombs
School of Business at the University of Texas at Austin. He received his PhD in Operations Research and Industrial
Engineering from Cornell University. Dr. Morrice’s research includes the modeling and analysis of service
operations, risk management, and supply chain management. His research has received funding from SAP
America under the University Alliance Research Grant Program. Additionally, he was a Visiting Research Scientist
for Schlumberger from 1996-1998. Dr. Morrice has over 40 technical publications and 2 patents. His research
has appeared in Management Science, Operations Research, IIE Transactions, and the Journal of Production and
Operations Management. Dr. Morrice is an area editor for the Association for Computing Machinery Transactions
on Modeling and Computer Simulation, an associate editor for the Journal of Production and Operations
Management, and a contributing editor for International Abstracts in Operations Research. He is a member of
the Institute for Operations Research and Management Science and the Production and Operations Management
Society. He has designed and delivered executive development programs on operations modeling and supply
chain management for several organizations including Schlumberger, Pricewaterhouse Coopers, and Texas
Instruments.

14 • Project Management Journal September 2004

THROUGH THE LOOKING GLASS: EXAMINING
THEORY DEVELOPMENT IN PROJECT MANAGEMENT
WITH THE RESOURCE-BASED VIEW LENS
ABSTRACT
Project management is a young discipline
and young disciplines tend to lack well-
developed theories. This paper examines
several topics that help with theory devel-
opment – the use of a common terminolo-
gy and holistic frameworks, the
importance of avoiding tautologies, and
the merits of analogies. To guide the
process, the paper draws from a recent
empirical study that used the Resource-
Based View to study project management
as a strategic asset. The paper discusses
how these four topics that contribute to
theory development were managed in the
study. Applying theory construction prac-
tices enables us to be more aware of the
challenges related to research and
improves our understanding of variables
as used in conceptual and empirical
papers. By applying the Resource-Based
View to project management, the paper
also shows how we can improve our
understanding of project management as
a source of competitive advantage.
Keywords: Resource-Based View; strate-
gy; theory construction; project manage-
ment; competitive advantage
©2004 by the Project Management Institute
Vol. 35, No. 3, 15-26, ISSN 8756-9728/03
KAM JUGDEV, Athabasca University, 8311-11 Street SW
Calgary, Alberta, Canada T2V 1N7
Introduction
Globally, companies are increasingly relying on project management to help
them complete projects more efficiently and effectively (Cleland & Ireland, 2002;
Project Management Institute, 2000). Although project management has been
around unofficially since time immemorial when people started to coordinate
tasks and activities with each other (e.g., from building shelters to ships to pyra-
mids), it is a young discipline (Verzuh, 1999). Evolving disciplines, such as proj-
ect management, often lack a fully developed theoretical base and tend to draw
from more established fields. As is often heard, “There is nothing more practical
than a good theory” (Meredith, 2002, p. 47). But what does this mean and how
is this relevant to project management?
This paper begins with a brief overview of the perceived state of theory devel-
opment in project management. Then, the paper examines some concepts relat-
ed to theory construction with an emphasis on several topics that help develop
theories - the use of a common terminology and frameworks, avoiding tautolo-
gies, and the merits of analogies. Thereafter, the paper introduces readers to the
Resource-Based View of the firm. The issue is that project management has not
been studied using the Resource-Based View and the dimensions of a strategic
asset in the discipline remain to be understood. This is an important topic
because it will help us understand the facets of project management that lead to
or contribute to a competitive advantage, so that companies can invest in the
appropriate practices and develop those internal assets relevant to positioning
project management strategically. This paper examines steps taken to study proj-
ect management using the Resource-Based View lens. It discusses some of the the-
ory development issues encountered and explains how they were managed. The
paper concludes with summary comments on theory development in project
management.
Project Management
Project management evolved in the 1950s from mathematical research based on
algorithms and project planning techniques (e.g., network planning efforts
involving Program Evaluation and Review Techniques and Critical Path Methods)
(Engwall, 2003; Packendorff, 1995). Over the years, a stream of literature devel-
oped in operations management and was devoted to computer applications and
September 2004 Project Management Journal • 15
expert systems for project planning, control, and risk analy-
sis (Packendorff, 1995). The planning approach was criti-
cized in the 1960s for an overemphasis on the rationalistic
approach, and this gave rise to a body of literature on proj-
ect organizational structures and project leadership. A num-
ber of the emerging areas of study in project management
are based on theories from management and focus on such
topics as human resources, organizational change, quality,
and business process reengineering (Kloppenborg & Opfer,
2002; Ulri & Ulri, 2000). However, a recent scientometric
study of project management for the 1987-1996 period of
over 3,565 articles indicated that the primary areas of publi-
cation continue to be in the technical domains of opera-
tions research, cost engineering, business process
reengineering, and infrastructure (Ulri & Ulri, 2000).
Kloppenborg and Opfer’s assessment of the project manage-
ment literature identified over 19,000 books on project
management within the 1960-1999 timeframe
(Kloppenborg & Opfer, 2002). The project management
textbooks tend to focus on normative advice on planning
and managing projects and the systems approach leads to
viewing project management as simply a tool based on
rationalistic views (Morris, 2001a).
In the United States, the Project Management Institute
first developed A Guide to the Project Management Body of
Knowledge (PMBOK® Guide) in 1976. The PMBOK® Guide
emphasizes time, cost, and scope, and uses a systems
approach based on inputs, processes, and outputs. Over
the years, revisions to the PMBOK® Guide did not change
the structure of the body of knowledge, nor did the revi-
sions introduce or support the need for a project manage-
ment theoretical foundation. Nonetheless, the PMBOK®
Guide continues to be put forth as the “generally accepted”
project management practices to its over 130,000 members
(Project Management Institute, 2000, p. 3). Similar bodies
of knowledge were developed by the Association of Project
Management in England and variations evolved in Austria,
France, Germany, Switzerland, and the Netherlands
(IPMA, 2000; Morris, 2001a, 2001b). The various bodies of
knowledge address professional qualifications and compe-
tences, but do not address project management in terms of
a theoretical base. It would be beneficial for a holistic body
of knowledge to address the importance of theory devel-
opment to the discipline and make efforts to delineate
these theories.
As most of the project management knowledge is based
on practitioner-driven normative approaches, there is a need
for a theory in the discipline (Engwall, 2003; Packendorff,
1995). “Most descriptive research on the management of
projects is relatively young and suffers from a weak theoret-
ical basis” (Engwall, 2003, p. 792). “Descriptive empirical
research grounded in theoretical problems is rare”
(Packendorff, 1995, p. 325). Recently, there have been calls
for better theory generation in project management
(Engwall, 2003; Meredith, 2002; Morris, 2001a;
Packendorff, 1995). Others have made this same criticism
over the years and the Project Management Journal® is cur-
16 • Project Management Journal September 2004
rently conducting a self-review to seek ways of improving its
academic quality.
Meredith recently assessed papers published in the
Project Management Journal from 1995-2001 (Meredith,
2002). In his critique, Meredith categorized the Project
Management Journal articles in terms of high, medium, and
low rigor. 63% of the Project Management Journal articles
were deemed to be of low methodological rigor and only
3% of the articles fit the high rigor category. Assessments of
this nature raise the question of how we can improve theo-
ry development and research within project management.
The two go hand-in-hand as a study’s theoretical underpin-
ning helps shape the research question and, in turn, the
methodology used. In his assessment, Meredith indicated
that the tutorial, background, and war story approaches are
not recognized as research methodologies (Meredith, 2002).
He further explains that research involves the use of a good
literature review to describe and understand phenomena
followed by an examination of cause-and-effect relation-
ships. Then, research builds on the literature and models to
develop frameworks that address the issue of “why.”
“Frameworks are essentially pre-theories” that are then test-
ed (Meredith, 2002, p. 48).
The above overview relates to the question of why we
need good theory in project management. Theories help us
advance the body of knowledge. Rigorous research encour-
ages the use of evidence-based practice (Meredith, 2002)
and can better position project management as a value-
adding practice with enhanced credibility.
Some Concepts from the Philosophy of Science:
What is Theory?
In general terms, a theory is a description of a new idea or
phenomenon, supported with empirical evidence, and
described in ways that others understand it. Theories are
important because they help explain and predict events, pat-
terns, and trends. Theories go beyond descriptions and
debunk common stereotypes and myths. Theories also help
develop and refine bodies of knowledge (Olszewski-Walker
& Coalson-Avant, 1995). In this section, the paper briefly
introduces readers to several definitions of theories and
research epistemologies. Then, the paper discusses how
models (paradigms) help with theory construction, the
challenges of working with partial theories, the importance
of avoiding tautologies in partial theories, and the value of
analogies to help explain concepts in partial theories.
Theories can be defined in specific or broad ways.
Specific definitions limit the term to the natural sciences,
and involve empirical verifications based on direct observa-
tion and experience as per the laws of physics (Herzog,
1996). For example, Kerlinger defines a theory as “a set of
interrelated constructs (variables), definitions, and proposi-
tions that presents a systematic view of phenomena by spec-
ifying relationships among variables, with the purpose of
explaining natural phenomena” (Kerlinger, 1973, p. 64).
However, a broader definition states that scientific knowl-
edge is a typology to organize things, explain past events
and predict future events, explain or help with understand-
ing event causes, and potentially control events (Reynolds,
1971). A definition of theory from the social sciences states
that theory is narrative and describes social process
(DiMaggio, 1995). Such more-inclusive definitions are
applicable to the types of studies conducted in the social
and management sciences, including project management.
Theory development also involves different ontologies
and research epistemologies. Ontology has to do with the
nature of reality regarding the phenomenon researchers seek
to understand. Ontological perspectives can range from
objective worldviews supporting one reality to subjective
worldviews supporting multiple realities. For example, the
scientific approach supports one worldview (or world
truth), whereas phenomenological studies, which are quali-
tative in nature, involve multiple realities as presented by
the various participants. Epistemology has to do with the
study of knowledge, that is, how we know what we know.
The long-standing epistemological debate on how to con-
duct research is rooted in two different inquiry paradigms.
“Logical-positivism, which uses quantitative and exper-
imental methods to test hypothetical-deductive general-
izations, versus phenomenological inquiry, using
qualitative and naturalistic approaches to inductively
and holistically understand human experience in con-
text-specific settings” (Patton, 1990, p. 37).
Research epistemologies also describe the relationship
of the researcher to subjects. In the positivist approach (as
per the scientific approach), the researcher is distant and
independent (or objective, context-free, and deductive) of
the subjects. In contrast, within the phenomenological par-
adigms, the researcher interacts more with those being
researched (Cresswell, 1998), and the approach is viewed as
subjective, contextual, and inductive. For example, in ethno-
graphic studies, the research is viewed as “insider” fieldwork
and the researcher interacts more with the participants
(Patton, 1990).
During theory development, models (paradigms)
evolve and help refine the theories. Paradigms are collec-
tions of shared beliefs. Some theoretical ideas are known as
radical changes in orientation or Kuhn’s paradigms to con-
note revolutionary changes. To exemplify, Newtonian
physics explains how planets orbit the sun, whereas quan-
tum physics explains how electrons surround the nucleus of
an atom. The shift from Newtonian physics to quantum
physics was a Kuhn’s paradigm because the common laws of
physics did not apply at the atomic level. Less radical
changes in orientation may involve refinements to detail
and can be called paradigm variations (Reynolds, 1971).
Mature sciences, such as the natural sciences, tend to have
well-developed paradigms as per the laws of physics. As the
social sciences are evolving, they typically lack well-devel-
oped paradigms or theories (Chalmers, 1999). Instead,
evolving disciplines tend to have perspectives that are in the
theory development stages. For example, the Resource-Based
View is a perspective from strategy with its roots in industri-
al organization economics (Conner, 1991).
Another aspect of theory building has to do with tau-
tologies. Tautologies involve circular reasoning. Tautologies
lack empirical content and some view them to be conceptu-
al problems that arise primarily in partial theories (Arend,
2003; Fahy, 2000; Powell, 2001; Priem & Butler, 2001b).
Tautologies are helpful, though, because they engender
debate on the theoretical perspective, challenge researchers
to revisit their methodologies, and bring to the forefront
issues that need to be researched to advance theories and
perspectives.
Theories do not have to be complete to make significant
contributions to knowledge (Herzog, 1996). Even partial
explanations are appropriate, since they serve as “pieces of
the puzzle” towards theory construction. Unlike studies
based on developed theories, where hypotheses are used to
test “if/then” relationships, researchers frequently use
propositions when the theoretical underpinning of the
study involves a perspective. Propositions allow researchers
to examine relationships that may not be of the causal
nature. Propositions can be thought of as hunches or ideas
and are appropriate to use with perspectives where research
on the relationships between variables is evolving
(Reynolds, 1971).
Partial explanations tend to lack a clear consensus on
terminology and concepts. Such terminology issues can be
problematic in developing a common understanding. In
addition, researchers may not always consider the work oth-
ers have done. This can make it difficult for researchers and
readers to grasp how the terms relate to each other and can
hinder theory development. Debate, discussion, and further
building on ideas put forth by others help develop theory,
and can lead to consensus on terms and concepts. For exam-
ple, well-developed literature reviews and meta-analyses
extend our understanding of the prevailing ideas on specific
topics and work done to date.
The process of theorizing involves simultaneous activi-
ties of abstracting, generalizing, relating, selecting, explain-
ing, synthesizing, and idealizing (Weick, 1995). These
activities contribute to the creation of reference lists, data,
variable lists, diagrams, and hypotheses. Although Sutton
and Staw (1995) reject these activities as non-theory, Weick
(1995) argues that these practices help build theory
because they help create frameworks that show relation-
ships, develop concepts and the interconnections between
them, and enable researchers to look at alternative expla-
nations. Arnoult (1976) identifies four characteristics of a
good theory:
• Testability refers to a theory that clearly ascertains
wrong predictions
• Power means that the theory correctly explains a
variety of phenomena
• Fertility means that the theory offers new ideas to
explore further
• A simple theory is elegant and parsimonious; it
involves few concepts and relationships.
September 2004 Project Management Journal • 17
 These characteristics can be considered the gold stan-
dard for theories, and are equally relevant to the social sci-
ences as they are to the natural sciences.
On the continuum of scientific to social science
approaches to theory development, there is no one right
way to develop theory, but there are many different and
acceptable ways of doing so, and many studies involve both
quantitative and qualitative methods (Chalmers, 1999;
Cresswell, 1994, 1998; Denzin, 1996; Hardy & Clegg, 1997;
Mason, 1998; Maxwell, 1996; Reynolds, 1971; Wacquant,
1993; Weick, 1995). Although there are different viewpoints
on the concept of theory, as well as research paradigms,
there is agreement that a theory explains phenomena and
the properties involved. It is important to be aware of the
different ways in which reality can be construed and knowl-
edge created as individual worldviews and beliefs influence
our approaches to theory development. Appraising a theory
or perspective through different worldview lenses helps us
see its various facets.
“Perspectives or views need not be complete theories in
order to contribute to our understanding of strategic man-
agement” (Priem & Butler, 2001a, p. 32). The Resource-
Based View is one such perspective and is gaining increasing
attention within strategy. The next section introduces the
Resource-Based View of the firm. This sets the stage to dis-
cuss the topics related to theory development that were
encountered and addressed in the empirical study on proj-
ect management.
The Resource-Based View of the Firm
Strategy is a vast field with a rich theoretical and empirical base.
“It’s a field where everything matters. Choices matter,
the leader matters, the culture matters, the values mat-
ter, random events matter, and so on. Strategy is inher-
ently an integrative subject that has to allow for
complexity” (Argyres & McGahan, 2002, p. 52).
Within strategy, the competitive advantage literature is
increasingly important as compressions of distance and
time intensify competition and focus managerial attention
on multiple internal and external factors (Thomas, Pollock,
& Gorman, 1999). To outperform rivals, firms must deliver
greater value to customers or create comparable value at a
lower cost (Porter, 1996). Firms must make choices in areas
such as products, services, goals, and industry positioning,
to name a few. In practical terms, strategic management is
about “the direction of organizations” and deals with com-
petition and firm success and failure (Rumelt, Schendel, &
Teece, 1994, p. 9). Firm performance depends on a number
of factors including those factors external to the company,
such as the macro-economic environment (Porter, 1996)
and those internal to it, such as the company’s internal
assets and management practices (Barney, 2002). The per-
spective of looking at factors external to the firm is called the
Industry View and the perspective of examining factors with-
in the firm is called the Resource-Based View.
18 • Project Management Journal September 2004
The Industry View looks to the marketplace to help
firms determine the areas in which they want to compete
(Andrews, 1980). Porter introduced the industry-focused
strengths, weaknesses, opportunities, and threat model,
along with the five structural forces techniques that can
erode a firm’s long-term industry average profitability
(Porter, 1991). The five forces are threats of new entrants,
bargaining power of suppliers, rivalry among existing com-
petitors, bargaining power of buyers, and threats of substi-
tute products or services (Collis & Montgomery, 1995;
Porter, 1991). The Industry View is a good ex-post descrip-
tion of market conditions and allows firms to identify some
of the conditions for normal rent returns, but does not pro-
vide information on above-normal profits (Chakraborty,
1997). The Industry View downplays resource heterogeneity
and immobility, which are two key tenets of examining
competitive advantage in terms of a firm’s internal assets
(Amit & Schoemaker, 1993; Barney, 1991). In contrast, the
Resource-Based View underscores the need for a balanced
approach to competitive advantage involving a firm’s
strengths and weaknesses, along with industry components
(Barney & Zajac, 1994). In the Resource-Based View, com-
panies focus on their asset mix (organizational, physical,
financial, human, and social assets) and determine which
ones they should develop and invest in further and which
ones to de-emphasize.
A crucial question in the strategy literature asks, “Why
do firms differ?” In contrast to the Industry View that
emphasizes the environment, the Resource-Based View
explains firm existence based on internal assets that are
scarce, difficult to trade, imitate, appropriate, and give a firm
its competitive advantage (Amit & Schoemaker, 1993;
Madhok, 2002; Porter, 1991). The Resource-Based View
emphasizes the creation, maintenance, and renewal of a
competitive advantage through a firm’s unique resources,
their characteristics, and how they change over time (Foss,
1997; Schulze, 1994). According to Barney, a strategic asset
is characterized by being valuable, rare, inimitable, and hav-
ing an organizational focus, otherwise known as the VRIO
framework (Barney, 2001). According to Barney, a firm
needs four attributes for a sustained competitive advantage.
In his original framework, Barney indicated that resources
must be valuable, rare, inimitable, and non-substitutable
(VRIN) (Barney, 1991). In a later model, Barney revised the
framework and replaced the term “non-substitutable” with
“organizational focus,” which referred to managerial sup-
port (Barney, 1998; Chakraborty, 1997).
A firm’s strategic assets may not always be visible; exam-
ples of strategic assets include quality, reputation, manageri-
al skills, brand recognition, patents, culture, technological
capability, customer focus, and superior managerial skills
(Barney & Zajac, 1994; Castanias & Helfat, 1991;
Chakraborty, 1997; Hawawini, Subramanian, & Verdin,
2002; Kogut & Zander, 1993). A firm’s formal processes and
production functions are the backbones that support strate-
gic assets, and firms protect their assets through business
practices. Strategic assets are unique to the firm. Strategic
assets signify complex, higher order interactions between
resources, processes, and knowledge (Eisenhardt & Martin,
2000; Grant, 1991; Henderson & Cockburn, 1994; Kaplan,
Schenkel, von Krogh, & Weber, 2001; Kogut & Zander, 1992;
Liedtka, 1996; Teece, Pisano, & Shuen, 1997). Strategic
assets involve a mix of explicit and tacit knowledge that is
embedded in a company’s unique internal skills, knowl-
edge, and resources (Foss, 1997; Rumelt et al., 1994). Such
strengths are difficult to purchase, let alone copy, so they can
contribute to a firm’s ability to move beyond competitive
convergence toward a competitive advantage or strategic
position. Other synonyms for strategic assets include core
competences (Prahalad & Hamel, 1990), distinctive compe-
tence (Selznick, 1957), dynamic capability (Teece et al.,
1997), dynamic routines (Collis, 1991), indivisible assets
(Teece, 1980), integrative capabilities, implicit / social
knowledge, meta-capability (Kaplan et al., 2001), organiza-
tional architecture (Henderson & Cockburn, 1994), and
organizational capability. However, the distinctions between
these terms are not always obvious.
Within the Resource-Based View literature, there is a
lack of clarity on resource characteristics that help develop
versus sustain a competitive advantage (Amit & Schoemaker,
1993; Barney, 1991, 1998; Chakraborty, 1997; Collis &
Montgomery, 1995; Grant, 1991; Mata, Fuerst, & Barney,
1995; Peteraf, 1993; Priem & Butler, 2001a). For example,
Amit and Schoemaker view scarcity as a sustaining feature,
but Barney and Peteraf view it as a feature that develops a
competitive advantage. As these distinctions can be confus-
ing, I decided to use Barney’s VRIO framework because it
covers the main characteristics of strategic assets addressed
in the literature and has been used for empirical studies
(Barney, 1991, 1998, 2002; Ray, Barney, & Muhanna, 2004).
The VRIO concepts are presented in the following table.
Valuable? Rare Difficult to Supported by Competitive
Imitate? Organization? Implications
No Competitive
Disadvantage
Yes No Competitive Parity
Yes Yes No Temporary Competitive
Advantage
Yes Yes Yes Sustained Competitive
Advantage
Table 1: VRIO Framework (adapted from Barney, 1998)
If a resource is only valuable in the VRIO framework, it
leads to competitive parity. Both value and rarity are
required for a temporary competitive advantage and value,
rarity, and inimitability are required for a sustained compet-
itive advantage (Barney, 1998). An organizational focus is
necessary to both develop a competitive advantage and sus-
tain it (Barney, 1998).
An extensive literature review did not indicate that proj-
ect management had been assessed with the Resource-Based
View, and only one article discussed project management in
terms of core competences (DeFillippi & Arthur, 1998). The
Resource-Based View is appropriate to use as a theoretical
underpinning for project management for several reasons.
First, the Resource-Based View has a rich 20-year history of
development that includes the VRIO framework. Second,
the Resource-Based View addresses knowledge and process
assets and this fits with the exploration of project manage-
ment as a complex resource. The Resource-Based View has
relevance to project management because it emphasizes the
less tangible human and organizational assets that also
involve social and intellectual capital (knowledge, skills,
and know-how) (Lundvall & Johnson, 1994). Finally, one
way of achieving theory status is to conduct studies with per-
spectives in the theory construction stage.
Examining Theory Development in Project Management with
the Resource-Based View Lens
In 2003, I completed an empirical study on how companies
develop and sustain project management as a strategic asset
(Jugdev, 2003). The study began with an extensive literature
review of several fields - project management, strategy, and
the Resource-Based View. This led me to appreciate the com-
plexities of the Resource-Based View and to better under-
stand the difference between a theory and perspective. The
literature review enabled me to develop two research ques-
tions to investigate - what characterizes project management
when it is considered a strategic asset, and how do firms
develop and sustain a competitive advantage in project
management? The literature review and understanding of
this key concept enabled me to translate the concept into
constructs for which data was gathered using a case study
approach.
The multiple case study explored the characteristics of a
strategic asset in project management and the processes
companies use to develop and sustain the advantage. The
study involved four international companies from the finan-
cial, telecommunications, manufacturing, and utility indus-
Performance tries. Senior, middle, and project managers from the
companies participated in interviews and a survey. Sixty-
Below Normal
seven interviews were conducted and 28 responses to a proj-
ect management maturity survey were analyzed. The
Normal
qualitative data was coded and textually analyzed using
Above Normal
Atlas.ti® software and the survey results were analyzed using
Above Normal non-parametric tests with SPSS® software.
Initially, I considered using project management matu-
rity models to assess project management within companies
as a source of competitive advantage. The project manage-
ment maturity models are based on the Carnegie-Mellon
Software Engineering Institute’s Capability Maturity Model
for software development (Carnegie-Mellon, 2002). The
models consist of five linear stages reflecting software
processes and practices that are increasingly more defined
and repeatable. The five levels are: Level 1: Initial Process;
Level 2: Structured Process and Standards; Level 3:
Organizational Standards and Institutionalized Process;
Level 4: Managed Process; Level 5: Optimizing Process
(Pennypacker, 2001). The software development maturity
models are based on a theoretical foundation of software
process improvement and total quality management
September 2004 Project Management Journal • 19
(Ngwenyama & Nielsen, 2003), yet others argue that the
models are atheoretical (Carnegie-Mellon, 2002; Dymond,
1995). The models use a technical, rational, and mechanis-
tic view of organizations because they do not address the
social aspects of companies (Ngwenyama & Nielsen, 2003).
The project management maturity models address tan-
gible assets, but not intangible assets (knowledge assets).
The models do not emphasize organizational processes and
practices. The models typically lack a connection between
operations management and strategy. Few project manage-
ment maturity models have been empirically tested and
many are based on anecdotal material, case studies, or
espoused best practices (ESI-International, 2001; Hartman &
Skulmoski, 1998; MicroFrame, 2001; Pennypacker, 2001;
Schlichter, 2000; Skulmoski, 2001). In addition, as these
models do not draw from the economic or strategy literature
on competitive advantage, or meet the VRIO criteria, the
arguments put forth towards winning in the marketplace
with such models are weak at best (Jugdev & Thomas,
2002). Consequently, I turned to the Resource-Based View
as a theoretical underpinning for the study.
The study brought to light several issues in terms of the-
ory development including the need for common terminol-
ogy, the importance of an accepted framework, tautologies,
and the value of an analogy to show connections between
constructs.
Creating a Common Terminology to Help Build Theory
A perspective differs from a theory in that it still involves
issues of terminology and concept confusion, whereas a the-
ory has generally addressed many of these concerns.
Terminology issues are common in theory construction and
reflect an ongoing exploration of conceptual frameworks
(Shaw & Gaines, 1995). Terminology issues were evident in
the Resource-Based View. For example, when it comes to the
word “resource,” some authors use the term to refer to tan-
gible assets and others use it to refer to both tangible and
intangible assets. Barney (2001) describes resources as all
the firm’s assets, capabilities, competencies, organizational
processes, firm attributes, information, and knowledge,
while Schulze (1994) defines resources as “the set of assets
and skills which are employed to create and support a com-
petitive advantage” (p. 37). Some researchers interpret
resources as basic building blocks (assets, inputs, primary
resources), and others view them as more complex combi-
nations (e.g., resource bundles, complementary assets,
strategic assets, stocks, competences, capabilities, meta-capa-
bilities, and routines) (Foss, 1997).
The literature also uses different terms to describe tan-
gible and intangible resources, and not all authors distin-
guish between the two. Some group resources as tradable
and non-tradable assets (Dierickx & Cool, 1989). Non-trad-
able assets (i.e., reputations, dealer loyalty, trust, and know-
how) cannot be bought or sold and are non-transferable or
non-appropriable (Kogut & Zander, 1993). They are in fixed
supply, and build up over time through training and learn-
ing as successful strategies are implemented. Non-tradable
20 • Project Management Journal September 2004
assets are viewed as sources of durable firm heterogeneity
and are known as “sticky” resources because of their close
connections to the firm or person (Collis, 1994; Grant,
1991; Priem & Butler, 2001a). In other words, they cannot
be exchanged like commodities, but can offer social benefits
or costs (Tcha, 1997). The Resource-Based View literature
also involves many synonyms for strategic assets, as men-
tioned earlier.
The literature review confirmed the lack of consensus
regarding terminology within the Resource-Based View, and
this enabled me to develop an extensive table on the syn-
onyms and terms used for tangible resources, intangible
resources, and isolating or protective mechanisms, identify
who coined certain terms, and refine my understanding of
the terms. The review confirmed that my understanding was
evolving as I encountered other publications with similar
frameworks once I had completed mine (Fahy, 2000).
Although Barney states that “distinctions among
resources, capabilities, and competencies can be drawn in
theory, it is likely that they will become badly blurred in
practice” (Barney, 2001, p. 157). Such terminology issues
can be problematic in developing a common understanding
among scholars and must be addressed as part of develop-
ing theory. Further research, debate, and dialogue can help
explore such issues and eventually lead to disentangling ter-
minology issues. On this basis, I decided to use the words
“resource” and “asset” as synonyms and used the term
“strategic asset” to encompass the synonyms for this term in
the literature. Since completing the strategic asset study and
in discussions with other academics in strategy, I am further
refining my understanding that perhaps in project manage-
ment, we should use the term “meta-resource” until we have
a more solid base of empirical studies supporting it as a
strategic asset.
The breadth of terms used in the Resource-Based View
underscores the importance of a common language, as it
was sometimes a difficult perspective to understand. With
an appreciation of some of the concepts from philosophy of
science, I was able to understand the rationale for using
propositions and not hypotheses. Propositions examine
relationships that may not be of the causal nature.
Propositions are like hunches or ideas and are appro-
priate to use with perspectives where research on the rela-
tionships between variables is evolving (Reynolds, 1971). I
also learned the importance of conducting a thorough liter-
ature review. This helped me understand the breadth and
depth of publications on the Resource-Based View, the
degree to which theories relevant to the topic have been
developed, the extent of empirical studies in the field, and it
improved my understanding of the issues that are being
debated. The array of Resource-Based View terms used in the
literature also reflected the need for a holistic resource
framework.
Developing Frameworks to Help Build Theory
Theory building involves creating reference lists, data, vari-
able lists, diagrams, and frameworks (Weick, 1995). The
Resource-Based View provides evidence of the use of frame-
works. Within the perspective, resource frameworks show
preliminary groupings of elements in a logical order and
depict how various components fit into an overall structure.
The resource frameworks map components into an organ-
ized and logical whole to explain phenomena. Some exam-
ples of resource groupings include physical, organizational,
human, social, financial, technological, and intangible
assets (Barney, 1991, 1998; Brush, Greene, Hart, & Haller,
2001; Marino, 1996; Thomas et al., 1999). Other examples
of groupings in the Resource-Based View literature focus on
grouping assets hierarchically, such as in the pyramid of
value creation. The pyramid of value creation involves
generic resources at the base, followed by core competences,
and strategic assets at the apex (Brush et al., 2001).
A review of the literature involved an assessment of
some of the Resource-Based View frameworks used to
explain how firms develop and sustain a competitive advan-
tage (Amit & Schoemaker, 1993; Barney, 1991, 1998;
Chakraborty, 1997; Collis & Montgomery, 1995; Grant,
1991; Mata et al., 1995; Peteraf, 1993; Priem & Butler,
2001a). The resource frameworks covered the breadth of
characteristics deemed relevant in developing and sustain-
ing a competitive advantage including valuable (important),
rare (unique), inimitable (difficult to copy), organizational
focus (management support), low-tradable (non-transfer-
able), durable (long-lasting), and non-substitutable (noth-
ing comes close to replacing a certain resource). Most of the
authors consistently cited inimitability in their frameworks.
Surprisingly, there was a lack of consensus among the
authors on the relative use of the array of Resource-Based
View characteristics. The resource characteristics that some
authors identified as helping develop a competitive advan-
tage were viewed by others as sustaining a competitive
advantage. In addition, few authors discussed organization
focus as explicitly as Barney did (1991, 1998, 2002). On this
basis, I was able to focus on the VRIO framework with con-
fidence, as it has been extensively developed and empirical
studies conducted with it. The VRIO framework served as
the basis for the research methodology for the strategic asset
study on project management. It guided the development of
the semi-structured interview instrument to explore the two
research questions on project management as a strategic
asset.
In terms of radical or less dramatic theoretical ideas, the
VRIO framework is a paradigmatic variation of the
Resource-Based View as it involves refinements to details
(Reynolds, 1971). It took a decade or so to develop the VRIO
framework. Like one piece of the puzzle, the VRIO frame-
work is a contribution toward the efforts of other researchers
to develop the Resource-Based View and adds to our under-
standing of strategic assets. In terms of the merits of frame-
works, I was able to appreciate that when frameworks (such
as the VRIO framework) have been developed and are used
repeatedly, it makes sense to conduct further studies using
them, rather than proposing dramatic changes without an
empirical basis of support.
In addition to terminology issues and the need for a
holistic framework, some perspectives often involve tautol-
ogy issues. In the course of the strategic asset study, I
encountered some readings on tautology issues and these
are discussed next.
Avoiding Tautology Traps to Help Build Theory
Within the Resource-Based View, strategic assets are often
described as being valuable, rare, inimitable, and having an
organizational focus (Barney, 2002). A tautology may exist
within the Resource-Based View if a competitive advantage
is defined in terms of value and rarity, and the same terms
are used as the resource characteristics (Barney, 2001). A tau-
tology can be avoided if resource characteristics are defined
independent of firm performance so that empirical falsifica-
tion is possible. Another way of dealing with tautology
issues is to develop proxy variables.
As evident from debates such as those between Priem
and Butler and Barney, the tautology issue in the Resource-
Based View continues to gain prominence (Barney, 2001;
Priem & Butler, 2001a). Priem and Butler argued that
Barney’s 1991 publication on the Resource-Based View in
the Journal of Management was tautological. This may be the
case, but Barney’s contributions have helped advance the
Resource-Based View, and those that conducted empirical
studies thereafter were more cautious about how they
defined resource variables independent of firm perform-
ance. Barney continues to be a noted authority on the
Resource-Based View.
As scholars grapple with theoretical concepts and the
challenges of operationalizing them, they often discuss tau-
tologies and ways in which they can be avoided. These dis-
cussions reflect different views on how tautologies are
defined and are useful in helping scholars revisit how
Resource-Based View terms are operationalized. The discus-
sions also heighten awareness on key issues within the per-
spective. Dealing with tautologies will help develop the
Resource-Based View according to Arnoult’s characteristics
of a good theory having increased testability (Arnoult,
1976). Over time, a simple and elegant theory of the
Resource-Based View may emerge.
The tautology issue was circumvented in the strategic
asset study as it was an exploratory study that did not strive
to measure a dependent variable (such as superior perform-
ance), but focused instead on the range of independent vari-
ables (antecedents) such as resource characteristics and
organizational practices that potentially contribute to a
competitive advantage.
The strategic asset study resulted in a rich dataset of
qualitative findings on the main constructs and has brought
me to the point now where I can focus on a quantitative
study that builds on the first study. The next research project
(which is currently underway) involves operationalizing the
VRIO constructs with a survey to assess the extent to which
project management is a meta-resource within companies in
North America. This study will use two dependent variables
- firm performance and project management performance.
September 2004 Project Management Journal • 21
The study is being structured to avoid the tautology trap, as
the measures for the independent and dependent variables
will not be the same.
Firm performance has long been assessed using finan-
cial and economic measures. Performance has many
meanings and can be assessed with long versus short and
financial versus relationship building approaches
(Deshpande, Farley, & Webster, 1993). I plan to use a com-
bination of financial and subjective measures. Various
databases offer company financial data and I can also
measure firm performance with Likert scale questions ask-
ing participants to assess their company on profitability,
size, market share, and growth rate in comparison to the
largest competitors for that particular business
(Deshpande et al., 1993).
However, using a highly aggregate dependent variable,
such as firm performance, may not be the optimal way of
examining the Resource-Based View and could lead to mis-
leading conclusions, so some recommend the use of less
aggregate (intermediate) variables (Ray et al., 2004). As a
result, I plan on using project management performance as
another dependent variable. This variable will be assessed
using the VRIO framework. The findings from the prior
study from four international companies enabled me to
operationalize the VRIO concepts and awareness on tau-
tologies; such experience is enabling me to avoid this trap in
the upcoming empirical study. To summarize, when work-
ing with complex perspectives such as the Resource-Based
View, frameworks and tautologies help elucidate compo-
nent parts and interrelationships. Another way of under-
standing complex topics within theory development
involves the use of analogies.
Genes: Strategic Asset
Genes:
Tangible Resources
Hydrogen Bonds:
Intangible Resources
(i.e., Knowledge-Based
Resources)
Helices:
Business & Process
Backgrounds
Cellular Structures:
Isolating Mechanisms
Figure 1: Strategic asset genome
22 • Project Management Journal September 2004
Developing Analogies to Help Build Theory
Based on Arnoult’s system of classifying theories into one of
four categories - metaphors, analogies, reductionist (sim-
pler) concepts, and abstract (mathematical) relationships,
the strategic asset study made use of the analogical approach
(Arnoult, 1976). The use of metaphors, similes, and analo-
gies is helpful in theory development because they encour-
age thinking in different ways and draw on symbolic
constructs to help explain reality (Tsoukas, 1991).
Metaphors, similes, and analogies are not widely used in the
natural sciences where theories are well developed.
Metaphors, similes, and analogies help us make infer-
ences about something that we know less about on the basis
of what we know of other things. In the project management
study, a genome analogy was used to depict the strategic
asset genome and this is discussed next.
Just as scientists around the world are working to crack
the biochemical code for each chromosome to determine
human characteristics, strategy and economics researchers
are working to unlock the code for competitive advantage
creation (Boulton, 2000). Unlocking the genome of a com-
pany’s competitive advantage involves defining, grouping,
and characterizing the component parts used to develop a
competitive advantage. Figure 1 portrays this analogy.
Within cells of complex organisms, the nucleus consists
mainly of the chromosome (genome) that carries the genet-
ic information within the genes, and cellular structures that
power chemical reactions. In humans, the chromosomes
consist of thousands of genes made up of deoxyribonucleic
acid (DNA).
In turn, each DNA structure consists of chemical
nucleotide bases linked by hydrogen bonds. The chromo-
some is like a twisted stepladder or helix, with the genetic
information located on the rungs (genes) (DNASciences,
2003). Although the genes are an important building
block, the chromosome, as a whole, is vital in passing on
the genetic code to the next generation. As well, the hydro-
gen bonds may appear to be nondescript and secondary,
but they are crucial in how the DNA nucleotide bases
combine with each other. Understanding how all the
genetic components interact is paramount to breaking the
genetic code.
By using an analogy, some of the earlier concepts of
the Resource-Based View may be easier to understand, as
well as the relationships among them. In relating the
genome to the strategic asset genome, a firm’s tangible
resources are analogous to the vast number of genes. The
business and process backbones within the firm are like
the helix. They support strategic assets through deci-
sions, organizational policies, procedures, and practices
to create products and services (Brush et al., 2001).
Capabilities, competences, and routines are knowledge
assets and are analogous to the gene’s hydrogen bonds.
They hold the tangible resource genes together and are
crucial in creating strategic assets. Akin to cells protect-
ing their genomes through cellular structures and
processes, firms protect their assets through isolating
mechanisms such as history and causal ambiguity. The
strategic assets are equivalent to the chromosomes.
Strategic assets are as unique to each firm as a genetic
blueprint for reproduction is to each person. Additional
parallels between the human and strategic asset genome
follow:
• Just as the human genome consists of more genes
than chromosomes, a firm’s genome consists of more
resources than strategic assets.
• Just as individuals differ in their genetic makeup, so
do organizations and their resource mix; some assets
are also more important than others.
• Although the chromosome is ultimately more impor-
tant than the individual genes for reproductive pur-
poses, it requires the genes and hydrogen bonds.
Similarly, strategic assets are more important than
individual resources, but basic resources are essential
in creating strategic assets.
• Just as chemistry determines how the DNA bases
bond specifically to certain bases through hydrogen
bonds, a company’s intangible resources (invisible
assets, capabilities, routines, processes, or compe-
tences) are like knowledge bonds that serve a similar
function. They are vital in linking resources and creat-
ing meta-resources.
• Similar to the human genome involving long DNA
strands that comprise chromosomes, a firm’s formal
processes and production functions are the back-
bones supporting strategic assets.
• Just as cells protect their genome with cellular struc-
tures and chemical processes, firms protect assets
from being copied by rivals through isolating mech-
anisms (i.e., history, causal ambiguity, and social
complexity).
• As part of evolution, the genetic code passes from par-
ents to offspring (e.g., hair color, eye color, and other
traits). Similarly, a firm’s strategic assets (chromo-
somes) characterize its strengths and can be passed to
spin off companies, joint ventures, and partnerships.
Taking this analogy one step further helps us understand
project management in terms of the genome analogy. The
following is a brief excerpt of findings from the strategic asset
study, reflecting practices that make project management
valuable, rare, inimitable, and with an organizational focus.
• As a knowledge-based discipline, project management
consists of tangible assets (genes) such as tools, tech-
niques, templates, methodologies, and other forms of
explicit knowledge called “know-what.”
• In terms of project management business and process
backbones (helices), the strategic asset study identifies
the use of leadership and solid support for the disci-
pline, the establishment of organization-wide project
management programs, continuous improvement
practices, and connections between project manage-
ment and business outcomes, as examples of man-
agement practices supporting the discipline.
• The intangible resources (hydrogen bonds) within
project management are evident in such practices as
staff embracing the discipline, clear links between the
organizational and project management cultures, and
staff sharing tacit knowledge or “know how,” through
social capital practices.
• Regarding isolating mechanisms (cellular structures),
there is evidence that companies that view and sup-
port project management as a strategic asset experi-
enced periods of stabilization in the history of how
the discipline evolved. One company went from
“project management strength to strength” over a 20-
year period. There were also indications that the disci-
pline involved social complexity. The financial
institute exhibited the strongest profile of project
management as a strategic asset and it capitalized on
its knowledge-sharing practices. It had a better appre-
ciation of the importance of social capital as a way of
sharing tacit knowledge.
To summarize, I used the VRIO framework to empiri-
cally examine project management as a strategic asset. I had
developed the analogy in the course of data analysis. As I
reviewed the data for different themes, the analogy helped
me further examine aspects of intangible assets within proj-
ect management. The genome analogy helped depict the
connections between the various components of the
Resource-Based View and helped portray the constructs in
relation to project management. The analogy was also use-
ful because it drew my attention to the importance of intan-
gible assets (hydrogen bonds) in project management. What
was striking about the value of the analogy was that it did
not simply depict the relationships between the genome
and strategic asset. Instead, the analogy also focused my
attention on what appears to be a small aspect of the dia-
gram - the hydrogen bonds. The hydrogen bonds are almost
invisible, yet vital to how the genome functions. Similarly,
intangible assets seem to be undervalued in project man-
agement, yet they potentially play a crucial role in project
management as it is a knowledge-based discipline, and tacit
knowledge and social capital are significant sources of
knowledge exchange. The theme of intangible assets is
emerging as a crucial component of the current study I am
working on, whereby I am assessing both the tangible and
intangible assets in project management and their connec-
tions to project management as it meets the Resource-Based
View criteria of being valuable, rare, inimitable, and having
an organizational focus. In particular, my focus will be on
tacit knowledge and social capital in project management.
Conclusions
Developing good theory takes time and is an incremental
process. This paper examined the topics of a common ter-
minology, frameworks, tautologies, and analogies in rela-
tion to theory construction. The paper used some lessons
learned from a prior empirical study to exemplify how these
theory development concepts were used.
The Resource-Based View is work in progress towards a
September 2004 Project Management Journal • 23
holistic theory. Over 100 articles are published on this per-
spective each year (Rouse & Daellenbach, 2002). More
recently, an increasing number of publications are taking
empirical approaches to the Resource-Based View (Castanias
& Helfat, 2001; Collis, 1991; Cool & Schendel, 1987;
Henderson & Cockburn, 1994; Levinthal & Myatt, 1994;
Lopez, 2001; Montealegre, 2002; Ray et al., 2004; Wiggins &
Ruefli, 2002). This groundswell of support and research will
help raise awareness of the contributions being made to the
field, foster idea sharing, and encourage researchers to build
on previous works. Over time, a clearer and more complete
picture of the Resource-Based View perspective will emerge.
In parallel, we can capitalize on the advances made
within the Resource-Based View. We can continue to do
research on project management using the Resource-Based
View lens. By doing so, we can contribute to knowledge
development in project management and strategy. By exam-
ining project management with the Resource-Based View
lens, we can develop a clearer picture of the characteristics of
project management that contribute to a competitive advan-
tage and better understand the connections between project
management and strategy.
References
Amit, R., & Schoemaker, P. J. H. (1993). Strategic assets and
organizational rent. Strategic Management Journal, 14(1), 33-46.
Andrews, K. (1980). The concept of corporate strategy. In N.
Foss (Ed.), Resources, firms, and strategies: A reader in the resource-
based perspective (1st ed., Vol. 1, pp. 52-59). Oxford, UK: Oxford
University Press.
Arend, R. J. (2003). Revisiting the logic and research con-
siderations of competitive advantage. Strategic Management
Journal, 24(3), 279-284.
Argyres, N., & McGahan, A. M. (2002). An interview with
Michael Porter. Academy of Management Executive, 16(2), 43-52.
Arnoult, M. D. (1976). Fundamentals of scientific method in
psychology (2 ed.). Dubuque, IN: William C. Brown.
Barney, J. B. (1991). Firm resources and sustained competi-
tive advantage. Journal of Management, 17(1), 99-120.
Barney, J. B. (1998). On becoming a strategic partner: The
role of human resources in gaining competitive advantage.
Human Resource Management, 37(1), 31-46.
Barney, J. B. (2001). Is the resource-based “view” a useful
perspective for strategic management research? Yes. Academy of
Management Review, 26(1), 41-56.
Barney, J. B. (2002). Gaining and sustaining competitive advan-
tage (2nd ed.). Upper Saddle River, NJ: Prentice-Hall, Inc.
Barney, J. B., & Zajac, E. J. (1994). Competitive organiza-
tional behaviour: Toward an organizationally-based theory of
competitive advantage. Strategic Management Journal, 15(1), 5-9.
Boulton, R. E. S. (2000). A business model for the new
economy. The Journal of Business Strategy, 21(4), 29-35.
Brush, C. G., Greene, P. G., Hart, M. M., & Haller, H. S.
(2001). From initial idea to unique advantage: The entrepre-
neurial challenge of constructing a resource base. The Academy of
Management Executive, 15(1), 64-78.
Carnegie-Mellon. (2002, November 15). Carnegie Mellon
Software Engineering Institute: Capability maturity models. Retrieved
November 20, 2002, from http://www.sei.cmu.edu/cmmi/
24 • Project Management Journal September 2004
Castanias, R. P., & Helfat, C. E. (1991). Managerial resources
rents. Journal of Management, 17(1), 155-171.
Castanias, R. P., & Helfat, C. E. (2001). The managerial rents
model: Theory and empirical analysis. Journal of Management,
27(6), 661-678.
Chakraborty, K. (1997). Sustained competitive advantage: A
resource-based framework. Advances in Competitiveness Research,
5(1), 32-63.
Chalmers, A. F. (1999). What is this thing called science? (3rd
ed.). Indianapolis, IN: Hackett Publishing Company Inc.
Cleland, D. I., & Ireland, L. (2002). Project management:
strategic design and implementation (4th ed. Vol. 1). New York:
McGraw-Hill.
Collis, D. J. (1991). A resource-based analysis of global com-
petition: The case of the bearings industry. Strategic Management
Journal, 12(Summer, special issue), 49-68.
Collis, D. J. (1994). Research note: How valuable are organi-
zational capabilities? Strategic Management Journal, 15(1), 143-
152.
Collis, D. J., & Montgomery, C. A. (1995). Competing on
resources: Strategy in the 1990s. Harvard Business Review, 73(4),
118-128.
Conner, K. R. (1991). A historical comparison of resource-
based theory and five schools of thought within industrial organ-
ization economics: Do we have a new theory of the firm? Journal
of Management, 17(1), 121-154.
Cool, K. O., & Schendel, D. E. (1987). Strategic group for-
mation and performance: The case of the U.S. pharmaceutical
industry, 1963-1982. Management Science, 33(9), 1102-1124.
Cresswell, J. W. (1994). Research design: Qualitative & quanti-
tative approaches. Thousand Oaks, CA: Sage Publications, Inc.
Cresswell, J. W. (1998). Qualitative inquiry and research design:
Choosing among five traditions. Thousand Oaks, CA: Sage
Publications, Inc.
DeFillippi, R. J., & Arthur, M. B. (1998). Paradox in project-
based enterprise: The case of film making. California Management
Review, 40(2), 125-139.
Denzin, N. K. (1996). The epistemological crisis in the
human disciplines: Letting the old do the work of the new. In R.
Jessor, A. Colby & R. A. Shweder (Eds.), Ethnography and human
development: Context and meaning in social inquiry (pp. 127-151).
Chicago: University of Chicago Press.
Deshpande, R., Farley, J. U., & Webster, F. E. (1993).
Corporate culture, customer orientation, and innovativeness in
Japanese firms: A quadrad analysis. Journal of Marketing, 57(1),
23-27.
Dierickx, I., & Cool, K. (1989). Asset stock accumulation and
sustainability of competitive advantage. Management Science,
35(12), 1504-1514.
DiMaggio, P., J. (1995). Comments on “What theory is not”.
Administrative Science Quarterly, 40(3), 391-397.
DNASciences. (2003, January 25). DNA Sciences Inc.
Retrieved January 21, 2003, from
h t t p : / / w w w. d n a . c o m / h o m e / h o m e . j s p ? j s e s s i o n -
id=1150011043159995906&link=Home.htm
Dymond, K. M. (1995). A guide to the CMM: Understanding
the capability maturity model for software. Retrieved November 20,
2002, from http://home.okstate.edu/homepages.nsf/
toc/level2.info.book1.html
Eisenhardt, K. M., & Martin, J. A. (2000). Dynamic capa-
bilities: What are they? Strategic Management Journal, 21(10-11),
1105-1121.
Engwall, M. (2003). No project is an island: Linking proj-
ects to history and context. Research Policy, 32(5), 789-808.
ESI-International. (2001, November 22). ESI-International.
Retrieved November 20, 2002, from http://www.esi-intl.com/
Fahy, J. (2000). The resource-based view of the firm: Some
stumbling-blocks on the road to understanding sustainable
competitive advantage. Journal of European Industrial Training,
24(2/3/4), 94-104.
Foss, N. J. (Ed.). (1997). Resources, firms, and strategies: A
reader in the resource-based perspective (1st ed. Vol. 1). Oxford,
UK: Oxford University Press.
Grant, R. M. (1991). The resource-based theory of compet-
itive advantage: Implications for strategy formulation. California
Management Review, 33(3), 114-135.
Hardy, C., & Clegg, S. (1997). Relativity without relativism:
Reflexivity in post-paradigm organization studies. British Journal
of Management, 8(Special Issue), S5-S17.
Hartman, F. T., & Skulmoski, G. (1998). Project manage-
ment maturity. Project Management, 4(1), 74-78.
Hawawini, G., Subramanian, V., & Verdin, P. J. (2002). Is
performance driven by industry - or firm-specific factors? A new
look at the evidence. Strategic Management Journal, 24(1), 1-16.
Henderson, R., & Cockburn, I. (1994). Measuring compe-
tence? Exploring firm effects in pharmaceutical research.
Strategic Management Journal, 15(1), 63-84.
Herzog, T. R. (1996). Research methods in the social sciences
(1st ed.). New York: Harper Collins College Publishers.
IPMA. (2000, Nov 20). IPMA Certification. Retrieved
December 27, 2002, from http://www.ipma.ch/site/certifica-
tion.htm
Jugdev, K. (2003). Developing and sustaining project manage-
ment as a strategic asset: A multiple case study using the resource-
based view. Unpublished doctoral dissertation, University of
Calgary.
Jugdev, K., & Thomas, J. (2002). Project management
maturity models: The silver bullets of competitive advantage.
Student paper award. Project Management Journal, 33(4), 4-14.
Kaplan, S., Schenkel, A., von Krogh, G., & Weber, C. (2001).
Knowledge-based theories of the firm in strategic management: A
review and extension (MIT Sloan Working Paper No. 4216-01).
Kerlinger, F. N. (1973). Foundations of behavioral research (2
ed. Vol. 1). New York: Holt, Rinehart & Winston.
Kloppenborg, T., & Opfer, W. (2002). The current state of
project management research: Trends, interpretations, and pre-
dictions. Project Management Journal, 33(2), 5-18.
Kogut, B., & Zander, U. (1992). Knowledge of the firm,
combinative capabilities, and the replication of technology. In
N. Foss (Ed.), Resources, firms, and strategies: A reader in the
resource-based perspective (1st ed., Vol. 1, pp. 306-325). Oxford,
UK: Oxford University Press.
Kogut, B., & Zander, U. (1993). Knowledge of the firm and
the evolutionary theory of the multinational corporation.
Journal of International Business Studies, 24(4), 625-644.
Levinthal, D. A., & Myatt, J. (1994). Co-evolution of capa-
bilities and industry: The evolution of mutual fund processing.
Strategic Management Journal, 15(1), 45-62.
Liedtka, J. M. (1996). Collaborating across lines of busi-
ness for competitive advantage. The Academy of Management
Executive, 10(2), 20-46.
Lopez, V. A. (2001). An overview review of the resource-
based view (RBV) of the firm, drawing on recent Spanish man-
agement research. Irish Journal of Management, 22(2), 105-120.
Lundvall, B. Å., & Johnson, B. (1994). The Learning
Economy. Journal of Industry Studies, 1(2), 23-42.
Madhok, A. (2002). Reassessing the fundamentals and
beyond: Ronald Coase, the transaction cost and resource-based
theories of the firm and the institutional structure of produc-
tion. Strategic Management Journal, 23(6), 535-550.
Marino, K. E. (1996). Developing consensus on firm com-
petencies and capabilities. The Academy of Management Executive,
10(3), 40-50.
Mason, J. (1998). Qualitative research (1st ed. Vol. 1).
London: Sage Publications, Inc.
Mata, F. J., Fuerst, W. L., & Barney, J. B. (1995). Information
technology and sustained competitive advantage: A resource-
based analysis. MIS Quarterly, 19(4), 487-507.
Maxwell, J. A. (1996). Qualitative research design: An interac-
tive approach (1 ed. Vol. 41). Thousand Oaks, CA: Sage
Publications, Inc.
Meredith, J. R. (2002, July). Developing project management
theory for managerial application: The view of a research journal’s
editor. Paper presented at the Frontiers of Project Management
Research and Application, Seattle, WA.
MicroFrame. (2001, October 1). Project management maturi-
ty model. Retrieved November 20, 2002, from
http://www.microframe.com/
Montealegre, R. (2002). A process model of capability
development: Lessons from the electronic commerce strategy at
Bolsa de Valores de Guayaquil. Organization Science, 13(5), 514-
531.
Morris, P. W. G. (2001a). Science, objective knowledge, and the
theory of project management. Unpublished manuscript, London.
Morris, P. W. G. (2001b). Updating the project manage-
ment bodies of knowledge. Project Management Journal, 32(3),
21-30.
Ngwenyama, O., & Nielsen, P. A. (2003). Competing values
in software process improvement: An assumption analysis of
CMM from an organizational culture perspective. IEEE
Transactions on Engineering Management, 50(1), 100-112.
Olszewski-Walker, & Coalson-Avant. (1995). Strategies for
theory construction in nursing (3rd ed. Vol. 1). Norwalk, CT:
Appleton & Lange.
Packendorff, J. (1995). Inquiring into the temporary organ-
ization: New directions for project management. Scandinavian
Journal of Management, 11(4), 319-333.
Patton, M. Q. (1990). Qualitative evaluation and research
methods (2nd ed.). Newbury Park, CA: Sage Publications, Inc.
Pennypacker, J. (2001). Project management maturity bench-
mark: A Center for Business Practices (CBP) research report
(Published manuscript). Havertown, PA: Project Management
Solutions, Inc.
Peteraf, M. (1993). The cornerstones of competitive advan-
tage: A resource-based view. In N. Foss (Ed.), Resources, firms,
and strategies: A reader in the resource-based perspective (1st ed.,
Vol. 1, pp. 187-203). Oxford, UK: Oxford University Press.
Porter, M. E. (1991). Towards a dynamic theory of strategy.
Strategic Management Journal, 12(5), 97-117.
Porter, M. E. (1996). What is strategy? Harvard Business
September 2004 Project Management Journal • 25
Review, 74(6), 61-78.
Powell, T. C. (2001). Competitive advantage: Logical and
philosophical considerations. Strategic Management Journal,
22(9), 875-888.
Prahalad, C. K., & Hamel, G. (1990). The core competence
of the corporation. In N. Foss (Ed.), Resources, firms, and strate-
gies: A reader in the resource-based perspective (1st ed., Vol. 1, pp.
235-256). Oxford, UK: Oxford University Press.
Priem, R. L., & Butler, J. E. (2001a). Is the resource-based
“view” a useful perspective for strategic management research?
Academy of Management Review, 26(1), 26-40.
Priem, R. L., & Butler, J. E. (2001b). Tautology in the
resource-based view and the implications of externally deter-
mined resource value: further comments. Academy of
Management Review, 26(1), 57-66.
Project Management Institute. (2000). A guide to the project
management body of knowledge (PMBOK® Guide) 2000 ed.
Newtown Square, PA: Project Management Institute.
Ray, G., Barney, J. B., & Muhanna, W. A. (2004).
Capabilities, business processes, and competitive advantage:
Choosing the dependent variable in empirical tests of the
resource-based view. Strategic Management Journal, 25(1), 23-37.
Reynolds, P. D. (1971). A primer in theory construction (1st
ed.). New York: MacMillan Publishing Company.
Rouse, M. J., & Daellenbach, U. S. (2002). More thinking
on research methods for the resource-based perspective.
Strategic Management Journal, 23(6), 963-967.
Rumelt, R. P., Schendel, D. E., & Teece, D. J. (1994).
Fundamental issues in strategy. In R. P. Rumelt, D. E. Schendel
& D. J. Teece (Eds.), Fundamental issues in strategy (pp. 9-47).
Cambridge, MA: Harvard Business School Press.
Schlichter, J. (2000). Personal communication. Newtown
Square, PA: Project Management Institute.
Schulze, W. S. (1994). Two resource-based theories of the
firm: Definitions and implications. In P. Schrivistava, A. Huff &
J. E. Dutton (Eds.), Advances in Strategic Management (Vol. 10,
pp. 127-151). New York: JAI Press.
Selznick, P. (1957). Leadership in administration: A socio-
logical interpretation. In N. Foss (Ed.), Resources, firms, and
strategies: A reader in the resource-based perspective (1st ed., Vol. 1,
pp. 21-26). Oxford, UK: Oxford University Press.
Shaw, M. L. G., & Gaines, B. R. (1995, November 24).
Comparing conceptual structures: Consensus, conflict, correspon-
dence, and contrast. Retrieved February, 2002, from
http://ksi.cpsc.ucalgary.ca/articles/KBS/COCO/
Skulmoski, G. (2001). Project maturity and competence
interface. Cost Engineering, 43(6), 11-19.
Sutton, R. I., & Staw, B. M. (1995). What theory is not.
Administrative Science Quarterly, 40(3), 371-384.
Tcha, M. (1997, November 15). Externality and pro-
tection: Online international trade policy course.
Retrieved November 20, 2002, from
http://www.firststep.com.au/econ/plain/externalities.html
Teece, D., J., Pisano, G., & Shuen, A. (1997). Dynamic capa-
bilities and strategic management. Strategic Management Journal,
18(7), 509-533.
Teece, D. J. (1980). Economies of scope and the scope of
the enterprise. In N. Foss (Ed.), Resources, firms, and strategies: A
reader in the resource-based perspective (Vol. 1, pp. 103-116).
Oxford, UK: Oxford University Press.
Thomas, H., Pollock, T., & Gorman, P. (1999). Global
strategic analysis: Frameworks and approaches. The Academy of
Management Executive, 13(1), 70-82.
Tsoukas, H. (1991). The missing link: A transformational
view of metaphors in organizational science. Academy of
Management Review, 16(3), 566-585.
Ulri, B., & Ulri, D. (2000). Project management in North
America: Stability of the concepts. Project Management Journal,
31(3), 33-43.
Verzuh, E. (1999). The fast forward MBA in project manage-
ment. New York: John Wiley & Sons, Inc.
Wacquant, L., J. D. (1993). Positivism. In W. Outhwaite, T.
Bottomore & W. Cuthwaite (Eds.), The Blackwell dictionary of
twentieth century social thought. Oxford, UK: Blackwell
Publishers, Ltd.
Weick, K. E. (1995). What theory is not, theorizing is.
Administrative Science Quarterly, 40(3), 385-390.
Wiggins, R. R., & Ruefli, T. W. (2002). Sustained competi-
tive advantage: Temporal dynamics and the incidence and per-
sistence of superior economic performance. Organization
Science, 13(1), 82-104.

DR. KAM JUGDEV, Assistant Professor of Project Management and Strategy in the MBA program at
Athabasca University in Alberta, has extensive university teaching experience in online and traditional
formats. She has over 12 years of experience as a senior project manager in public and private sector
organizations. Kam’s current areas of research include project management as a source of competitive
advantage, and the Resource-Based View of the Firm as it applies to project management.

Kam holds a PhD in Project Management (Engineering) from the University of Calgary, a Master of
Engineering in Project Management (Civil Engineering, Calgary), a Master of Health Services Administration
(Alberta) and two undergraduate science degrees. She has over 25 publications, some of which are in PM
Network, Project Management Journal, and the International Journal of Project Management. Kam is also a
reviewer for the International Journal of Project Management. As a member of the Project Management
Institute, Academy of Management, Strategic Management Society, Administrative Sciences Association of
Canada, and the Western Academy of Management, Kam actively contributes to the advancement of
academic and professional communities of management practice.

26 • Project Management Journal September 2004

THE IMPACT OF OWNER REPRESENTATIVES IN A
DESIGN-BUILD CONSTRUCTION ENVIRONMENT
ABSTRACT
This research asserts that owner repre-
sentatives have a positive impact on bal-
ancing project constraints such as time,
cost, quality, and scope in a design-build
construction environment. It further
asserts that soft skills are the most
important skills for owner representa-
tives on design-build projects. This
research focuses on a specific owner rep-
resentative, the U.S. Army Health Facility
Planning Agency, to determine if its proj-
ect integration concept has a significant
impact in balancing constraints on Army
healthcare renewal projects. It also illus-
trates the importance of soft skills for
project integrators, as they execute
owner representative duties in a design-
build environment.
Keywords: project management; project
integrator; project integration concept;
project constraints; time; cost; quality;
scope; design-build construction; renewal
program; soft skills; owner representative
©2004 by the Project Management Institute
Vol. 35, No. 3, 27-38, ISSN 8756-9728/03
DENIS R. PETERSEN, PMP, Milestone Management Consultants LLC, 8030 South 2350 East,
Ogden, Utah 84405
E. LILE MURPHREE JR., PE, PhD, The George Washington University, 1776 G Street, NW, Suite 110,
Washington, DC 20052
THE PROBLEM AND ITS SETTING
Introduction
ne of the most difficult yet thrilling aspects of project management in a
O design-build construction environment is trying to consistently and suc-
cessfully balance project constraints to produce project success. This
seemingly simple concept of balancing time, cost, quality, and scope has been
either the savior or downfall for many design-build project managers. In fact, this
delicate balance affects project managers of all types, including a special breed
called owner representatives. While some may believe that owner representatives
have a minimal impact on project performance and balancing project constraints,
the authors believe the influence of owner representatives on the success or fail-
ure of a design-build project is significant. They also believe owner representatives
must excel in a specific type of integration and management skills—soft or people
skills—to positively contribute to the production of success on design-build proj-
ects.
The Setting
The U.S. Army Health Facility Planning Agency (HFPA) is an organization that is
learning the importance of balancing project constraints as an owner representa-
tive in a modern design-build environment. HFPA was originally formed because
the Army Medical Command (MEDCOM) was not satisfied with the consistency
of the U.S. Army Corps of Engineers (USACE) in successfully delivering quality,
high-tech, Army medical facilities on time and under the programmed budget
(Kurmel, 2000). Since its creation, HFPA has partnered with USACE in develop-
ing a massive inventory of Army medical facilities. Most of the facilities built dur-
ing the first 20 years of this partnership were built through the Military
Construction (MILCON) program as new, inpatient-oriented facilities. The
responsibilities of HFPA on these projects leaned solely toward the owner repre-
sentative skill set, with separate managers focusing on each phase of the project
life cycle. USACE was responsible for handling most of the direct project admin-
istration and contract management duties on these projects.
The past 5 to 10 years have produced major changes in the modes of health
care delivery and the associated requirements for health facilities. The paradigm
for health care has shifted from inpatient-oriented care, to a focus on wellness
and outpatient care. This paradigm shift has had an overwhelming effect on the
September 2004 Project Management Journal • 27
use and functionality of health facili-
ties throughout the world. Facilities
originally designed to hold patients in
bed space for extended periods of time
are now being rapidly converted into
outpatient clinic space. As one can
imagine, inpatient bed space does not
easily convert into efficient outpatient
clinic space without major functional,
architectural, mechanical, electrical,
and plumbing modifications. This
trend has had a significant impact on
Army medical facilities and on HFPA
as it seeks to modernize MEDCOM
facilities to meet the requirements of
current business practices.
The MILCON program, which was
instrumental in developing much of
the original inpatient health care infra-
structure, is a very rigorous design-bid-
build process and is steeped in
bureaucracy. It is not uncommon for
MILCON projects to take as much as 7
to 10 years to go from programming
through completion of construction.
With such a rapid swing in methods of
health care delivery, the MILCON
process has not been able to keep up
with the time and quality demands for
modifying facilities to meet business
requirements, thus throwing projects
out of balance with respect to quality
and time constraints on MILCON
healthcare modernization projects.
As HFPA has worked to improve
the processes associated with the deliv-
ery of medical facilities, it have seen its
role expand to become the action arm
of the Medical Command and the
Surgeon General in managing the
entire capital investment program for
Army health facilities. This role has
forced the agency to explore many pos-
sibilities for delivering medical facili-
ties within the new paradigm of
wellness and outpatient care. Its explo-
ration has led it into the design-build
arena as a means of meeting the busi-
ness and facility needs of MEDCOM,
and of achieving more control over
quality and time constraints.
One of HFPA’s new methods of
delivering health care projects is the
renewal program. The renewal pro-
gram is based on the design-build con-
cept, which has much shorter turn
around times and is a critical vehicle
28 • Project Management Journal September 2004
for meeting the business needs of
Army health care providers. Through
this program, areas of a facility are
quickly renovated to assist health care
providers in modernizing their opera-
tions and in repairing an aging infra-
structure. This program is still done in
partnership with USACE. However,
one of the differences between this and
the MILCON program is that USACE
only handles the contracting aspect of
renewal projects. HFPA, in its expand-
ed role, now provides both owner rep-
resentation and project management
expertise through each phase of the
project life cycle, utilizing staff mem-
bers given the title project integrator.
HFPA has provided varying levels
of owner representation and project
management services on behalf of the
Army Surgeon General throughout its
existence. However, the project integra-
tion concept, in conjunction with the
renewal (design-build) program, takes
the responsibilities of HFPA managers
to a new level. Project integrators are
now charged with managing a pro-
gram of projects through every aspect
of the project’s life cycle. Their respon-
sibilities now involve moving a renew-
al project seamlessly from the master
planning and concept stages all the
way through design, construction,
transition, and post-occupancy
processes. They now ensure proper
integration of the project at every level
and through every key milestone.
Furthermore, they are charged with
properly integrating each project with-
in the total MEDCOM capital invest-
ment program to ensure proper
justification, prioritization, authoriza-
tion, funding, and execution through
the project’s life cycle. Given this new
integration concept and their expand-
ed role in design-build project delivery,
it is imperative that HFPA project inte-
grators develop skills and techniques
which positively assist the renewal
project teams in providing seamless
project delivery and a better balance of
project constraints on design-build
projects.
The Problem Statement
Project integration and the renewal
program are cutting-edge concepts
among government agencies, with
HFPA blazing the trail. This agency is
forging partnerships and developing
unprecedented contracts through
USACE, with a focus on delivering
quality, cost-effective, and timely facil-
ity solutions. However, as with any
new concept there are inherent con-
cerns. There are questions as to whether
the project integration concept positive-
ly assists in balancing the demands of
project constraints. Furthermore, HFPA
project integrators find that many of the
project management principles and
practices applied in the traditional
design-bid-build (MILCON) scenario
do not necessarily apply on medical
projects in a design-build arena. They
also find that the project integration
concept involves new leadership and
management processes with which they
are unfamiliar.
This research attempts to quantify
the impact the project integration con-
cept has in balancing project con-
straints and producing success on
Army healthcare renewal projects. It
also attempts to determine which proj-
ect integration and management tech-
niques best help integrators achieve
success.
The Subproblems
• The first subproblem is to quan-
tify the impact the project inte-
gration concept has in balancing
project constraints and produc-
ing success on Army health care
renewal (design-build) projects.
• The second subproblem is to
determine which project integra-
tion and management tech-
niques are most important for
HFPA project integrators as they
seek to balance project con-
straints and produce success on
Army healthcare renewal
(design-build) projects.
The Hypotheses
• The first hypothesis is that the
project integration concept has a
significant, positive impact in
balancing project constraints
and producing success on Army
health care renewal (design-
build) projects.
 • The second hypothesis is that
soft or people skills are the most
important project integration
and management techniques for
HFPA project integrators as they
seek to balance project con-
straints and produce success on
Army healthcare renewal
(design-build) projects.
The Delimitations
For the purposes of this research pro-
posal, successfully balanced project
constraints have the following charac-
teristics or results:
• Project and product scope are
managed to the satisfaction of
primary stakeholders
• Project is completed within a
schedule agreed upon by pri-
mary stakeholders
• Project is completed within a
budget agreed upon by primary
stakeholders
• Product at completion meets a
level of quality agreed upon by
primary stakeholders, desired by
customers, and required by
codes, standards and regulations
• Project is completed with mini-
mal operational impact to all
primary stakeholders.
The study does not attempt to
delineate or predict any level of success
in balancing project constraints or cre-
ating project success beyond the defi-
nition posed above.
The study focuses solely on project
management principles and practices
applied by HFPA project integrators
working in the Army health care
renewal (design-build) program.
The study does not include
information associated with the
Army Medical MILCON program, or
any other military construction pro-
gram outside the healthcare renewal
program.
The study does not determine
which principles and practices can
be extrapolated to design-build proj-
ect management and owner repre-
sentation in the civilian sector,
although this information will most
likely be applicable to management
in that sector.
The Importance of the Study
The U.S. Army Health Facility
Planning Agency is entrusted with a
direct fiscal and moral responsibility
to provide world-class health care facil-
ities on behalf of taxpayers, govern-
ment officials, military beneficiaries,
and hospital staff members. Its mis-
sion involves “Providing readiness
support and world class facilities to
enhance the accountable delivery of
seamless, quality, health care in sup-
port of the Nation’s war-fighters,
worldwide” (USAHFPA, 1997, p. 1).
The agency’s ability to deliver timely,
cost-effective, quality solutions in car-
rying out its duties is what determines
organizational success or failure and
allows it to continue operating in the
military health facilities arena.
Project integrators are key assets to
this agency in its pursuit of mission
fulfillment. One of their specific roles
is to provide “seamless” delivery of
renewal projects. In this role, they are
becoming HFPA’s front-line interface
with project stakeholders for all health
care renewal (design-build) projects.
Their success or failure is a direct
reflection on the agency in the fulfill-
ment of its stated mission. In essence,
their success is critical to the current
and future viability of the U.S. Army
Health Facility Planning Agency.
Therefore, this study may be critically
important to the furtherance of the
mission, values, and goals of the U.S.
Army Health Facility Planning Agency.
While this research is limited to
the constraints outlined above, the
authors believe the results are applica-
ble to a wide range of government and
non-government organizations that
work within the design-build construc-
tion arena. In fact, the authors believe
the principles illuminated by this
research could have far-reaching effects
on the practices applied by project
managers on design-build projects in
all sectors of the construction industry.
REVIEW OF RELATED LITERATURE
Project Management –
A Historical Basis
Modern project management is a rel-
atively young profession with a
bright future. It has experienced rapid
September 2004 Project Management Journal • 29
evolution from its conceptual roots
as a set of skills to its emergence as a
legitimate profession (Coates and
Jarrett, Inc., 1999, p. 8). This evolu-
tion is expected to continue as
change becomes the norm in our
society, and as the need for “a profes-
sion that essentially manages
change” (PMI, 1999, p. xi) becomes
even more crucial.
The roots of project management
are found in the works of visionaries
such as Frederick Taylor and Henry
Gantt. These pioneers are responsible
for the development of skills that later
became the foundation of a profes-
sion. Taylor was generally acknowl-
edged as the “Father of Scientific
Management.” His theory of scientific
management and his studies of work
showed that the productivity of labor
can be measured, analyzed, and
improved by focusing on its elemen-
tary parts (Drucker, 2001, pp. 6-7).
Gantt, a contemporary of Taylor, stud-
ied the sequence and duration of tasks
in work processes. He is the creator of
the Gantt chart, which uses task bars
and milestones to synthesize project
activities, sequences, durations, and
resources into a coherent project
schedule (Kerzner, 2001).
The skills and processes developed
by Taylor, Gantt, and many others, are
believed to be the impetus behind the
development of modern processes for
managing projects. Many believe that
modern project management was born
between the 1940s and 1960s, when
massively engineered, complex mili-
tary and government projects prompt-
ed managers to develop management
techniques such as the Program
Evaluation and Review Technique
(PERT) and the Critical Path Method
(CPM). These techniques and others
assisted project managers with plan-
ning and control aspects of complex
projects. (Van der Merwe, 1998).
In his text Project Management: A
Systems Approach to Planning,
Scheduling, and Controlling, Kerzner
provides an excellent outline of the
evolution of project management
between the years 1960 and 2000. He
illustrates how slow growth and lack of
acceptance of project management
techniques in many industries domi-
nated the first 25 years of this period.
He then shows how the business need
for project management experienced
exponential growth over the final 15
years of this time period. This growth
was primarily due to executives recog-
nizing the implementation of project
management as a necessity, driven by
variables such as complex capital proj-
ects, competitiveness, effectiveness,
efficiency, and customer expectations.
Kerzner’s outline also highlights the
development of skills and methodolo-
gies during the past 40 years, such as
Triple Constraint Management, Total
Quality Management, Risk
Management, and Concurrent
Engineering, which have led to the
recognition of project management as
a professional career path (Kerzner,
2001).
The evolution of project manage-
ment continues today. Wideman
(1996, p. 1) states, “In the last couple
of decades, project management has
emerged as a business process tool
with broad application in the corpo-
rate world. It is seen as the manage-
ment approach of choice for dealing
with an ever-shifting business environ-
ment.…We have a much improved
understanding of project management
tools and techniques, and this decade
has focused on the importance of the
behavioral and organizational aspects
of projects.” Many authors concur with
Wideman’s view that we have devel-
oped a good understanding of the
technical tools of the trade, while the
past few years have illuminated the
importance of the behavioral, or soft,
skills of project management.
The recent focus on the behav-
ioral, or soft skills of project manage-
ment is a realization by project
management professionals that people
and teams perform the work of proj-
ects. This realization has spawned
much thought from authors that lead-
ership is a significant aspect of project
management. Covey (1989) states,
“You manage things, you lead people.”
Drucker (2001, p. 81) echoes this
thought, “One does not ‘manage’ peo-
ple. The task is to lead people. And the
goal is to make productive the specific
30 • Project Management Journal September 2004
strengths and knowledge of each indi-
vidual.” Eisner (1997, pp. 95-103) also
appears to agree with this philosophy
as he devotes an entire chapter in his
text to “The Project Manager and
Leadership.” In this chapter, Eisner
lists 20 attributes of a successful proj-
ect manager. Fifteen of these attributes
deal specifically with leading people.
The balance of these attributes address
personal management skills, which
ultimately impact a manager’s ability
to lead project team members.
Along with the need for leader-
ship skills, the Project Management
Institute (PMI) recognizes the need
for integration across all aspects of a
project life cycle. The first of nine pub-
lished Knowledge Areas in PMI’s A
Guide to the Project Management Body
of Knowledge focuses on “Integration
Management.” This is the concept
that a project manager facilitates the
seamless delivery of a project from its
planning and development stages,
through project execution and com-
pletion, while also managing the
scope and integrating change (PMI,
2000, pp. 41-49).
Kerzner (2001), Coates & Jarret,
Inc., (1999), Haldane (1998), Pells
(1998), Foti (2001), and many others
see globalization, change and multina-
tional project teams as the norm for
projects in the future. They also seem
to agree that the need for leadership,
team building, and integration skills
among project managers is becoming
essential to successful project execu-
tion in this environment of global
change. Thus, as the profession of proj-
ect management continues its evolu-
tion, it appears that leadership and
integration are becoming much more
important to project success - perhaps
even more important than the tools
and techniques which defined project
management in the past.
Design-Build Project Management
Design-build “…is a method of proj-
ect delivery, which at its core posits a
new relationship between owner and
builder” (Belle, 2002, p. 1). It does so
by placing responsibility on a single
contractor to facilitate project com-
pletion all the way from initial con-
cept stages through design, construc-
tion and sometimes transition, oper-
ation, and maintenance of the
facility. Bearing the responsibility for
all of these processes places more risk
on a contractor and heightens the
need for innovation, integration, and
relationships with the owner in seek-
ing to meet project performance
requirements.
Design-build is not a new concept,
but like project management its use
has seen exponential growth over the
past few years. It is rapidly becoming
the construction management method
of choice for markets and projects that
require timely, cost-effective, quality,
project delivery. More than ever, the
norm within both public and private
markets is change. Belle (2002, p. 1)
states, “There is greater receptivity than
ever to systems that can respond quick-
ly and effectively to new market trends.
Flexible, single-source delivery systems
such as design-build are gaining
increased market share precisely
because they are best situated to
respond to these emerging trends.”
One of the inherent strengths of
design-build is the ability of the con-
tractor to shorten time frames and
costs by running the design in parallel
with construction. This requires
involvement of the owner for deci-
sions, providing them with more flexi-
bility to make changes during each
phase of the project. (Oberlander
2000). It also provides the contractor
with more flexibility in “real-time” to
attempt new delivery methods, try new
materials, implement new technolo-
gies and better meet the performance
requirements of the owner (USACE,
1994, p. B-1). Many traditional con-
struction delivery processes are with-
out this advantage.
Design-build began to take hold
as a single-source construction delivery
system in the United States in the
1970’s; since then its popularity has
grown tremendously. The Design-
Build Institute of America (DBIA) pre-
dicts that design-build will overtake
traditional design-bid-build delivery
methods in the construction industry
by the year 2005 (HBE, 2002). The
DBIA itself was recently formed, in
1993, “in response to the emergence of
design-build project delivery as a sig-
nificant force in the design and con-
struction field.” (DBIA, 2002). Its
emergence as a “significant force”
came about because owners saw it as
an integrated and innovative process
that created a relationship, or partner-
ship, with a single project delivery
source. Halpin and Woodhead (1998,
p. 72) believe this relationship is a
“natural evolutionary step beyond the
negotiated contract.”
Many private organizations agree
that integration, innovation, and pre-
ferred partnerships, naturally created
by design-build delivery processes, are
the foundation for future success in
the construction marketplace. Tarpey
(2000) believes that successful client-
provider relationships of the future
will be achieved through processes
such as design-build, because they
match client expectations with
provider competencies, creating cer-
tainty in relationships and a “partner
of choice” environment. Hoffman, as
quoted by Tulacz (2001, p. 49), agrees
that remaining competitive in today’s
changing environment requires
design-build firms to become integra-
tors by “…understand[ing] the
owner’s business and provid[ing]
advice on everything from financing
options to needs analysis to regulatory
compliance.…”
Public agencies are turning to
design-build because it provides an
avenue through which they can break
from the bureaucracy of traditional
project delivery means. It also provides
innovative, integrative, cost-effective,
and timely methods of meeting their
asset management and capital invest-
ment needs. Goode (2000) illustrates
that innovation and integration are
becoming a requirement in many gov-
ernment projects. Public agencies are
looking for single source contractors
that not only provide design and con-
struction services, but those who will
participate in programming, planning,
operations, and maintenance activi-
ties. Government agencies also like
design-build because it moves more of
the risk from the agency to a single
contractor, “…sharpen[ing] attitudes
to risks, methods, time, cost, quality,
and warranties” (Jones, 1984, preface).
Thus, public agencies are embracing
the design-build process because it
inherently provides incentives to the
contractor for innovatively meeting
performance requirements and effec-
tively balancing time, cost, and quality
parameters.
The future appears to be unlimited
for the design-build project delivery
method in both public and private sec-
tors. Literature on this subject is grow-
ing at a rapid pace. The U.S.
Government Federal Acquisition
Regulations, Part 36.104, and Title 10
United States Code section 2862 have
been rewritten to include design-build
as an authorized delivery method for
government projects (USACE, 1999).
Major public events such as the Winter
Olympics in Salt Lake City have high-
lighted advantages of design-build
delivery methods (Cho & Sawyer,
2001, p. 13). Projects such as a $90
million state-owned hospital in
Kazakhstan (Reina, 2001) and a $14
billion high-speed rail project in
Taiwan (McManamy, 2001) are part of
an international surge in design-build
delivery methods. And finally, design-
build has been hailed in the wake of
terrorist attacks at the Pentagon as the
“secret weapon” for managing con-
struction in a crisis situation (Tanner,
2001). Flexibility, innovation, integra-
tion, and single-source delivery are
keys to successful project delivery in
future changing environments; design-
build project delivery methods appear
to offer all of these attributes.
Owner Representation in a Design-
Build Construction Environment
There are hundreds, perhaps thou-
sands, of construction management
companies offering owner representa-
tive services for design-build construc-
tion contracts in the United States
today. There are also many public enti-
ties that act as construction agents or
owner representatives for other govern-
ment agencies. One needs only to type
“owner representatives in design-build
construction” into a Web browser to
view the vast opportunities available
for owners in selecting an appropriate
agency to represent their facility needs.
In contrast to these vast opportunities
is a significant void in information
quantifying the impact owner repre-
sentatives have on design-build
processes. There is also a significant
lack of information regarding the spe-
cific skills needed by owner representa-
tives to assist in the production of a
successful design-build project. Of the
information available regarding owner
representation in design-build con-
struction, much of it appears to agree
that owner involvement is key to suc-
cessful project completion.
One example showing the need
for owner involvement in design-build
construction is illustrated by a study at
Penn State University. This study used
data from 351 U.S. building projects to
show that design-build projects proved
to be the most efficient of the current
construction project delivery methods
- with a caution that, in order for this
to be true, owners must think and act
differently than they have in past
design-bid-build environments. From
this study, Sanvido and Konchar
(1999) conclude that design-build
projects require owners to forge a
team-centered relationship based on
qualifications and best value to suc-
ceed. Since owners must be involved, it
goes without saying that their project
representatives must have the same
abilities in their approach to managing
a project.
Another example of the need for
owner involvement in design-build
projects is outlined in the opinions of
numerous construction professionals
in New York. In a report of the pro-
ceedings of a meeting of the Greater
New York Construction User Council,
members of the Design-Build Institute
of America, New York State American
Institute of Architects, General
Contractors Association of New York,
and many other professional and legal
authorities discussed “The Ten Most
Deadly Design/Build Construction
Sins.” In their discussion, they deter-
mined that two of these “construction
sins” were an owner’s “failure to
assemble the right project team” and
“failure to coordinate the project
team.” The consensus for choosing
September 2004 Project Management Journal • 31
these as problems came from the per-
ceived tendency of an owner to “walk
away” from the process once a design-
build contractor has been selected
(Osborn, 2000, p. 26). In a discussion
apparently produced by caution
toward the design-build process, this
group appears to firmly advocate the
need for owner involvement through-
out the design-build process in order
to provide a complete and successful
project. This need again translates to a
requirement for the owner representa-
tive to be very involved and to have the
skills necessary to assemble and coor-
dinate the project team in order to pro-
duce success.
Public agencies also appear to
agree that owner representatives are
key to design-build project success.
The U.S. Army Corps of Engineers acts
as an owner representative for multiple
government agencies in executing
design-build construction contracts.
They certainly see the value of owner
representation in this process, as they
put together an entire “Design-Build
Management Team” (USACE, 1994,
pp. 2-3) composed of USACE man-
agers and employees for each design-
build project. This team works
hand-in-hand with each of the con-
tractor trades throughout acquisition,
design, and construction processes
with the objective of producing suc-
cess. Many of the state transportation
departments also see the value of
owner representation on public works
design-build projects. In its recently
published Guidebook for Design-Build
Highway Project Development, the
Washington State Department of
Transportation recognizes its responsi-
bility in the design-build process to
“…make it work satisfactorily for all
parties involved, including WSDOT,
design firms, construction contractors,
bonding and insurance companies,
local jurisdictions, and other affected
state and federal agencies” (WSDOT,
2001, p. 10). It appears that this organ-
ization, which acts as owner represen-
tative on design-build highway
projects, believes that its responsibility
involves creating an environment in
which the entire design-build team can
be successful. Thus, it perceives its
32 • Project Management Journal September 2004
involvement in the process is impor-
tant to its success.
While all of the publications listed
above promote the importance of
owners and owner representatives to
the success of design-build project
delivery, none of them attempts to
quantify the effectiveness and influ-
ence of owner representation on
design-build processes. Furthermore,
the authors had a very difficult time
finding information regarding specific
skills required of owner representatives
in building the design-build team and
creating an environment in which a
successful balance of the project con-
straint condition can be achieved.
Project Integration from an Owner
Representative Perspective
While much has been written regard-
ing project management, design-build
delivery, and the need for innovation
and integration, few seem to discuss
the importance of project integration
from an owner representative stand-
point in a design-build construction
environment. In fact, the notion of
project integration, or integrative man-
agement, is a concept the literature
associates primarily with executing
projects in a systems engineering and
research environment, rather than that
of design-build construction.
As mentioned earlier, the notion
of project integration, or integrative
management, is a concept primarily
associated with the systems engineer-
ing and research community. NASA, a
highly technical-systems and engineer-
ing-based organization, preaches the
importance of integration and verifica-
tion as a key component of their proj-
ect cycle “Vee Chart” (NASA, 1995, p.
20). The United States Department of
Defense specializes in the acquisition
of high-tech weapons systems, and is
therefore a keen advocate of the
importance of integrative management
in many of its new acquisition require-
ments (DODI, 2001; see also DOD,
2001). The ISO 9000 series, produced
by the International Organization for
Standardization, is based on funda-
mental principles requiring the inte-
gration and alignment of quality
processes through all aspects of prod-
uct and project life cycle (ISO, 2000).
Total Quality Management,
Configuration Management, and
Quality Assurance are all systems engi-
neering concepts spawned by a desire
to integrate different aspects of a sys-
tem and project in order to produce a
higher quality product. These exam-
ples are truly indicative of the impor-
tance of integration and integrative
management in the systems engineer-
ing community. Yet still, the literature
rarely crosses over to illustrate how
these principles might apply in a proj-
ect management environment in
which design-build construction and
owner representation are discussed in
conjunction with the concept of proj-
ect integration.
One of the first texts to create an
excellent bridge between systems engi-
neering and project management con-
cepts is Eisner’s Essentials of Project and
Systems Engineering Management.
Within this text, Eisner illustrates the
importance of integration and integra-
tive management for both the project
management and systems engineering
communities. He defines integrative
management as “…a set of practices
whereby people, processes, tools, and
systems are brought together into har-
monious interoperation so as to maxi-
mize their efficiency and
effectiveness.” (Eisner, 1997, p. 312).
He also states that managers operating
as integrators must “…pay attention to
the interactions between people, the
systems they are in, and the systems
they are trying to build” (Eisner, 1997,
p. 313). These are excellent definitions
of integrative management; they are
very applicable to both project and sys-
tems engineering environments.
Furthermore, they are very applicable
to project management in a design-
build environment. Eisner finishes this
discussion by outlining how teams,
plans, the systems approach, methods,
standards, information systems, and
enterprises can all act as integrators on
a project (Eisner, 1997).
During approximately the same
period that Eisner was writing and
publishing his text, the Project
Management Institute (PMI) was in
the process of editing and publishing
their 1996 version of A Guide to the
Project Management Body of Knowledge.
This text also recognizes the impor-
tance of project integration, as it
names Project Integration
Management as one of the nine Project
Management Knowledge Areas and
devotes an entire chapter to this topic.
This chapter stresses the importance of
Project Plan Development, Project
Plan Execution, and Overall Change
Control as the primary integrative
management processes (PMI, 1996).
In 2000, PMI released an updated ver-
sion of this standard. The chapter on
integrative management is still fairly
similar to the 1996 version, with some
minor changes to integration tech-
niques and a title change for one of the
sections from Overall Change Control
to Integrated Change Control (PMI,
2000).
The authors recommend both of
these texts as two of the very best in the
fields of project and systems engineer-
ing management; however, one con-
cept noticeably missing from both
texts is how the owner and owner rep-
resentative fit into the project integra-
tion equation. Understandably, these
texts are written from a performing
organization perspective. Furthermore,
both texts address the importance of
the owner as one of the primary stake-
holders in a project. But during the dis-
cussion of project integration
management, neither the owners nor
their representatives are mentioned as
a key aspect of integrative manage-
ment. The authors find that this omis-
sion occurs in many texts that discuss
integrative management. Yet, as shown
in earlier discussion, there appears to
be a significant need to include owners
and their representatives as a key part
of the integrative management team,
particularly in a design-build construc-
tion environment.
HFPA – Owner Representatives and
Project Integrators
The U.S. Army Health Facility Planning
Agency was formed as a key field- oper-
ating agency of the Office of The
Surgeon General of the Army in 1975
(OTSG, 1975). At the time of its organ-
ization, its mission was “…to execute
The Surgeon General’s responsibilities
for health facility programming, plan-
ning, design, and construction…”
(USAHFPA, 1975, p. 1). Given this
mission, the commander of the agency
was designated as the “…program
manager of the Army health facilities
construction program” (DA, 1988, pp.
3-4). In effect, the owner, The Surgeon
General of the Army, designated this
agency as his representative for the
health facility construction (MILCON)
program. Hence, the role in the early
stages was based around the MILCON
or traditional design-bid-build con-
struction program.
As stated earlier, the mission of
HFPA has been greatly expanded to
include the role of executing the entire
MEDCOM capital investment pro-
gram, including the renewal (design-
build) program. Its responsibilities
now involve moving a renewal project
seamlessly from the master planning
and concept stages, all the way through
design, construction, transition, and
post-occupancy processes, ensuring
proper integration of the project at
every level and through every key mile-
stone (Chong, 2001).Furthermore, it is
charged with properly integrating each
project within the total MEDCOM cap-
ital investment program to ensure
proper justification, prioritization,
authorization, funding, and execution
throughout its life cycle (Pauli, 2002).
In light of its new role in the
renewal (design-build) arena, this
agency recognizes the need for man-
agers that can integrate projects at
every level and at key decision points
during a project’s life cycle. As stated
earlier in the document, HFPA calls
these managers project integrators. The
project integration concept in its cur-
rent form creates an entity to represent
the owner (The Surgeon General of the
Army), in balancing project constraints
and producing success on renewal
(design-build) projects. This concept
requires HFPA project integrators to
understand the delicate balance
among project constraints and to
determine how best to apply project
integration and management tech-
niques to help the project teams
achieve this balance.
September 2004 Project Management Journal • 33
While the literature regarding
HFPA and the project integration con-
cept is very limited, it will most likely
increase over the next few years. The
efforts of HFPA continue to gain the
attention of many in the government
facilities procurement arena. In fact,
HFPA has been recognized by the U.S.
General Accounting Office as a sole
“…Defense Department entity…using
life-cycle principles…to optimize
investments in operations, mainte-
nance, repairs, and minor construc-
tion” (SFWMD, 2001, p. 8). The
project integration concept plays a sig-
nificant role in the facility and project
life cycle management processes HFPA
employs. Furthermore, the HFPA
Project Integration Division is piloting
many facility procurement strategies
and relationships with government
agencies and design-build contractors.
The division’s efforts appear to place it
at the forefront of government agen-
cies with respect to owner representa-
tion and project integration in a
design-build environment.
The project integration concept
and the renewal (design-build) pro-
gram adopted by HFPA have roots in
project management, design-build
project delivery, systems engineering,
project life-cycle management and
owner representation disciplines. This
literature review has outlined the
impact of these philosophies in detail,
illuminating key concepts that have
played, or will play, an important role
in the development of the HFPA proj-
ect integration concept. This develop-
ment will continue as USAHPFA
project integrators better understand
the impact they have on balancing
project constraints and learn to apply
project integration and management
techniques that help promote this bal-
ance to produce success.
RESEARCH METHODOLOGY
AND RESULTS
General Approach
The authors looked at many different
methods of gathering this informa-
tion, such as interviews, on-site case
studies, and anonymous question-
naires. However, the expert population
working with HFPA project integrators
is spread throughout the world. Given
this broad dispersion of the intended
expert population, the authors used
the most practical avenue, an anony-
mous questionnaire distributed via e-
mail, to elicit responses from this
broad-based expert population.
The target population for this
study was not only dispersed widely
across the globe, but also varied
according to organization and posi-
tion. Potential respondents included
design-build construction contractors,
architecture firms, planning firms,
USACE offices, HFPA staff members
and other government employees.
Given this diverse grouping of respon-
dents, the authors recognized the need
to employ screening or evaluation cri-
teria for acceptance of responses from
the target population.
Three primary criteria were estab-
lished for acceptance of responses in
testing the first hypothesis.
Respondents must have worked direct-
ly with an HFPA project integrator dur-
ing the past year, they must have had at
least two years’ experience working
with health care construction manage-
ment, and they must have had a mini-
mum of two years’ experience in the
design-build construction arena.
Only one restriction was
employed for responses for testing the
second hypothesis. Each respondent
must have worked directly with an
HFPA project integrator during the
past year. The authors felt it important
to include the responses of some of the
less-experienced respondents, since
project integrators have worked exten-
sively with many of these individuals
during the past couple of years to help
them understand the renewal pro-
gram. The authors believed these inter-
actions provided enough information
to less experienced respondents to
make a reasonable determination
regarding skills that are important for
HFPA project integrators.
The survey instrument was
designed in close coordination with
the Commander of the U.S. Army
Health Facility Planning Agency. It was
crafted to be easy to use, simple in pur-
pose, and clearly explained. Every
attempt was made to use common
34 • Project Management Journal September 2004
terms, phrases, and tones that would
elicit proper responses and avoid an
introduction of bias. The authors also
provided a “Definitions” sheet for
instances in which acronyms and other
terms used may have been unfamiliar
to respondents.
Data Collection and Disposition
Meeting the research goals outlined
above required a strategy to gather,
record, store, manipulate, and analyze
the data produced by the survey
instrument. The authors utilized the e-
mail system to gather data. They used
common database and spreadsheet
programs to record, manipulate, and
store the data. Finally, they imple-
mented the Bradley and Terry (1952)
method of paired comparisons to ana-
lyze the data.
As mentioned earlier, the authors
felt a questionnaire transmitted via e-
mail to potential respondents world-
wide was the best method of gathering
information. Thus, they developed an
e-mail message to provide directions
to potential respondents and attached
an information sheet and the ques-
tionnaire to the e-mail message. This
message was distributed worldwide via
e-mail to many locations in the conti-
nental United States, Alaska, Hawaii,
Korea, Japan, Germany, and elsewhere.
The questionnaire was distributed to a
very high percentage of the target pop-
ulation. E-mail not only provided an
excellent medium for transmitting and
receiving the questionnaire, but it also
provided a method of improving the
overall response rate. Gentle reminders
sent via e-mail significantly increased
the number of responses. The ques-
tionnaire was sent to approximately
200 potential respondents. The
authors received and processed 102
total responses, for a response rate bet-
ter than 50 percent.
The questions outlined in the sur-
vey instrument were designed to pro-
duce raw, ordinal data as respondents
selected an appropriate, “Likert-type”
(Cooper & Emory, 1995, p. 180) scaled
answer consistent with their opinion
of the project integration concept. The
Bradley and Terry (1952) method of
paired comparisons was used for deter-
mining statistical significance and sto-
chastic, or probabilistic, dominance
between the attributes or skills
addressed in each question. This
method also provided an excellent
avenue for rank ordering attributes or
skills with each other.
Key Assumptions
Four assumptions were made that were
important to the statistical analysis in
this research. These assumptions dealt
with survey techniques, combination
and comparison of results, and signifi-
cance levels for statistical tests.
The first assumption involved the
survey process. In developing the sur-
vey instrument, the authors assumed
better results would be drawn from
respondents by using clearly defined
multiple-choice questions and scaling
techniques. They also assumed that
limiting answers to simple choices
allowed experts to respond quickly
without worrying about multiple pos-
sibilities for answers.
Next, the authors assumed that
the responses, or votes for specific
attributes, could be combined into dis-
tinct subsets to allow for better statisti-
cal analysis. Originally, the survey was
created to allow respondents to select
one of five multiple-choice responses
or scale values for each question. In
analyzing the data, the authors found
it much easier to quantify and analyze
the responses if they were grouped in
subsets. Doing so provided a simpler
picture of the results without harming
the purity of the data.
The third assumption was that cer-
tain statistical scores or measures pro-
duced by the Bradley & Terry (1952)
method could be used to compare sets
of data that were not directly com-
pared by respondents. This assump-
tion provided an excellent avenue to
rank order skills and attributes based
on statistical scores for their set of
responses. For example, the survey
instrument did not ask experts to com-
pare the skills of “communication”
and “engineering experience” with
each other. It only asked the respon-
dents to rank the importance of each
on a given scale. However, statistical
scores can be calculated for each of the
data sets associated with these skills.
These scores provide a total, statistical
picture of how the respondents feel
about each particular skill. If one skill
has more high-ranking votes than the
other, the statistical scores reflect that
difference. The basis of this assump-
tion comes from David (1963) and
Bradley and Terry (1952) as they com-
pare and rank statistical outcomes of
sets of responses using statistical
scores. Hence, this assumption provid-
ed the ability to directly rank order
skills and attributes that may not have
been directly compared by respon-
dents.
The final assumption involved the
levels at which the results of statistical
tests are considered significant. Cooke
(1991) chooses to reject the null
hypothesis for the Coefficient of
Agreement at the five percent level on
standard chi-square distribution
tables. The authors assumed five per-
cent was an appropriate level and
chose to do the same for this research.
For all other statistical tests in this
research, the authors used the 0.01 per-
cent level on standard chi-square dis-
tribution tables.
The assumptions above served as
guiding principles for the statistical
analysis of the results of this research.
The results of the research incorporate
these assumptions.
Results – Hypothesis Number One
The objective of hypothesis number
one was to determine if the HFPA
project integration concept had a sig-
nificant, positive impact in balancing
project constraints and producing
success on Army health care renewal
(design-build) projects. Four ques-
tions on the survey asked respon-
dents to determine how well HFPA
project integrators assist the project
team in producing successful scope,
time, cost, and quality management
on renewal projects. A fifth question
asked how well HFPA project integra-
tors assist the project team in mini-
mizing operational impacts caused
by a renewal project. These five ques-
tions focused on core project con-
straints that must be balanced to
produce success on renewal (design-
build) projects. Thus, responses to cost, quality, and in minimizing oper-
these questions assist in answering ational impacts.
hypothesis number one. Statistical
results of the responses to these five Results – Hypothesis Number Two
questions are summarized as follows: The objective of hypothesis number
Constraint u value B1 score Tn score pi value
Scope 0.54 -319.45 2078.00 1.000
Operational Impacts 0.23 -108.71 1260.67 0.991
Quality 0.22 -97.88 1152.67 0.988
Time 0.19 -90.14 1112.00 0.984
Cost 0.18 -80.17 1010.67 0.979
Table 4.1. Project integrator impact: statistical results for test of hypothesis one
The results in the u value column two was to determine whether soft or
(Table 4.1) showed positive agreement people skills were the most important
by respondents that HFPA project project integration and management
integrators perform “well” and had a techniques for HFPA project integra-
strong positive impact in assisting the tors. The authors used two primary
project team at producing success with sections of the questionnaire to gather
scope, time, cost and quality manage- data for testing this hypothesis. The
ment on renewal projects. HFPA first section included four multiple-
Integrators also performed “well” with choice questions dealing with the
minimizing operational impacts dur- importance of leadership, partnership,
ing projects. The results in the pi value management, and technical skills. The
column showed that respondents second section included the ranking
believed this at a probability level of of 24 individual skills broken into the
90% or better for each constraint list- four broad skill sets mentioned above.
ed in Table 4.1. At this level, the First, respondents were asked to
authors concluded that the “well” rank the importance of the four skills
selection stochastically dominated the sets: leadership, management, part-
other options for all five constraints nership, and technical skills. The
listed above. Finally, Group Preference authors believed the skills listed in
and Goodness of Fit tests showed statis- the “leadership” and “partnership”
tical significance of the responses and categories were primarily soft or peo-
appropriateness of the Bradley and ple oriented. Thus, using these ques-
Terry (1952) method in all five situa- tions, he desired to show that
tions. Thus, the data illustrated that leadership and partnership skill sets
respondents felt HFPA project integra- ranked higher than management and
tors performed “well” and had a technical skill sets. Results of respons-
strong positive impact in producing es were summarized and rank ordered
success and balancing scope, time, as follows in Table 4.2:
Rank Skill Sets u value B1 score Tn score pi value
1 Partnership 0.87 -1007.56 4760.67 1.000
2 Management 0.75 -704.45 4382.00 1.000
3 Leadership 0.64 -526.67 3926.00 1.000
4 Technical 0.30 -194.15 2562.00 0.997
Table 4.2. Rank ordering of skill sets: statistical results for test of hypothesis two – stage one
September 2004 Project Management Journal • 35
 The results in the u value col-
umn (Table 4.2) showed positive
agreement by respondents that the
skill sets outlined in this table all
had “High Importance” for HFPA
project integrators as they execute
their owner representative duties on
renewal projects. However, the data
also showed that “partnership” was
the most important skill set, fol-
lowed by “management,” “leader-
ship” and “technical skills”
respectively. The results in the pi
value column showed that respon-
Rank Skill
1 Communicating (L)**
2 Project Mgmt Exp (T)
3 Building Trust (P)
4 Team Building (L)
5 Planning (M)
6 Managing Change (M)
7 Improving Relations (P)
8 Aligning Resources (M)
9 Collaborating (P)
10 Creating Alliances (P)
11 Knowledge of Criteria (T)
12 Assuring Quality (P)
13 Empowering Others (L)
14 Measuring Progress (M)
* This value represents the probability that respondents would choose the attribute “>=High” over the
attribute “Moderate” and the attribute “Moderate” over the attribute “<=Low” for these fourteen skills.
** The letters in parentheses following each skill in the list represent the four skills sets (L) Leadership,
(P) Partnership, (M) Management and (T) Technical.
Table 4.3. Rank ordering of skills: statistical results for test of hypothesis two – stage two
dents believed this at a probability
level of 90% or better for each con-
straint listed in Table 4.1. At this
level, the authors concluded that the
“high importance” selection sto-
chastically dominated the other
options for all four skills sets listed
above. Finally, Group Preference and
Goodness of Fit tests showed statisti-
cal significance of the responses and
appropriateness of the Bradley and
Terry (1952) method in all four situ-
ations. Thus, the data illustrated that
respondents felt that “partnership”
36 • Project Management Journal September 2004
skills were most important for HFPA either partnership- or leadership-ori-
project integrators. This skill set was ented. This means that 57% of the
considered a soft skill by the authors skills listed above could be consid-
from the outset of the research. ered, “Soft” or people skills. One final
Next, respondents were asked to consideration is that 3 of the top 4
rank the importance of 24 skills that skills listed in Table 4.3 were clearly
were broken into the four broad soft or people oriented. These skills
skill sets mentioned above. Results were “communication,” “building
of the responses that showed statis- trust” and “team building.” The
tical significance and agreement remaining skill in the top four, “proj-
among respondents are summarized ect management experience,”
and rank ordered as follows (those required frequent use of people skills.
that lacked agreement were not The authors concluded that the
included): results shown above provide strong
support for hypothesis number two.
The authors also believe that having 8
u value B1 score Tn score pi value of these 14 skills fall in the leadership
0.81 -934.192 9233.000 1.0 or partnership categories showed a ten-
dency for experts to select soft skills as
0.70 -690.004 8514.000 1.0
higher in importance than other skills.
0.61 -547.421 8201.000 1.0 Furthermore, partnership skills
0.49 -457.109 7686.000 1.0 appeared to dominate both lists, gar-
nering more than one-third of the
0.52 -453.775 7892.000 1.0
spots on the top 14 skills list. Also,
0.46 -388.483 7403.000 1.0 while leadership skills only filled 3 of
0.43 -373.384 7225.000 1.0 the top 14 spots, 2 of these were in the
top 4. Each of these conditions sup-
0.36 -285.826 7022.000 0.999
ported hypothesis number two. Thus,
0.35 -271.168 6698.000 0.999 the authors believed respondents
0.33 -257.137 6522.000 0.999 showed a tendency to select leadership
and partnership or soft skills as most
0.23 -157.438 6251.000 0.989
important for HFPA project integrators
0.16 -112.302 5433.000 0.974 as they execute their responsibilities on
0.11 -78.659 5281.000 0.938 renewal (design-build) projects.
0.09 -63.049 5113.000 0.906
Conclusions and Recommendations
The purposes of this research were as
follows: first, to determine if the HFPA
project integration concept had a sig-
nificant, positive impact on balancing
project constraints and producing suc-
cess, and second, to identify which
skills were most important to HFPA
The information from Table 4.2 project integrators as they execute in a
appeared to agree with the results design-build environment. The
from Table 4.3. The results in Table authors believe these research objec-
4.3 showed that 5 partnership skills tives were achieved through this
made the top 14, followed by 4 man- research.
agement skills, 3 leadership skills and Given the results shown earlier in
2 technical skills. The results in Table this document, the authors believe this
4.2, which came from an entirely dif- research has assisted in proving
ferent set of questions, rank ordered hypothesis number one to be true.
these skill sets in exactly the same Respondents concur that project inte-
manner: partnership, management, grators assist the project team in pro-
leadership, and technical skills ducing successful scope, time, cost, and
respectively. Another interesting note quality management, and in minimiz-
is that 8 of the top 14 skills were ing operational impacts to the project,
at a level equal to or greater than
“Well.” The research demonstrates this
to be true and statistically significant
for all five of the constraint areas men-
tioned above. Therefore, HFPA project
integrators have a significant, positive
impact on balancing project con-
straints and in producing success on
renewal (design-build) processes.
This research has also been key in
showing that soft or people skills are a
highly important focus for HFPA proj-
ect integrators. Hypothesis two asserts
this is true; the data analysis appears to
support the hypothesis. The analysis
shows that partnership skills are vital
to success for project integrators, fol-
lowed closely by management and
leadership skills. In fact, partnership
and leadership skill sets hold 8 of the
top 14 spots on the statistically signifi-
cant, rank-ordered listing of skills in
Table 4.3. Furthermore, partnership
skills hold more than one-third of the
positions on the top 14 skills list.
Given that partnership and leadership
skills are soft or people-related skills,
the data shows support for hypothesis
two. Thus, soft or people skills are
some of the most important project
integration and management tech-
niques for HFPA project integrators as
they assist the project team in balanc-
ing project constraints and producing
success on Army health care renewal
(design-build) projects.
Although the results of this
research provide support to the stated
hypotheses, the authors understand
that this research only opens the door
to a myriad of questions for future,
potential research. Some of these ques-
tions might lead to a direct compari-
son of each skill or skill set. Others
might focus more on the management
and technical aspects of owner repre-
sentation. Still others might provide a
basis for extending the scope beyond
the HFPA project integration concept
to other types of projects, environ-
ments, and industries. The authors rec-
ommend the development of further
research that delves into owner repre-
sentation and project management
skills across many industries. They
believe it is important for owner repre-
sentatives and project managers work-
ing with design-build construction (or
with the design-build concept in any
industry) to understand which skills
are most important in creating project
success. They also believe it is impor-
tant to continually quantify and moni-
tor the impact of project team
members at all different levels and
processes. Improvement only comes
through measurement and course cor-
rection.
In particular, the authors recom-
mend that the U.S. Army Health
Facility Planning Agency continue
their cutting-edge efforts at developing
the project integration concept. This
concept has a positive impact on
renewal processes, which can certainly
be extended to the MILCON program.
HFPA has been effective in using the
project integration concept to balance
project constraints and promote suc-
cess. The authors believe it will
become more effective as project inte-
grators understand the need to develop
and utilize soft or people skills. HFPA
can also improve success by helping
project stakeholders better understand
the project integrator’s role and
embrace the benefits that come from
working with them. Overall, the
authors believe the efforts of HFPA and
the project integration concept are
truly leading the way to more effective
owner representation in the renewal
(design-build) construction arena.
REFERENCES
Belle, R.A. (2002). Critical market
forces and design-build. DBIA online.
Available from http://www.dbia.org/pubs/
belle01.html. (Accessed April 16, 2002).
Bradley, R.A. & Terry, M.E. (1952).
The rank analysis of incomplete block
designs: The method of paired compar-
isons. Biometrika, 41, 502-537.
Cho, A. & Sawyer, T. Bulk of ambi-
tious $1.6-billion I-15 design-build job
complete. Engineering News Record, May
14 2001, 13.
Chong, N. Integrator training.
Presentation given at the Project
Integrator Training Conference, Falls
Church, VA. November 5, 2001.
Coates and Jarratt, Inc. (1999).
Forces and factors shaping the future of
the project management profession and
its professionals. In The Future of Project
Management, ed. Project Management
Institute. Newtown Square, PA: Project
Management Institute.
Cooke, R.M. (1991). Experts in uncer-
tainty. New York: Oxford University Press.
Cooper, D.R. & Emory, W.C. (1995).
Business research methods. 5th ed. United
States: Richard D. Irwin, Inc.
Covey, S.R. (1989). The seven habits of
highly effective people: Taught by the author.
Audio cassette presentation. Provo, UT:
Franklin Covey Co.
David, H.A. (1963). The method of
paired comparisons. New York: Hafner
Press.
David, H.A. (1988). The method of
paired comparisons. New York: Oxford
University Press.
Department of Defense (DOD).
(2001). Mandatory procedures for major defense
acquisition programs and major automated
information system acquisition programs. DOD
5000.2-R. Washington, DC.
Department of Defense Instructions
(DODI). (2001). Operation of the defense
acquisition system. DODI 5000.2.
Washington, DC.
Department of the Army (DA).
(1988). Field operating agencies of the office
of the Surgeon General. AR 10-54.
Washington, DC.
Design Build Institute of America
(DBIA). (2002). Design-build institute of
America: About DBIA. DBIA online.
Home page on-line. Available from
http://www.dbia.org/au/index.html.
(Accessed April 16, 2002).
Drucker, P.F. (2001). The essential
Drucker. New York: HarperCollins
Publishers, Inc.
Eisner, H. (1997). Essentials of project
and systems engineering management. New
York: John Wiley & Sons, Inc.
Foti, R. Forecasting the future of proj-
ect management. PM Network, October
2001, 29-31.
Goode, M.G. Introduction to man-
agement of construction. Presentation
given at The George Washington
University. September 6, 2000.
Haldane, D. (1998). Megatrends in
project management. In The Future of
Project Management, ed. Project
Management Institute. Newtown Square,
PA: Project Management Institute.
Halpin, D.W. & Woodhead, R.W.
(1998). Construction management. 2nd ed.
New York: John Wiley & Sons, Inc.
HBE. (2002). Design build continues to
September 2004 Project Management Journal • 37
grow in popularity. HBE Blueprint online, vol 1-
no 1. [e-journal] <http://www.hbecorp.com/
Healthcare/pdfs/Blueprint.pdf> (Accessed
April 16, 2002).
International Organization for
Standardization (ISO). (2000). ISO
9000–quality management principles. ISO
online. (Accessed 03 Oct 2002). Available from
http://www.iso.ch/iso/en/iso9000-
14000/iso9000/qmp.html.
Jones, G.P. (1984). A new approach to
the JCT design and build contract. New
York: Longman, Inc.
Kerzner, H. (2001). Project manage-
ment: A systems approach to planning, sched-
uling, and controlling. 7th ed. New York:
John Wiley & Sons, Inc.
Kurmel, T. (2000). Introductory
remarks. As recorded by Denis Petersen.
U.S. Army Health Facility Planning Agency
(USAHFPA) 25th Anniversary Conference.
San Antonio, TX.
McManamy, R. (2001). Globalization
means collaborative skills will be more
prized than ever. It’s a small world, after all.
Design-Build Magazine [e-journal]
<http://designbuild.construction.com/Aug
ust01/globalization801.asp>
National Aeronautics and Space
Administration (NASA). (1995). NASA
systems engineering handbook. SP-610S.
Washington, DC.
Oberlander, G.D. (2000). Project
management for engineering and construc-
tion. 2nd ed. Boston: The McGraw-Hill
Companies, Inc.
Office of The Surgeon General
(OTSG). (1975). Department of the
Army. General Orders Number 29.
Organization of Unit – United States
Army Health Facility Planning Agency.
Washington, DC.
Osborn, J.E. Esq. (2000).
Design/build: Will it lead to even more
construction contract litigation? The
Metropolitan Corporate Counsel, Dec 26,
2000.
Pauli, M. Roles and lines of commu-
nication. Presentation given at the
Contractor’s Training Course, Mobile, AL.
March 18, 2002.
Pells, D. (1998). Global tides of
change: Significant recent events and
trends affecting globalization of the proj-
ect management profession. In The future
of project management. Newtown Square,
PA: Project Management Institute
Project Management Institute, Inc.
(PMI). (1996). A Guide To The Project
Management Body Of Knowledge
(PMBOK® Guide). 1996 ed. Newtown
Square, PA: Project Management
Institute.
Project Management Institute, Inc.
(PMI). (2000). A Guide To The Project
Management Body Of Knowledge.
(PMBOK® Guide). 2000 ed. Newtown
Square, PA: Project Management
Institute.
Project Management Institute, Inc.
(PMI). (1999). The Future of Project
Management. Newtown Square, PA:
Project Management Institute.
Reina, P. (2001). Kazakhstan invests
in Astana’s future. Design-Build Magazine
[e-journal] <http://designbuild.construc-
tion.com/August01/Astana801.asp>
Sanvido, V., & Konchar, M. (1999).
Selecting Project Delivery Systems.
Pennsylvania State University: Project
Delivery Institute.
South Florida Water Management
District (SFWMD). (2001). Improvements
needed in capital maintenance protocol for
the central and southern Florida Flood
Control System. Office of the Inspector
General, Audit #01-18. West Palm Beach, FL.
Tanner, V. (2001). Pentagon gets
back to business. Design-Build Magazine
[e-journal] <http://designbuild.construc-
tion.com/dec2001/pentagon1201.asp>
Tarpey, J. Industry Trends and
Challenges. Presentation given at The George
Washington University. May 20, 2000.
Tulacz, G.J. The top 100: Design-
build firms. Engineering News Record, June
18, 2001, 48-51.
U.S. Army Corps of Engineers
(USACE). (1994). Design Build-
Instructions. Washington, DC.
U.S. Army Corps of Engineers (USACE).
(1999). Design-Build Contracting. ER1180-1-9.
Washington, DC.
U.S. Army Health Facility Planning
Agency (USAHFPA). (1975). USAHFPA
Memorandum Number 10-1.
Organization and functions–United States
Army Health Facility Planning Agency. Falls
Church, VA.
U.S. Army Health Facility Planning
Agency (USAHFPA). (1997). Brochure.
USAHFPA–Worldwide facilities planning,
design, construction, and sustainment servic-
es for military health care missions. Falls
Church, VA.
Van der Merwe, A.P. (1998). The
history of project management: a
search for world best practices.
Infinite.org online. Available from
http://www.infinite.org.za/PM%20His
tory.htm (Accessed May 11, 2001).
Washington State Department of
Transportation (WSDOT). (2001).
Guidebook for design-build highway proj-
ect development. Olympia, WA.
Wideman, R.M. (1996). Improving
PM: Linking success criteria to project type.
PMForum online. PMForum Homepage.
Available from http://www.pmforum.org/
l i b r a r y / p a p e r s / i m p r ov i n g p m . h t m
(Accessed April 16, 2002).

DENIS PETERSEN is a doctoral student at
the George Washington University in
Washington, DC. This document is a
report of research he conducted in partial
fulfillment of his doctoral program. Denis
holds a Bachelor of Science degree in
Civil Engineering from Brigham Young
University, a Master of Public
Administration degree from Golden Gate
University, and is completing his doctoral
work in Engineering Management and
Systems Engineering. He earned his
PMP® certification in March 2001.
E. LILE MURPHREE JR. was educated at The
University of Mississippi (BS Civil Engineering,
MA Mathematics), The Massachusetts
Institute of Technology (SM Civil Engineering),
and The University of Illinois at Urbana-
Champaign (PhD Civil Engineering Systems).
His academic specialties include operations
research, project management, constuction
management, computer applications,
information technology, and management of
technology. In addition to his academic career,
he spent six years as a Division Chief at the
U.S. Army Construction Engineering Research
Laboratory and ten years in private practice.

38 • Project Management Journal September 2004

UNDERSTANDING THE ROLE OF “VISION”
IN PROJECT SUCCESS
ABSTRACT
Rigorous applications of project
management methodologies are
responsible, though only partially, for
project success. We argue, however, that a
significant driver of project management
success is effective and intelligent
leadership communicated through an
inspiring vision of what the project is
meant to achieve and how it can make a
significant positive impact.
An information technology case
study project is presented to illustrate how
project vision provided and maintained
commitment to a complex project that was
judged successful when compared to
similar projects despite the difficulties
described. This success was substantially
attributed to the project leadership
group’s use of a vision.
Keywords: project management leader-
ship; vision; stakeholder management
©2004 by the Project Management Institute
Vol. 35, No. 3, 39-52, ISSN 8756-9728/03
DALE CHRISTENSON, Executive Director, Leadership and Learning Centre, Public Service Agency,
Provincial Government of British Columbia, Canada and DPM (candidate), RMIT University
DEREK H.T. WALKER, Professor of Project Management and Program Director, Doctor of Project
Management at RMIT University
Introduction
Leadership forms a significant body of knowledge, but we will only be focusing
on a very small but important area of this discipline due to limitations of the
scope of this paper. This paper focuses on an examination of the impact of the
leadership construct of project vision on expected project outcomes.
There has been considerable attention given to contributing factors to project
success and failure. Included in these have been the examination of project man-
ager competence, but little has been written about the concepts and constructs of
the role of project leaders in developing and communicating a project vision and
the impact of this on project success. One of the most significant contributions
that any leader can make to an organization or project is that of creating and clear-
ly communicating a shared vision. Therefore, we argue in this paper that, “project
vision” is a significant contributing factor to project success, and, the communica-
tion and maintenance of a project vision will impact project outcomes.
Evidence from a case study of a major information technology project is
offered to support this argument to reveal the importance of project vision and
clarify the impact of this influence upon successful project outcomes. We present
evidence that while many of the accepted critical success factors were absent at
one time or another in this project, a strong project vision may have been signif-
icantly responsible for the ultimate success of the project. We also demonstrate
that while many variables are at play in any project, the development, communi-
cation and maintenance of a strong project vision may be the key to creating suc-
cessful project outcomes. Therefore, development, communication and
maintenance of the project vision become a primary mission for project man-
agers when considering stakeholder management. As such, the traditional stake-
holder analysis must address the stakeholders’ influence on the project’s vision.
We conclude that to completely understand the impact of the project vision and
those who influence its evolution, one must also understand the organizational
structure, culture and stakeholder power relationships.
The remainder of this paper is structured as follows: First, we discuss the con-
cept of project vision, drawing upon the literature with particular relevance to cor-
porate vision followed by how a vision may be developed. In that section we
summarize and present a model for project vision development. We then provide
September 2004 Project Management Journal • 39
example from the literature of both fail-
ures and successes of vision develop-
ment and successes linked to our vision
model. We follow this with a discussion
of a case offered from one of the
author’s direct experience, and in so
doing, draw upon rich contextual first-
hand knowledge. While we acknowl-
edge the potential danger of bias that
using personal experience, we argue
that our opinion-related data has been
subject to intense reflection and discus-
sion with other direct project partici-
pants to validate opinion and to
unearth any in-built biases of interpre-
tation. We then conclude our discussion
and draw attention to implications for
project management practice.
Vision – Not Just a Series of Fancy
Words But a Meaningful Model
What is vision? The dictionary defini-
tion of vision is “the ability to think
about or plan the future with imagina-
tion or wisdom” (Oxford, 2001, p.
2066). Kotter (1995) describes vision
in terms of something that helps clari-
fy the direction in which to proceed—
this makes sense as the word has
implication of the sense of sight.
However, Kotter like many other writ-
ers on this subject, imbue vision with a
transformational quality that enables
not only pure change of X into Y but
doing so with committed purpose and
enthusiasm. As Bennis and Nanus
(1987, p. 82) explain, “…vision articu-
lates a view of a realistic, credible,
attractive future for the organisation
[sic], a condition that is better in some
important ways than what now exists.”
The concept of vision becomes one of
a tool or means to engender passion
and meaning to a project to meet the
envisioned end manifestation.
Taken from the perspective of a
tool to make sense out of a plan, Karl
Weick provides many examples and
illustrations of the power of vision to
enable people to make sense out of a
plan of action. He discusses how sys-
tems of sense-making are vitally
important when specialization and
decentralization results in segregation
of people and teams and differentia-
tion of processes in undertaking an
activity. He talks about highly interre-
40 • Project Management Journal September 2004
lated and dependent sub-systems
being tightly coupled (linearly linked,
so that a break in one link severely
affects other in the chain) and loosely
coupled (parallel links with opportu-
nities to switch emphasis and direction
of energy while maintaining momen-
tum) (Weick, 2001, p. 384).
Project management is about
effecting a change through first segre-
gating tasks into modules undertaken
by specialists with skills to undertake
those tasks and then integrating these
modules into a coherent whole
(Morris, 1994; PMI, 1996; Pinto &
Trailer, 1998; Cleland, 1999;
Dinsmore, 1999; Turner, 1999; Walker
& Hampson, 2002). To understand a
loosely coupled system in a project
management context, we could see a
sequence of events unfolding uneven-
ly, sporadically, discontinuously, and
unpredictably—this is one view of the
challenge of project management.
From a tightly coupled perspective, we
could see project management excel-
lence as a smooth transition of logical-
ly linked activities, each
interdependent and each converging to
be actioned when and where required.
The intelligent use of both tight and
loose coupling also conforms to the
concept of project management matu-
rity continuums (Kerzner, 2001, pp.
1046-1049) where maturity is
described at five levels of attainment.
Unfortunately, as the above-cited proj-
ect management experts attest, fre-
quently subsystem parts are very
difficult to coordinate and align, par-
ticularly as people and teams have a
variety of motivations, aspirations and
agenda that might and often does
clash with best-for-project objectives.
One of the principal tasks that a
project management leader needs to
effectively perform is ensuring align-
ment of goals and resultant commit-
ment to project goals. Because projects
are tightly coupled, this situation
requires all team members to be able
to make sense of project goals so that
they can be prepared to support proj-
ect goals and internalize these as being
aligned with their own. Further,
because much of the interaction
between project teams and subsystems
is loosely coupled, there needs to be
clear understanding of the cause-and-
effect loops that exist so that adverse
action by one group can be traced in
the minds of that group to the impact
the adverse action may have on others
connected through the project process
and how that might adversely affect
them later on. This requires project
participants or stakeholders to under-
take a sense-making exercise (Weick,
2001, p. 184) and an envisioning exer-
cise focusing on what the end point
should be, so that the weaving of
seemingly unconnected actions can be
clarified to understand how the parts
form the whole.
When this occurs, project partici-
pants might better see the logic of
mutual adjustment and enacting cop-
ing mechanisms to provide a loosely
coupled system for the required flexi-
bility needed for projects that are high-
ly complex and tightly bound while
living in a tightly coupled framework.
As Weick argues, “The real trick in high-
ly reliable systems is somehow to achieve
simultaneous centralization and decen-
tralization” (Weick, 2001, p. 340). This
idea of systems reliability being linked
to tight-loose coupling also relates to
the notion of underlying assumptions
being the foundation of culture (dis-
cussed later in this paper–see Figure 4).
Unspoken assumptions can generate
confusion, whereas explicit artifacts
such as a vision statement provides the
link between assumption and a means
to provide both tightly and loosely
coupled systems.
The organizational culture litera-
ture identifies vision as being an
important contributor to the character-
istics of a culture. Organizational cul-
ture’s most visible manifestation (often
undecipherable) is artifacts such as sto-
ries, history, image, identity and organi-
zational structures. Underpinning these
are organizational values, and deeper
again are the underlying assumptions
(generally invisible) that are shared by
the group in that culture. A vision state-
ment may become an artifact, a docu-
ment describing project goals and
aspiration. This will not have meaning
unless it reflects the values of the cul-
ture concerned.
 Values are those behaviors that are
cherished by members of the culture or
subculture. So that, for example, a
project vision may state that people
will treat each other with integrity and
respect, avoiding blame and litigation.
However, if that organization has a his-
tory of scapegoating, seeking opportu-
nities for making claims against
project supply-chain members and
conducting a paper trail to cover them-
selves at every opportunity, then a con-
flict arises between the vision
statement artifact and the culture’s
value system. Clearly, there is a palpa-
ble link between values and artifacts.
Similarly, values are palpably
linked to assumptions. The above-
illustrated example of defensive behav-
iors, such as creating a paper trail,
being unhelpfully bureaucratic or
identifying scapegoats, may be derived
from a strong foundation in assump-
tions that other supply-chain partners
will take advantage of an organization.
The need to expose and unearth such
assumptions so that they can be under-
stood and dealt with is a fundamental
feature of partnering and alliancing —
an increasingly popular approach to
managing complex projects (Bennett &
Jayes, 1995; CII, 1996; Lenard, Bowen-
James, Thompson & Anderson,. 1996;
Doz & Hamel, 1998; Lendrum, 1998;
Walker & Hampson, 2002).
Creating an effective project vision
requires excellent communication
skills and a deep understanding of
both organizational culture and the
history and trigger mechanisms that
create underlying assumptions of indi-
viduals and groups comprising project
teams. A project vision can be an arti-
fact that defines the project’s soul so
that it anchors project participants
through their core values to a project
outcome that all can relate to. Clearly,
this is a difficult task requiring intelli-
gence and wisdom on the part of proj-
ect leaders. Crafting a vision requires
insights into the underlying assump-
tions that determine values and calls
for creating artifacts that can be accept-
ed and internalized.
The role of the vision artifact is
well recognized as a focus for manag-
ing projects. Leading project manage-
ment commentators have observed,
“The most significant success factor for
project teams is that they have a com-
mon and shared idea of what difference
they are trying to make as a result of the
project. Such a vision can be built up by
exploring questions with stakeholders
and project team members, such as:
• How will this project make a dif-
ference to the organisation [sic]?
• How would we know that this
project has been highly successful?
• What in our wildest dreams
would you like this project to
achieve?” (Briner, Hastings, &
Geddes et al. 1996, p. 89).
It is important to discuss the basis
of change management because proj-
ects are about instigating a process or
product change (Cleland, 1999). The
interesting link between change and
vision is that the vision statement clear-
ly seeks to convince its audience that
the change is worthwhile and indeed
bestows benefits — it also indicates
what will be different or describes a dif-
ferent end state (change). Change also
invokes anxiety because it challenges
the status quo and requires expenditure
of transformational energy.
Two different types of anxiety have
been identified as affecting people’s
readiness for change (Schein, 1993). In
this paper’s context of discussing proj-
ect vision, we can also look at project
vision as fulfilling a changed approach
to participating in the project. This
change is usually moving from a busi-
ness-as-usual expectation to striving
for excellence by aspiration to make a
palpable and positive impact. A project
vision concept, encapsulated in a
vision statement, often seeks to moti-
vate and inspire team members and
other stakeholders to positively partic-
ipate and shed negative behaviors
(such as naked self-interest) in favor of
the common good, resulting in wide-
spread benefit.
Anxiety Type 1 is the feeling associ-
ated with an inability or unwillingness
to learn something new because it
appears too difficult or confronting
(Schein, 1993, p. 86). In this situation
we deny the problem exists, search to
blame others for the symptoms requir-
ing the change, or simplify the per-
ceived problem triggering change in
terms that when seen in retrospect,
appear ridiculous. The phrase “who
would have foresaw this … or who could
have anticipated that…” are often used,
yet in such cases the problem and
change action is often quite clear to
those not clouded by Anxiety Type 1.
Unfortunately, Anxiety Type 1 behav-
iors are universal and all too evident
with a management response to
mount more pressure to conform to
the expected response. This can exacer-
bate the situation as it drives people
towards panic, and when people are
under severe stress (panic) they revert
to earlier patterns of learning even
when these patterns are no longer
effective or appropriate (Weick, 2001).
This leaves people in a bind. They need
to change and update their knowledge,
but this is a painful and energy absorb-
ing process.
Anxiety Type 2, the fear, shame, or
guilt associated with not learning any-
thing new, particularly when survival is
challenged without action being taken,
(Schein, 1993, p. 88) is the type of
anxiety that change activists and lead-
ers need to cultivate. Moreover, they
need to ensure that Anxiety Type 2
pressure is greater than Anxiety Type 1.
This is uncomfortable to many organi-
zations because it requires expensive
and extensive support and resourcing
to provide the escape route from this
form of anxiety. It is easy to see why
many organizations would opt for a
strategy of putting pressure on individ-
uals or business units (BUs) and then
leaving them to sort out the dilemma
”on the cheap” by not providing ade-
quate support systems. It is instructive
that this strategy seems to almost
always cost more through failed plans,
dreams and commitments inhibiting
delivery of the expected results. The
result is frequently blame and negativ-
ity. If Anxiety Type 2 is responded to
through a project vision, then we may
see that project leaders can make a
positive difference through providing
enabling support systems. Creating
Anxiety Type 2 grabs attention and is
consistent with what knowledge man-
agement gurus (Nonaka & Takeuchi
September 2004 Project Management Journal • 41
1995) refer to as providing a shock
impetus to trigger improvement.
Anxiety Type 2 provides a trigger to
search for a way out of the Anxiety
Type 1 dilemma. Anxiety Type 2
impacts must be greater than Anxiety
Type 1 so that the project leader needs
to prepare a general solution outline
(the project vision statement) that
enables people to find their own way
to channel their energies and commit-
ment to move from a position of
defensiveness to one of confidently
addressing the change deployment
that constitutes the project in question.
The characteristics of a company
vision make a useful model. A well-
constructed vision comprises two ele-
ments, the core ideology and
envisaged future (Collins & Porras,
1996). The core ideology comprises
core values and core purpose. Core
organizational culture values have
been discussed earlier in this paper.
Sometimes this is explained in terms
of the ”X way.” The envisioned future
is what has been termed a big, hairy,
audacious goal (BHAG). It is usually
described vividly and, as argued by
(Collins & Porras, 1996), BHAGs are
needed as stretch goals to provide a
framework for progressing towards the
envisioned future. Core purpose is
often defined in terms of a mission
statement. They also observe that the
basic dynamic of visionary companies
is to preserve the core culture and pur-
pose and to stimulate progress towards
the envisioned future. Stretch goals
should not encourage people to work
harder but work smarter often through
more effectively sharing and leveraging
knowledge (Nonaka & Takeuchi, 1995;
Davenport & Prusak, 2000).
Both a project vision and an orga-
nizational vision share many common
characteristics. However, a project
vision is more complex because projects
use multiple temporary organizations,
each with their own cultures and sub-
cultures. This notwithstanding, we
argue that a project vision should have
the following characteristics:
1. It must be understood: It must
capture the core purpose, preferred
future state and essence of the
project objectives, its raison d’être
42 • Project Management Journal September 2004
2. It must be motivational: It must
make a convincing case for fol-
lowing the project vision con-
cept that can be internalized by
project stakeholders and that
provides a compelling value
proposition
3. It must be credible: It must be
consistent with stakeholder cul-
tures or sub-cultures to appeal
at the assumptions and values
level so that the vision state-
ment artifact resonates with
them;
4. It must be demanding and chal-
lenging: It should be proactive
to facilitate teams to work
smarter and more effectively,
perhaps identifying stretch
goals.
Note that vision and mission state-
ments may get confused and under-
stood as synonymous by some
academics in this area of study. Mission
may be thought of more both in terms
of goals and objectives in an action
sense, whereas vision may be thought
of as a state of being or end state.
Vision and Transformational Leadership
The business literature contains count-
less examples of the nature of leader-
ship, and there appears to be a broad
consensus that leadership and man-
agement are both different in tasks
undertaken and have different aims.
Whereas managing is about the plan-
ning for the deployment of resources
to achieve an objective, leading is
about using “power to influence the
thoughts and actions of others”
(Zaleznik, 1998, p. 63). Management
is also seen to be about coping with
complexity and bring relative order out
of chaos, whereas leading challenges
the status quo and copes with change
by setting a direction and aligning peo-
ple’s motivation and abilities through
effective communication of the chosen
direction (Kotter, 1998, p. 41). Thus,
leaders use vision-making and vision-
communicating as their tools to
achieve strategic goals.
The link between leaders and fol-
lowers is an interesting one that has
also attracted much attention. Most
notably, the contingency approach to
leadership maintains that the style of
leadership is principally determined by
the maturity of followers. Briefly,
maturity relates to the degree of com-
mitment and the degree of capability
of followers to do what is required of
them (Hersey, Blanchard & Johnson,
1996, p. 208). Four leadership styles
were developed around this concept:
• Style 1 involves using direct
forms of power to coerce follow-
ers into following instructions in
situations where either commit-
ment is largely absent and abili-
ty is limited so specific
instructions must be given under
close supervision
• Style 2 involves explaining the
goals and providing opportuni-
ties for clarification and building
solutions to identified problems
• Style 3 involves sharing and can-
vassing ideas and cooperatively
making decisions with the lead-
ers as facilitators
• Style 4 involves identifying the
goals and then turning over
responsibility for finding ways
to achieve these goals by the
teams involved.
Styles 3 and 4 do represent a more
empowered manner for leaders to influ-
ence followers, however each of these
styles implies a somewhat transactional
leadership style—the leader provides
resources and rewards for followers to
accomplish the required tasks.
Transformational leadership is
defined in terms of the four I’s, (Bass &
Avolio, 1994, p. 2; Avolio, 1996 p. 5)
that is, as “when leaders:
• Stimulate interest among colleagues
and followers to view their work
from new perspectives (Individual
consideration)
• Generate awareness of the mission
or vision of the team and organiza-
tion (Intellectual stimulation)
• Develop colleagues and followers to
higher levels of ability and potential
and; (Idealised [sic] influence)
• Motivate colleagues and followers
to look beyond their own interest
toward those that will benefit the
group (Inspirational motivation).”
 The preceding four I’s also match
our four vision characteristics stated
earlier. The leadership literature
acknowledges advantages inherent in
adopting a transformational approach
as it appeals to more powerful intrinsic
motivational energy that transcends
self-interest, and thus is better placed
to be an effective vehicle for delivering
outcomes for multiple stakeholders
(Bass & Avolio, 1994; Avolio, 1996;
Bennis & Nanus, 1997). In at least one
study involving data gathered from
over 2,000 managers and employees at
a major Australian bank, researchers
found strong evidence associating
group cohesion with transformational
leadership (Carless, Mann & Wearing,
l1996). This indicates that this concept
can be translated effectively to devel-
oping an effective and appealing vision
for managing projects, because proj-
ects require a strong cohesive team
focus on the desired project outcomes.
However, what is particularly rele-
vant to project organizations where
multiple stakeholders have varying
specific ideals of their desired project
outcome is that a transformational
leadership is insufficient by itself.
Visions need to be communicated
clearly and simply, often in terms of a
metaphor with a powerful visual
image or sensual symbol of achieve-
ment that resonates with those stake-
holders required to commit to
delivering these outcomes (Bennis and
Nanus, 1997, p. 100). One vital skill of
project management leaders is to facil-
itate the development of a powerful
project vision as well as to develop a
communications strategy that effec-
tively explains and engenders commit-
ment, rather than just compliance to
the vision’s delivery.
This may be accomplished
through a charismatic approach where
the leader becomes teacher, mentor,
coach reformer or revolutionary, but
not all projects can expect to attract
such an individual to contribute these
powerful gifts. Transformational lead-
ership, however, is about leaders “struc-
turing and articulating problems for
followers, enabling the followers to more
easily comprehend problems so that they
can more effectively deal with them”
(Bass, 1985, p. 31). This seems to
address the Type 2 Anxiety discussed
by Schein (1993). Either a charismatic
individual or a coherently focused
stakeholder reference group can
achieve this. We argue that a well-cho-
sen stakeholder reference group might
be better positioned than a charismat-
ic leader to develop the vision because
of the wider experience, diversity of
knowledge and perspective and fuller
appreciation of stakeholder con-
stituencies that a truly representative
and intelligent reference group can
offer. Thus while transformational
leadership can be linked to charisma,
it is not necessarily dependent upon it.
Developing and Crafting a Project Vision
The development of a project vision is
complex and occurs in a number of
ways. One way is illustrated in Figure 1.
Possibilities for Project
project vision champion
Project leader
Type of project appointed
Develops a
project vision
+ brief
Figure 1. One situation for developing a project vision
The type of project influences the
way that a project vision may develop.
Project goals and delivery methods
may be well defined and this matrix
has been used to define project types
(Turner and Cochrane 1993). The sim-
plest situation is where the goals are
relatively clear and the delivery meth-
ods well understood. The most com-
plex and complicated projects are
characterized by ill-defined goals as
well as uncertainty surrounding deliv-
ery methods. Whatever the situation
regarding goal definition is has an
impact on the possibilities for develop-
ing a project vision. If the goals are rel-
atively ill-defined, any potential
project champion (i.e., that individu-
September 2004 Project Management Journal • 43
als or small groups who proactively
seek to establish a case for the project
development) will need to identify
and assemble a stakeholder reference
group to help define goals. This group
will then source and choose a project
leader and this reference group; the
project champion and project leader
will then define goals and develop the
project vision.
In the less-complex situation illus-
trated in Figure 2, a project champion
has a sound understanding and grasp
of the project goals and can source and
appoint a project leader. The leader
can then identify relevant stakeholders
with a view to form a reference group
to work with the project leader to
develop a vision. Whichever approach
is adopted, the vision can be devel-
oped with the insights of relevant
stakeholder representatives. This
Identifies
originator
stakeholders
Forms an
originator stakeholder
reference group
process facilitates not only group input
and diversity of insights for building a
credible solution, but also develop-
ment of stakeholder and team under-
standing. This is the process of culture
creation—of developing the assump-
tions that underpin values, which
results in identifying the artifacts of
culture (Schein, 1985). If this process
is effectively undertaken, then it can
facilitate a clear understanding of
stakeholder goals and their aspirations
for the project; it can inspired motiva-
tion and build credibility to provide
the impetus for designing challenges
that spur innovation.
Vision development appears to be
developed through a set stage of events
 groups to ensure that the vision is as
Identifies clearly understood in terms of impacts
originator and influences as possible.
Possibilities for Project stakeholders
project vision champion It is at this point that a vision real-
ity check can take place to ensure that
the vision is valid to stakeholders. This
Project leader is generally undertaken to varying lev-
Forms an
Type of project appointed originator stakeholder els of success. In some cases, the proj-
reference group ect management group assumes much
and stakeholders may not be consult-
Develops a ed. In other cases, feuding stakeholder
project vision
groups may generate a paralysis. In
+ brief
between lies the ”reasonable” path.
The outcome of step 5 is the develop-
Figure 2. Less complex situation for developing a project vision ment of the project vision as articulat-
ed (or often inferred). This has inputs
that occur in varying levels of intensity. in terms of impact upon stakeholder from management policy and external
The process begins either formally or groups and priority weightings can be management strategy, for example
informally with a stakeholder vision established. It is important to under- from legal or regulatory agencies. Once
beginning to emerge so that some pre- stand that negotiations should be the Vision is communicated as indicat-
ferred future becomes evident, general- undertaken on a win-win basis ed in step 12, then monitoring can
ly through a clouded or foggy process because once the implications of vari- take place. This can consist of checking
of struggling to understand fundamen- ous priorities are explored and under- how effective the vision communica-
tal issues affecting the project. If stake- stood, often unintended consequences tion strategy has been (step 13); the
holders are gathered together as can be avoided. A project manager implications of the vision through
indicated in Figures 1 or 2, then stake- who has the skills to both unearth con- monitoring project sub-culture reac-
holder representatives can either claim flicting priority impacts and facilitate tions (steps 14 and 15); and revisiting
a voice (if there is a need to be improved prioritization of objectives the periodically the relevance of the
assertive), or the project management can avoid many potential intra-stake- vision. It would be detrimental to the
representative can actively encourage holder disputes. The vision can then be project to overlook steps 16 and 17, as
and nurture this voice to be heard (if developed (step 5 in Figure 3) using a both priorities and relevance change
the stakeholder group is reticent or transformational leadership style that often during long or complex projects.
passive). Either way, project vision pri- addresses the four I’s and the suggested The model of project vision gener-
orities can be established. four characteristics of a good project ation illustrated in Figures 1 and 2 and
A process of negotiation can take vision. This may be best to be under- the development of a vision statement
place whereby these can be evaluated taken through stakeholder reference as illustrated in Figure 3 occurs in vary-
ing degrees from virtually never to very
substantially. We argue that the degree
1. Stakeholders
2. Identify 3. Stakeholder to which Figure 3 model is followed,
stakeholder representatives
image of preferred representatives rate their that is, the level of genuinely empow-
future priorities priorities 4. Negotiations ered and positive interaction between
and discussions about
6.1. Executive
implications of vision project stakeholders and the project
element priority
strategic alignment weightings management team, has a direct impact
check
6. Stakeholders
reference group
5. Develop
vision
upon the quality of a project vision.
reality check statement 17. Stakeholders This quality determines perceptions of
reference group
9. Training reality check how well the vision is understood by
8. Project and development
management policy stakeholders, how the vision appeals
strategy 7. Develop 16. Monitor
vision communication
10. Management vision to motivational instincts, how credible
strategy relevance
process the vision appears, and how challeng-
policy
11. External ing the vision can be without appear-
management 12. Deploy
strategy vision
ing too demanding to be seen as
15. Monitor
13. Monitor strategy achievable.
impact on project
vision deployment
sub-cultures
performance Intelligent project vision state-
14. Vision
deployment ments can be evaluated in this light for
implications
their effectiveness of being likely to
achieve the vision objectives. We argue
Figure 3. Model of vision development process that the key to developing an effective

44 • Project Management Journal September 2004

project vision is to make objectives
and purpose clearly understood, to
inspire motivation, and to ensure that
the project vision is credible and chal-
lenging. All this is accomplished
through organizational and ethnic cul-
tural frameworks and mental models
that enable people to relate to each
other, understand each other’s world-
views and share assumptions that cre-
ate shared values that make sense of
cultural artifacts such as stories, a proj-
ect vision and other forms of group
”branding.”
Figure 4 illustrates our model of
vision effectiveness. Understanding,
motivation, credibility and challenge
can move the emotional attachment
that people feel. This can range from a
vision statement being seen as a super-
ficial document or as a cultural artifact
through the vision providing a resonat-
ing truth that taps into the very core
depths of a culture or sub-culture. Each
and every project has a genesis in a real
need. This need must be articulated to
ensure that it receives its due and just
attention and resource allocation.
We argue that much of the skill of
project management leadership is
about ensuring that the project need is
adequately articulated into a project
vision statement that facilitates enthusi-
asm and commitment for its successful
realization. In this way, a deep under-
standing of the value of the project, its
motivational potential for those
involved, and its credibility as a worth-
while endeavor that aspires to achieve a
best-in-class outcome can be encom-
passed through the artifact of the vision.
Case Examples from the Literature of
Successful and Failed Vision
There are innumerable cases that could
be discussed here but we have chosen
to seek out only a few salient ones
from the literature that appear to typi-
fy examples of the good and the bad.
We have chosen one case where all
four of the above conditions seem to
have been globally met, and one case
where they all appear to be met within
Japan, but due to difficulties in the
Japanese historical and ethnic cultural
context being transferable, was less
successful than the first case. The third
case was an unmitigated disaster and
lack of a credible and reliable vision
was cited as a substantially cause.
One interesting example reported
by Nigel Holden tells how a company
vision for global knowledge-sharing
and use of a common approach to
operating procedures led to an innova-
tive process of creating the concept of a
corporate facilitator. These facilitators
were tasked with ensuring that the cor-
porate culture of a Danish-based glob-
al pharmaceutical and health product
enterprise. the ”Novo Nordisk Way of
Management,“ was spread across its
global operations (Holden, 2002, p.
110). The ”Vision 21 purpose” is cited
as “Our business is to discover and
market products which satisfy real
needs—improving the way people live
and work. We find better ways to fight
the burdens of disease and to provide
sustainable biological solutions to
industrial problems.” The company’s
way includes directions on how they
approach implementation of proce-
dures for accountability, being ambi-
tious for excellence, openness and
honesty, customer focus, readiness for
change, and social responsibility
(Holden, 2002, p. 107).
In this case, Vision 21 was imple-
mented through the widespread adop-
tion of the Novo Nordisk Way of
Management using 14 specially select-
Extent of
Extent of Motivation
Understanding
ARTIFACTS
VALUES
ASSUMPTIONS
Figure 4. Project vision effectiveness
ed senior level personnel called facili-
tators. These facilitators visited each
business unit (BU) throughout the
global organization to help the BUs
adopt standard approaches in a way
that shared knowledge about best
practices developed within other BUs,
and to improve communication and
transfer of values across the organiza-
tion. Thus, Novo Nordisk ”walked-the-
talk” by developing its vision and
acting upon it with credible, strong
leadership models that were credible,
and through its delivery model of facil-
itation and motivated BU commit-
ment and support. It should be noted
that this organization experimented
with the facilitator concept to activate
its vision in late 1996 and that as of
1999, the concept was judged a success
and it was continued and improved
(Holden, 2002, p. 129). Key aspects
relating to its vision were that the
change program could be viewed as an
initial three-year project that was suc-
cessfully completed. It was clearly
understood, inspired motivation, was
credible and challenging.
Another case study cited by
Holden (2002) relates to Konosuke
Matsushita (KM, as he became
known), the founder of the Japanese
global company Matsushita, as having
elaborated his organization’s raison
d’être in ”the basic management objec-
Degree of
Credibility Degree of Stretch Goals
Challenging the Status Quo
+ Encouraging Innovation
Low
ity
tens
of In
vel
e
Deeper
re L
ultu
C
red
Sha
Taproot
September 2004 Project Management Journal • 45
tive” and ”the company creed.” These
are quoted respectively as
“Recognizing our responsibility as
industrialists, we will devote ourselves
to the progress and development of
society and the well-being of people
through our business activities, there-
by enhancing the quality of life
throughout the world,” and “Progress
and development can be realized only
through the combined efforts and
cooperation of each employee of our
company. United in spirit, we pledge
to perform our corporate duties with
dedication, diligence and integrity”
(Holden, 2002, p. 141). This indicates
the types of messages and organiza-
tional culture artifacts that have been
developed to facilitate linking vision
with action.
An interesting aspect relating to
developing a vision is that KM devel-
oped Matsushita’s value system based
upon his own philosophy and experi-
ences of the Japan of the 1920s and
1930s. This creed was based upon a
humanist tradition and in fact pre-
dated constructs of social responsibili-
ty and triple bottom line (social,
economic, and environment) by over
50 years. His vision that he persuaded
his entire company to adopt was based
upon service to the public, fairness and
honesty, teamwork for the common
cause, uniting effort for improvement
and courtesy and humility (Holden,
2002, p. 143). He championed and
nurtured that vision to the extent that
it became a corporate scripture. It was
clearly understood by the Japanese
employees (assisted by their instinctive
and tacit knowledge of Japanese histo-
ry and culture), inspired motivation,
and was credible and challenging.
Holden (2002) makes a cautionary
note that is highly relevant to this
paper, namely, that the Matsushita
vision was poorly understood by its
non-Japanese employees, because
many of the embedded cultural values
and assumptions were purely Japanese
(and moreover related to the specific
historical context of Japan’s history of
1920s to post-war and through to the
modern day). This reinforces the
embeddedness of history in develop-
ing cultural values that support arti-
46 • Project Management Journal September 2004
facts such as a corporate vision.
We have provided two successful
examples of a vision deployment
(accepting the caveat of Matsushita’s
mixed success with transferring ethical
values from a core ethnic culture to
local affiliates in non-Japanese Bus).
We will now provide a well-cited exam-
ple of a spectacular failure of an IT
project to illustrate how a lack of
vision significantly contributed strong-
ly to it failure (Drummond, 1998).
The following contains a summary of
aspects of this case:
The Taurus project was proposed
for the UK Stock Exchange to provide
an online processing of share trading
and other ancillary activities. The £500
million project would later be aban-
doned in 1993. It was a highly complex
project that was demanded urgently
with the securities industry requiring
an 18-month time frame for design,
development, and deployment. A stan-
dard software package that was decided
upon as the basis for the system was
modified significantly, but the degree
of modification and the clarity of what
were to be the final features were
unclear, ill-defined, and subject to
much uncertainty. This added to the
confusion surrounding what the sys-
tem was meant to look like.
Further, the confusion and unreal-
istic demand on delivery time grew as
those responsible for delivery slipped
into being unmotivated and skeptical
about the prospect of any measure of
success. Each month that went by, the
project seemed to be no further
advanced because the needs and
demands for change increased and
much of the work produced became
useless due to these change requests.
Also, the market changed from ”bull”
to ”bear.” This highlights the impor-
tance of a vision being subject to ”real-
ity checks” as described in Figure 3
(steps 6 and 17). The delivery team
was certainly challenged, but not by
stretch targets designed to initiate
innovation, but by continual demands
to work harder to catch up with lost
progress. When a system is failing to
work due to a malfunctioning system
dynamic, then continued, increasing
resource inputs or making the system
more tightly coupled, it merely acceler-
ates the manifestation of the present-
ing problems (Weick, 2001, p. 235).
Thus the project as described by
Drummond (1998) based upon her
significant sources of data and reports
indicate that the project vision: was
poorly understood, inspired lack of
motivation, was not credible, and was
challenging in a negative and not pos-
itive way. The result of project failure
could be expected from the four
dimensions of vision development
suggested by our analysis. If all project
team members and stakeholders had a
clear vision of the purpose of the proj-
ect, Taurus’s spectacular failure could
have been avoided.
This case strongly indicates that
the role of project leadership is pivotal.
However, while the project manage-
ment literature appears to be dominat-
ed by the importance of techniques
used to manage projects, there is a
noticeable gap in the literature relating
to the impact of project leadership and
the development of a project vision
that unites project teams and supply
chain partners as a critical success fac-
tor for projects.
The Case Study
Joint Information Management System
(JIMS) was started in the early 1990’s
to address the business need for a cen-
tral repository of information for a
government agency involving four
interrelated elements of a government
service delivery system in which data
would be gathered about clients that
could be shared (subject to strict priva-
cy and confidentiality arrangements)
to enhance effectiveness of informa-
tion management—to benefit both
client and state interests. An informal
project was initiated to deal with the
issue of why individual key stakehold-
ers did not have an effective or uni-
form means to share information.
The project is typical of the need
for many service organizations with
many interlinked but separately oper-
ating entities to share vital authorized
information. A goal is to avoid dupli-
cation, information entry and mainte-
nance errors, and wasted management
energy in establishing duplicated
information systems. The original
vision was to develop a simple central-
ized case tracking database system.
However, as is often the case in devel-
oping such systems, as additional
requirements were identified, the
grander idea of a System for Case
Integrated Processing (SCIP) was
spawned, and an official project was
sanctioned.
Although the vision has remained
fairly constant over the 10-year life of
the project thus far, the project man-
agement structure matured from a
loose group of individuals with a good
idea (referring to SCIP) to a formal
project management structure with a
widely understood vision (referring to
JIMS). This is typical of what has been
described as working within the
”whitespace” in unofficial projects that
test the waters, with good ideas being
developed in an incubator before
being adopted as officially sanctioned
project ”blackspace”—fully funded,
resourced and subject to the relevant
corporate management systems
(Maletz & Nohira, 2001).
With this movement towards
greater structure came greater rigidity
of decision-making, less creativity in
problem solving, and ultimately an
advocacy positioning of stakeholders.
The movement to the “blackspace” was
required to develop the business
requirements and business case(s) so
the project could move forward to an
official funding stage of development.
The creation of the vision was a joint
effort and there was a reasonable
expectation that it was also a shared
vision. However, it quickly became
apparent that the mission, the ultimate
means to achieve this vision, was
mired in the dilemma of key stake-
holders all desiring separate business
requirement sessions.
The formal project management
structure was composed of a steering
committee and various subcommittees
(business change council, change
request, configuration, and training).
The chairperson of JIMS Steering
Committee, a senior Executive from
one of the core agencies, acted as the
executive sponsor. A full-time project
manager was assigned from within the
ranks of the Information Technology
Division, a central agency providing
technology support to all government
branches. Key stakeholders had repre-
sentation on all committees and sub-
committees. Project mangers were
selected for each core agency and des-
ignated as representative key stake-
holders. People selected to be project
managers, however, had no formal
training to meet their new responsibil-
ity and title. Their authority was uncer-
tain within their respective agencies
and dubious within the context of the
project management structure. It
appeared that the executive members
of the individual agencies still retained
conduct of decision-making and
resource allocation and left only rou-
tine management functions to the
project managers. Figure 5 illustrates
the project leadership structure.
Client
department
representatives
Software owners
Steering committee
General
community
Employees
Figure 5. JIMS stakeholder influence and impact map
In essence, assigned project man-
agers had none of the conditions,
including an organizational career
structure to reward project managers,
that Sauer, Liu and Johnston et al
(2001, p. 45) indicate is necessary to
increase the probability of project suc-
cess. The lessons to be learned here is
to separate the actual leadership struc-
ture from the individual skills and
capabilities of the managers in those
positions as important success factors.
The presence or absence of one or the
other will negate the positive impact of
the other: For instance, the right struc-
ture cannot work with the wrong lead-
ers, and the right leaders cannot make
a flawed leadership structure work.
Therefore, a successful project must
have both an appropriate leadership
structure and the individual leaders
must have the leadership skills and
abilities, and personal motivation, to
make it work.
Interestingly, the intent of JIMS
was not to have unskilled managers in
these roles. The assumption was that
they could benefit from “on-the-job”
training and be successful by learning
as they went. However, the intangible
cost of this steep learning curve and
professional stress resulting from on-
the-job project management training
is, in our view only recoverable in
organizations where project manage-
Subcommittees
departments,
specialized groups
Strong
Impact
Weak
project sponsor Impact
Strong
Influence
Project manager
Other
clients
ment recruits are cycled from one proj-
ect to the next regularly, and where a
peer support network is in place to
assist with this important mentoring
process. Also as a consequence, in
organizations where no recurring proj-
ect management assignments are pos-
sible, this investment in training
would be lost nevertheless.
The age-old problem of a func-
tional versus a matrix organization
also becomes apparent if full-time
resources assigned to a project remain
September 2004 Project Management Journal • 47
under the supervision of their own
organizations rather than reporting to
the overall project manager. In some
parts of the organizations this
changed, but for the most part this
dynamic remained throughout JIMS.
Assigned project managers therefore
only began to have greater moral
authority and control over the projects
as their experience grew and the host
agencies came to accept the expertise
of dedicated project managers and the
application of project management
principles when they experienced the
benefits of managing JIMS this way.
The credibility of the vision began to
unravel as the stakeholders began to
question the value of the integrated
information management solution,
while the ability of a project manager
with sufficient authority and influence
to maintain the original vision waned
and declined due in part to the per-
ceived relevance of the project manag-
er position and role.
Although a formal project man-
agement structure was adopted for
JIMS, the project seems to have
required a more flexible approach to
accommodate both the desire of indi-
vidual stakeholders to have personal
influence while mitigating the unfore-
seen risks that this structure actually
allows. As Maletz and Nohira (2001)
explain, to mitigate this often requires
a creative approach that necessitates
moving outside the formal structure
but remaining within the boundaries
of the original vision. Once again, this
reinforces the need for a central vision
that is compelling and which everyone
will support, regardless of the political
and organizational issues that arise.
Achieving this in practice was not easy
when the entire premise of the JIMS
system was to link a series of separate
and normally serial government
processes. For each part of the process,
a key stakeholder or core agency need-
ed to input data that was then accept-
ed by another key stakeholder or
agency to use or add value to the infor-
mation input flow continuing through
the service-provision business model.
Although the vision was apparent-
ly understood, business requirements
were identified, project charters signed,
48 • Project Management Journal September 2004
design specifications approved and
proof of concept modules displayed,
front end core agencies participating in
the system still had the option to walk
away from the project if they didn’t
like it. It soon became apparent to
everyone that the project had only cur-
sory buy-in from one key stakeholder
with no firm commitment to the
vision. Unfortunately due to the pro-
ject’s leadership structure, there was no
mechanism to preserve their continued
participation. Where they remained
conceptually committed to the vision,
they still walked away from the mis-
sion of creating a one-write system (to
enter data once then share it across the
wider system) in preference to their
own system.
In retrospect, this stakeholder may
not have understood its own needs or,
at a minimum, did not appreciate the
priority of its operational need for a
record management system rather than
a case tracking system. The difference
between the two systems is not neces-
sarily that significant, but this stake-
holder realized that it is investing in a
system that would address only 20 per-
cent of its workload, whereas 80 per-
cent of its workload is never forwarded
to one of the other prime stakeholder
agencies. This reduced the credibility
of the vision for that particular stake-
holder because it failed to motivate or
inspire it to be part of an information
system in which it may be generating
front-end effort for data and informa-
tion entry and yet gaining no tangible
benefits or rewards. This attitude is
typical in many processes where infor-
mation that one party can easily gain is
not ”paid for,” so that it makes little
sense to that group (based on a trans-
actional approach) to expend scarce
resources to gather data or information
that others along the supply chain may
benefit from. In this sense, the remedy
may be argued to lie in the system
rewards and resourcing design, and
not through demonizing or criticizing
a particular stakeholder for being ”silo-
minded” or selfish.
Attempts to influence continued
participation were unsuccessful, as no
structure of dispute resolution was in
place or contemplated in the leader-
ship terms of reference. Although
most agencies indicated they were
committed to the project, an inspec-
tion of project documents finds little
proof of action on this commitment.
An essential element of the founda-
tion necessary to support project suc-
cess was missing.
Apart from the overall vision
issue, there is always the seemingly
intractable problem of stakeholder
behavior being more reflective of their
individual agency or department
interests, rather than the good of the
overall project. Endemic leadership
behavior to accommodate these
requests, rather than to challenge
them effectively, meant there were
often questionable corporate benefits
for some aspects of the overall project.
Fortuitously, while the vision had
gaps, there was considerable docu-
mentation of end benefits to support
continued project funding.
Unfortunately, the dubious link
between these end benefits (stated pri-
marily to achieve funding) were not
entwined with the vision, creating
endless tension around internal proj-
ect decision-making.
At the time, project leadership was
clearly being driven by the central sup-
port agency based on technological
requirements to eliminate a number of
different legacy systems. These systems
had become undesirable as they were
costly to maintain, built with different
software applications, unreliable, and
did not interface with each other—yet,
executive sponsorship was not forth-
coming from all of these core agencies.
Again and most notably, one core
stakeholder was fractured in its sup-
port. Lack of executive sponsorship as
identified by Ulfelder (2001, p. 3) is
the number-one way information tech-
nology projects fail; further, “The prob-
lem is that too many business
executives view IT project, . . . as mere
(albeit expensive) technology chal-
lenges.” It is very conceivable that JIMS
was seen as just such an unwelcome
and threatening challenge, rather than
a welcome process improvement
opportunity. Thus, working smarter
with stretch goals to improve the over-
all system was not evident.
 The philosophies of independence
exacerbated the divisive nature of key
stakeholders’ internal agendas. The
project yearned for a transformational
leader (champion) that could trans-
form self-interest into a corporate mis-
sion. However, as the literature
suggests and Bass (1985, pp. 37 and
154) identifies, transformational lead-
ers often appear in a time of crisis. At
one point, the project was at risk of
stalling due to the lack of expected
funds in the next fiscal year. This was
compounded by the withdrawal of
sponsorship by one key and largely
uncommitted stakeholder, but trans-
formational leaders did appear in the
remaining core agencies. Self-interest
was put aside for an aggressive rollout
schedule. Sacrifices, cooperation and
creativity were the call of the day. A
renewed focus on the project vision
inspired all people in the core agencies
to get JIMS implemented to the extent
possible, given the remaining time-
frame. Thus, at this point of inflection,
both motivational and inspirational
vision characteristics were revived, and
the vision became credible again.
The idea of developing an inte-
grated case tracking system was as
much a high-risk endeavor as it was
innovative and untried, and it
involved a large number of stakehold-
ers (and, hence, presented some
intriguing stretch goals). The joint
vision was seemingly adopted in the
very early stages, quite likely before
individual stakeholders fully under-
stood their business needs and before
technological solutions had conceptu-
al specificity. The importance of the
shared vision cannot be underestimat-
ed in its importance to promote proj-
ect success and bind stakeholders to a
firm foundation.
A vision is inextricably linked to
the end benefits and should be clearly
understood at the outset of the project,
but with the foreknowledge that both
the vision and end benefits are likely
to change and evolve throughout the
project life cycle. The stakeholder
group highlighted in this case as being
problematic never took exception with
the vision of JIMS, it just no longer met
their specific needs. Perhaps the vision
or mission could have evolved to
retain their participation. It is possible
that another approach may have been
available but was not employed.
Remenyi and Sherwood-Smith (1998)
have proposed an ”Active Benefits
Realization” approach that is based on
a contingency philosophy, under
which an evolutionary approach is
used to remain focused on the delivery
of end benefits. Stakeholders play a
central co-evolutionary and participa-
tive role through the development of
the system. All stakeholders focus on
end benefits and not just the technolo-
gy that will get them there. Similarly, as
proposed by Ulfelder (2001) when
managing a system with a broad scope,
it may be beneficial to apply a micro-
project concept and allow the concept
of the larger undertaking to take shape
over time. With either proposed
approaches and regardless of their
champions’ intent, there is likely the
risk of losing clarity in the operational-
ization of objectives and the approach
may be somewhat susceptible to vision
creep.
Lastly, the leadership of the cen-
tral project manager was often
believed to be driven by technology
and resource concerns, and not always
the business requirements of the pro-
ject’s stakeholders. In a government
setting, this is not an atypical behav-
ior. This issue was never fully resolved,
but some of the others described in
the foregoing were. Generally, the
project suffered from a lack of project
management experience and knowl-
edge on many fronts, but the adoption
of well-documented best practices
may have addressed or prevented
some of the challenges experienced in
the JIMS project.
While there are many best prac-
tices documented in the current liter-
ature, a few examples will highlight
the value that applying such practices
may have had for the JIMS project.
Practices such as identifying the need
to conduct a stakeholder analysis
(Morris, 1994; Briner et al., 1996;
Cleland, 1999) may have alerted the
central project manager as to the
motivation of some project partici-
pants and created an opportunity to
September 2004 Project Management Journal • 49
better manage the risks that these
stakeholders presented to the project.
Briner et al. (1996) identify three
groups of stakeholders: internal,
external, and customers. Constituents
from all of these groups should be
considered when conducting a stake-
holder analysis. Such an analysis also
makes imminent sense, as the key to
project success is meeting stakehold-
ers’ expectations.
Similarly, the JIMS project adopt-
ed a vision but it was seldom referred
to except to justify funding or when
renewed energy was required in a time
of crisis. As many writers on leadership
agree, vision must be continuously
maintained foremost in the minds of
project team members and active
stakeholders. All too often, the project
vision is disregarded and attention is
misplaced on the business require-
ments or technical specifications of the
system. Similarly, the end benefits are
forgotten and all of the attention is
focused on the technology. For any
project to succeed, this cannot be
allowed to happen.
It is also unlikely that passion will
grow unless there is a strong, core
project team. The JIMS project saw lit-
tle in the way of scheduled, structured,
or prepared team building. Some
cathartic team cohesion occurred as a
result of people being together and
involved in a common purpose, but
this fell far short of what was required.
The term “project team” is actually a
misnomer for the central JIMS project
management group, as they had few
of the qualities necessary for an effec-
tive team as suggested by numerous
authors on teamwork. Katzenbach and
Smith (1993) noted such qualities as
problem solving, task effectiveness,
and maximizing the use of all mem-
bers’ resources to achieve the team’s
purpose. If a larger investment has
been made toward team building, the
opportunities for the creation of a
shared vision and mission may have
increased. Similarly, team building
and the dynamics of group cohesion
would likely have made it more diffi-
cult and less likely that the problemat-
ic stakeholder group would have
walked away from the project.
 Vision Case 1 Case 2 Case 3 Case 4
Characteristics Novo Nordisk Matsushita Taurus JIMS
1 Easily Yes – by all Yes – by Japanese No – by all participants, Yes – but tacitly at first;
understood employees, but different understanding of however, the vision and
No – by “foreign” employees. the core purpose by mission became later
different stakeholders. confused.

(part) (mainly) (mainly)

2 Motivational Yes – at first, many saw the
and inspirational process as intrusive and
greeted it with skepticism
and suspicion; later the
Knowledge Management
advantages were
appreciated and it was
highly supported.
Yes – by Japanese employees
the Japanese identity, history,
and cultural legacy was very
strong, but
No – for “foreign” employees.
No – it appeared that, in
general, the lack of focus,
the massive scope creep,
and difficulties in ever
seeing the light at the end
of the tunnel was a very
large negative.
No—for a crucial stakeholder
who “paid lip service” to be
inspired, but failed to commit
resources and energy.
Probably a system design
fault through failure to fully
consider participant rewards.

(part) (mainly)

3 Credible Yes—it made sense to Yes—for older and more
employees as a sound way senior Japanese
forward. No—for younger Japanese
and foreign employees due
to not “walking-the-talk”
No—it appeared that the
general confusion resulted
in unrealistic expectations
and ideas of what could be
achieved.
Yes—but one principle
stakeholder did not share the
belief in the project beyond
its own needs.

(part) (mainly) (part)

4 Working smarter, Yes—best practices
stretch goals identified and welcomed for
widespread uptake.
Yes—best practices within
Japan
No—for outside Japan, due
to the corporate dogma that
best practice can only come
out of HQ in Japan.
No—the chaos and
confusion accompanying an
ever-widening scope did
stretch people, but not to do
things smarter—there was
a lot of re-work and
abandoned work.
Yes—the concept for the
integration across the supply
chain of information to
provide superior service
delivery was recognized.

(part) (mainly)

Overall Rating Best practice vision A good vision for Japanese A weak and enfeebled A strong vision, but weakly
but not for outside Japan vision that failed to focus implemented at first through
the project. lack of central authority. A
strong transformational
leadership input emerged to
retrieve the situation.

Table 1. Vision characteristics from example cited in the literature

Lastly, project managers need to
have authority over their domains
(Briner et al., 1996). In JIMS the central
project manager did not have such
authority nor did many of the core
agency project managers. The central
project manager was not given the struc-
ture to control or manage the core
agency project managers or their indi-
vidual processes as is highly recom-
mended by (Sauer et al. 2001). This lack
of clear project management authority
in a single individual may have con-
tributed to the dissention of stakeholder
buy-in and general project ownership.
Any project, but especially a project
with multiple stakeholders who have
diverse interests, needs to have a shared
and prominent vision fostered by a rec-
ognized and empowered project manag-
er. This case study investigation
exemplifies a major need—it suggests
more attention should be paid to stake-
holder analysis and project-leader
behavior in the promotion and adop-
tion of a shared vision. In addition to
the impacts attributed to stakeholder
characteristics and project leader behav-
ior, we suggest that the organizational
environment in which a project exists
also impacts project vision. However,
more investigation is recommended.
While alternative approaches may be
required for different types of projects,
what does seem clear is that it is in every-
body’s interest to promote and sustain a
shared vision and to support leadership
behaviors that will encourage and
endorse this approach.
It should be noted, however, that
the project (despite not fully meeting all
the needs of all stakeholders) was
judged highly successful by the remain-
ing core stakeholders and the sponsor-
ing government. . The system does not
fully meet the anticipated functionality
for all stakeholders, but it was judged by
its peer-group developers from else-
where in the continent to have superior
performance to other similar projects at
a lower cost and in a shorter develop-
ment and delivery time. This was
believed to be due in no small part to
the vision being held together and
maintained despite setbacks and its
birth as a ”whitespace” project in which
substantial stakeholder commitment
and belief in its value was initially
placed.
Discussion and Conclusions
This paper has explored the role of a
project vision as a critical success factor

50 • Project Management Journal September 2004

to project outcomes. We have investi-
gated the concept of vision from a
number of perspectives. We defined it
as it occurs in the common dictionary
and also refined this definition in
terms of the business literature as
vision appears as a corporate tool. We
then extended our discussion of this
concept to apply to a project vision.
We identified four characteristics that a
vision should possess: it must be
understood; it must be motivational; it
must be credible; and it must be both
demanding and challenging. We then
linked vision to a transformational
leadership style, arguing that as proj-
ects are transformations and that as
vision is an idealization of a transfor-
mation, then transformational leader-
ship should be evident in successful
projects.
We then continued to explore four
case studies. The first three drawn from
the literature were chosen to provide
one successful example, one partially
successful example (from the organiza-
tion’s home country but not from its
foreign subsidiaries’ point of view),
and a spectacularly unsuccessful proj-
ect example. Our fourth case study was
discussed in some detail and related to
an interesting IT project that one of the
authors had an intimate involvement
with, and deep insights and sources of
information on.
In comparing and contrasting the
three examples from the literature and
the case study experienced by one of
the authors, the following observa-
tions and analysis can be made.
Table 1 provides a four-point vision
characteristic appreciation that helps us
to understand the important elements
of a vision that contributes to its suc-
cess. Clearly, not all elements need to
fully active constantly for project suc-
cess to be an outcome; however, they
should be present at some period dur-
ing the project’s execution and prefer-
ably, all present all the time. These
categories also matched well with a
transformational management style
and, as was shown, do so in terms of the
four I’s: individual consideration; intel-
lectual stimulation; idealized influence;
and inspirational motivation. Thus, we
argue for a link between development
of an effective project vision and trans-
formational management.
We also indicated how a project
vision could be developed, and
stressed that while a charismatic leader
may provide much energy and useful
motivational impulse to the develop-
ment of an effective vision, a stake-
holder reference group may form a
more enduring framework for vision
development and planned deploy-
ment. This is largely due to a more
inclusive and more diverse set of expe-
riences to be drawn upon in formulat-
ing both the vision itself and a
deployment strategy.
Our intended contribution is to
highlight best practice in vision devel-
opment and to a lesser extent, its
deployment. Our chosen case studies
provided useful evidence for our
analysis and the JIMS case provided an
in-depth example for us to draw upon.
References
Avolio, B. (1996). What’s All the
Karping About Down Under?
Leadership research and Practice. W. Parry
K. W. South Melbourne: Pitman
Publishing.
Bass, B. M. (1985). Leadership and
Performance Beyond Expectations, New
York: Free Press.
Bass, B. M., & Avolio, B. J. (1994).
Improving Organisational Effectiveness
Through Transformational Leadership.
London: Sage.
Bennett, J., & Jayes, S. (1995).
Trusting the Team. Reading, UK: Centre
for Strategic Studies in Construction,
The University of Reading.
Bennis, W., & Nanus, B. (1997).
Leaders Strategies for Taking Charge.
New York: Harper Business.
Briner, W., Hastings, C., & Geddes,
M. (1996). Project Leadership.
Aldershot, UK: Gower.
Carless, S. A., Mann, L., & Wearing,
A. J. (1996). Transformational
Leadership and Teams. Leadership
research and practice. W. Parry K. W.
South Melbourne: Pitman Publishing.
CII (1996). The Partnering
Process - Its Benefits,
Implementation, and Measurement,
Austin, Texas, CII, Bureau of
Engineering Resources, University of
Texas at Austin, CII Source Document
102-11.
Cleland, D. I. (1999). Project
Management Strategic Design and
Implementation. Singapore: McGraw-
Hill.
Collins, J., & Porras, J. I. (1996).
“Building Your Company’s Vision.”
Harvard Business Review. 74(5).
Davenport, T. H., & Prusak, L.
(2000). Working Knowledge - How
Organizations Manage What They Know.
Boston, Harvard Business School
Press.
Dinsmore, P. C. (1999). Winning
in Business With Enterprise Project
Management. New York: American
Management Association.
Doz, Y. L., & Hamel, G. (1998).
Alliance Advantage - The Art of Creating
Value Through Partnering. Boston,
Harvard Business School Press.
Drummond, H. (1998). Riding a
Tiger: Some Lessons of Taurus.
Management Decision. 36(3), 141-146.
Hersey, P., Blanchard, K., &
Johnson, D. E. (1996). Management of
Organizational Behaviour. London:
Prentice Hall International.
Holden, N. J. (2002). Cross-
Cultural Management - A Knowledge
Perspective. Harlow, UK: Pearson
Education.
Katzenbach, J. R., & Smith, D. K.
(1993). The Wisdom of Teams - Creating
the High-Performance Organization.
Boston: Harvard Business School
Press.
Kerzner, H. (2001). Project
Management- A Systems Approach to
Planning, Scheduling, and Controlling.
Toronto: John Wiley & Sons Inc.
Kotter, J. P. (1995). Leading
change - Why Transformation Efforts
Fail. Harvard Business Review. 73(2),
59-67.
Kotter, J. P. (1998). What Leaders
Really Do. Harvard Business Review on
Leadership. Review H. B. Boston, MA:
Harvard Business School Publishing.
Lenard, D. J., Bowen-James, A.,
Thompson, M., & Anderson, L.
(1996). Partnering - Models for
Success. Adelaide, Australia:
Construction Industry Institute of
Australia.
Lendrum, T. (1998). The Strategic
September 2004 Project Management Journal • 51
Partnering Handbook. Sydney: NC: Project Management Institute. (1993). The Goals and Methods Matrix:
McGraw-Hill. Remenyi, D., & Sherwood-Smith, Coping with Projects With Ill-defined
Maletz, M. C., & Nohira, N. M. (1998). Business Benefits from Goals and/or Methods of Achieving
(2001). “Managing in the Information Systems Through an Them. International Journal of Project
Whitespaces.” Harvard Business Review. Active Benefits Realization Management. 11(2): 93 -102.
79(2), 102-111. Programme. International Journal of Ulfelder, S. (2001). The Dirty Half-
Project Management. 16(2), 81-98. Dozen, Six Ways IT Projects Fail-and How
Morris, P. W. G. (1994). The
Sauer, C., Liu, L., & Johnston, K. You Can Avoid Them. Darwin
Management of Projects A New Model.
(2001). Where Project Managers are
London: Thomas Telford. Magazine (www.darwinmag.com), Project
Kings. Project Management Journal. 32(4),
Nonaka, I., & Takeuchi, H. Management. (June).
39-49.
(1995). The Knowledge-Creating Walker, D. H. T., & Hampson, K.
Schein, E. H. (1985). Organizational
Company. Oxford, UK: Oxford D. (2002). Procurement Strategies: A
Culture and Leadership. San Francisco:
University Press. Jossey Bass. Relationship Based Approach. Oxford,
Oxford (2001). The New Oxford Schein, E. H. (1993). How can UK: Blackwell Publishing.
Dictionary of English. Oxford, UK: Organizations Learn Faster? Lessons Weick, K. E. (2001). Making Sense
Oxford University Press. from the Green Room. Sloan of the Organization. Oxford, UK:
Pinto, J. K., & Trailer, J. W. Management Review. 34(2), 85-92. Blackwell Publishers.
(1998). Leadership Skills for Project Turner, J. R. (1999). The Handbook of Zaleznik, A. (1998). Managers
Managers. Sylva, NC: Project Project-based Management: Improving the and Leaders: Are They Different?
Management Institute. Processes for Achieving Strategic Objectives. Harvard Business Review on Leadership.
PMI (1996). A Guide to the Project London: McGraw-Hill. Review H. B. Boston: Harvard Business
Management Body of Knowledge. Sylva, Turner, J. R., & Cochrane, R. A. School Publishing.

DALE CHRISTENSON is Acting Executive Director of the Leadership and Learning Centre of the Public Service
Agency for the province of British Columbia. He is currently enrolled in the Doctor of Project Management
program at RMIT University in Melbourne, Australia. His undergraduate and graduate studies at Simon Fraser
University were in the field of Criminology. He also holds a post-graduate certificate in Project Management
from Royal Roads University.

Mr. Christenson’s current areas of research interest are critical success factors of project management, strategic
alignment of projects and organizational direction, and innovative application of project management
methodologies. Mr. Christenson has been responsible for a number of information technology projects. He has
successfully applied project management methodologies to the area of program development and strategic plan
implementation.

Mr. Christenson is also a seasoned lecturer and teaches at Athabasca and Royal Roads University in various
areas of project management. He has over 20 years of lecturing experience in a wide variety of forums and
venues. Mr. Christenson is also a Project Management Professional (PMP®) certificant, through the Project
Management Institute.

DEREK WALKER is Professor of Project Management and Program Director of the Doctor of Project Management
(DPM) program at the Faculty of Business, RMIT University. He worked in various project management roles in
the UK, Canada, and Australia for 16 years before commencing his academic career in 1986. He obtained a
Master of Science from the University of Aston (Birmingham) in 1978, and a PhD in 1995 from RMIT University
(Melbourne). He has written over 100 peer-reviewed papers and book chapters. His research interests center
on innovation diffusion of information and communication technologies, knowledge management, project
management, and project procurement systems.

52 • Project Management Journal September 2004

WHY MONTE CARLO SIMULATIONS OF PROJECT
NETWORKS CAN MISLEAD
ABSTRACT
Monte Carlo simulation of project net-
works is a standard project-modelling
technique. However, much of this analysis
is inadequate, as project managers always
take action to recover late-running proj-
ects, which is ignored in most models.
This paper shows the importance of this
omission (which generally gives unrea-
sonably wide probability distributions)
and discusses simple and easily coded
models of project-management actions.
The paper also notes a second flaw,
explaining why risk-analyses rarely pre-
dict catastrophic overspends that some-
times occur, namely the inability to
capture feedback loops resulting from
chains of causality from management
actions. The need to recognize these as
part of the modelling and then take steps
to avoid them is described.
Keywords: Monte Carlo; simulation; net-
work; management actions
©2004 by the Project Management Institute
Vol. 35, No. 3, 53-61, ISSN 8756-9728/03
TERRY WILLIAMS, Strathclyde University, UK
Department of Management Science, Strathclyde University, 40 George Street, Glasgow, G1 1QE, UK
Introduction
onte Carlo simulation of project networks is now a standard technique
M used by project managers and analysts. It has proved a powerful addition
to the project analysis toolbox; the technique allows modelling complex
combinations of uncertainties to show the risk to project deadlines. However,
there is an important flaw in much of this analysis, as actions taken by the proj-
ect manager to recover any project slippage are omitted from the model entirely.
This paper shows the importance of this omission and discusses simple models
of project-management actions. It goes on to note a second flaw which explains
why risk analyses also are not able to predict the catastrophic overspends that
sometimes engulf seemingly well-run projects. The effect of these flaws it that
results can often mislead project managers for whom the analyses are carried out.
The paper is written from the point of view of the author who was a project
risk manager for a decade before returning to academia a decade ago, where he
still consults on the modelling of major projects. Over these 20 years of experi-
ence, the author has seen these problems and errors in Monte Carlo time-risk
analyses presented to project managers time and time again. This has, in the opin-
ion of the author, either misled the project managers, or the results have been dis-
missed as unbelievable, therefore significantly weakening the credibility of this
valuable method of analysis.
Monte Carlo Time-Risk Analysis
Time analyses of projects are nearly always based on the network (PERT or CPM)
concept (e.g., Moder, 1988). In more recent years, project time-risk analysis using
Monte Carlo simulation, including uncertainty in the networks, has become a
standard technique in the project manager’s toolbox. While CPM served well to
structure a network and provide point estimates for the duration of a project, the
increasing complexity of projects and the increasing availability of computer
power has brought Monte Carlo simulation into common usage.
A Guide to the Project Management Body of Knowledge (PMBOK® Guide)
(Project Management Institute, 2000) describes Monte Carlo analyses of net-
works in Chapter 11, where it states:
Project simulation uses a model that translates the uncertainties specified at
a detailed level into their potential impact on objectives that are expressed at the
level of the total project. Project simulations are typically performed using the
September 2004 Project Management Journal • 53
Monte Carlo technique... For a sched-
ule risk analysis, the Precedence
Diagramming Method (PDM) sched-
ule is used.
In its Glossary, the PMBOK® Guide
defines Path Convergence as:
The node in the schedule where
parallel paths merge or join. At that
node, delays or elongations on any
converging path can delay the project.
In quantitative risk analysis of a sched-
ule, significant risk may occur at this
point.
The UK project management
community is in general agreement.
The Project Risk Analysis and
Management (PRAM) Guide (Simon,
Hillson, & Newland, 1997) says that
“The PERT technique has been
superseded by the more powerful
Monte Carlo simulation modelling
supported by computer-based tools,
and PERT is no longer considered to
be a suitable risk analysis tech-
nique”; in Appendix A3.3, it details
how to undertake such analyses by
simply listing the risks and opportu-
nities for each task, setting up a
Monte Carlo model and populating
it with data, running the simulation
analysis to determine the total dura-
tion of the project, and presenting
the results.
This technique can now be consid-
ered as common usage. It has for some
time been described in texts that could
be considered standard (e.g.,
Wideman, 1992 and Pritchard, 1997).
Extensions to network scheduling soft-
ware that can carry out such analyses
can be found through various sources,
including Palisade Corporation (a),
Primavera Systems Inc., ProjectGear
Inc., and Welcom Software Technology
(see References).
Furthermore, Monte Carlo simula-
tion of networks is considered easy to
carry out. All that is required, says
Cook (2001) is a microcomputer,
scheduling software, Monte Carlo risk
analysis compatible with the schedul-
ing software, and a set of activity-dura-
tion estimates; then, “the technical
expertise to conduct a Monte Carlo
analysis using the above tools is little
more than that required to generally
operate a microcomputer and schedul-
54 • Project Management Journal September 2004
ing or costing software.” The simple
usage described in the PRAM Guide is
taken as read (i.e., accepted without
question) in various PMI publications,
including articles (e.g., Schuyler, 2000)
and conference papers (e.g.,
Houghton, Kindingerm, Gemeinhart,
& Whyte, 2001).
Additionally, there are some
advanced usages. With this technique,
the manager or analyst can include in
the models some of the complexities
that must be included in an analysis
for the results to be both relevant and
credible. According to Williams
(1990), such complexities include:
Effects operating across a range of
activities or resources, such as com-
mon-cause effects
Unusual activity-duration-distri-
butions (e.g., weather windows in
deep-sea oil work)
Uncertain resource availabilities
or requirements that are uncertain
Uncertainties in the project net-
work structure (e.g., branching at proj-
ect choices).
Indeed, the errors that can be gen-
erated by not including one particular
more advanced feature, that of
dependence between activities, is now
becoming better known (e.g., Van
Dorp & Duffey, 1999) and software
packages will normally include a stan-
dard feature to include interactivity
relationships.
The Flaws in Such Analysis
These analyses might have become a
little more sophisticated, but they
still very frequently give results that
are not credible. Often, this is
because they have very wide proba-
bility distributions when project
management knows that the proba-
ble range is much smaller. Even when
the resulting probability distribution
is not particularly wide, management
will often see a fundamental flaw in
the analysis, namely that:
If a project is running late, manage-
ment does not sit idly by, but they do
something!
In other words, the Monte Carlo
simply runs through a network
assigning values to random variables
on each iteration of the simulation,
and ignores the fact that, in reality, at
each point, if the project was running
late, management would take actions
in order to bring the project back on
track.
Simulations rarely consider man-
agement actions, running through
each iteration in a “dumb” fashion.
This gives rise to results indicating
that projects might underrun or very
considerably overrun, whereas, in
reality, management actions would
often bring those overruns forward
even at significant cost. Indeed, many
of the actions management carry out
to catch up a late-running project
cannot be modelled in a naïve net-
work simulation, because they
involve changing the network (e.g.,
by bringing an activity forward to
start it earlier than would otherwise
be sensible, thereby breaking a
dependency; or taking serial activities
and making them parallel). Since a
Monte Carlo network simulation
assumes that the network is fixed and
unchangeable (except in some simu-
lations by branch points that allow
different paths to be taken), such
actions cannot easily be modelled.
The effect of this can be to offer
management an analysis showing
possibilities that management does
not intuitively believe, thus damag-
ing the credibility of what is, apart
from this flaw, a valuable analysis.
But is this omission that important?
And, if so, can management’s
actions be modelled? The answer to
both of these questions is “yes.”
Rather than look at actual examples
used by the author, the following
section looks at a very simple exam-
ple from the literature, and the
modelling has been performed in
@Risk for Excel (Palisade
Corporation (b)) to make the exam-
ple simple to replicate. This indi-
cates both how to model some
simple management actions, and
also the difference in results when
such actions are modelled. More
realistic examples show more strik-
ing differences, while not necessari-
ly requiring more complex models
of management actions.
 Engine Engine Engine
design development production

D/b engine
manufacture

General D/b airframe D/b airframe Assemble d/b Engine/frame Airframe Ready to
design design manufacture aircraft flight trials production assemble

Interim
avionics

Avionics Avionics Avionics Avionics

design test flight trials production

Figure 1. Example network

Example network is actual data from the UK calculations are given here so that the
The example is one that has been used Ministry of Defence (MoD). The exam- analysis can be easily replicated.
in a number of studies, taken from ple is therefore illustrative of how a Figure 1 shows the network of the
Bowers (1994). This is the simple net- Monte Carlo simulation is actually car- project. This shows parallel streams of
work describing the development of a ried out in practice, and the results activities for developing the airframe,
military aircraft; although the data are from a naïve analysis of the program engine, and avionics, as well as their
illustrative (and, in Bowers’ paper, are would be used in practice to inform integration. (“D/b” is an abbreviation
not completely given), the basis of the decision-making. Full details of the for “development batch”, or test-air-
craft, a key element in the develop-
Activity Distribution Triangular distributions Data
ment program). Table 1 shows the data
used to populate this network. As
Min Most Max shown on that table, some of these
Likely data are taken from Bowers; data for
General design Triangular 4 10 21 Bowers the other activities are purely illustra-
Engine design Triangular 21 32 55 Bowers tive. The time-unit throughout is
Avionics design Triangular 1 7 19 Bowers months.
D/b airframe design Triangular 6 15 32 Bowers The use of the triangular distribu-
D/b engine manufacture Triangular 7 9 11 Bowers tion in Bowers is fairly standard, and
Interim avionics Triangular 7 14 27 Bowers this has been extended to some of the
D/b airframe manufacture Triangular 8 11 17 Bowers other activities populated with illustra-
Assemble d/b aircraft Triangular 3 5 10 Bowers tive data. But two other distributions
Engine development Triangular 20 23 40 Bowers have been used for illustrative purpos-
Engine production Triangular 12 13 14 Illustrative es, both available in the @Risk pack-
Avionics test Gamma: mean 10 mode 5 Illustrative age: the gamma distribution for an
Avionics flight trials Discrete: relative probability activity with no obvious upper limit in
1:2:1:1 of 4,5,6,24 (*) Illustrative duration (as in a highly uncertain test
Engine/frame flight trials Discrete: relative probability situation), and a discrete distribution
1:2:2:1:0.5 of 5,6,7,8,13 (*) Illustrative for those two activities marked with a
Airframe production Triangular 12 14 18 Illustrative (*): these two activities illustrate typi-
Avionics production Triangular 14 16 24 Illustrative cal distributions used for trials activi-
Ready to assemble --- --- ties, showing a small probability of
significant problems which would lead
Table 1. Data on activities to considerable lateness.

September 2004 Project Management Journal • 55


Distribution for Ready to assemble
➱ ➱
X 77.??
5%
1 is having the most effect on network
0.8
0.6
0.4
0.2
0
60 80
Figure 2. Duration of project (unmitigated)
If each activity took exactly as long
as its “Most Likely” value (or “mode”),
then the project would take 78.5
months. For simplicity, let us assume
for the purposes of this example that
the target duration of the project is this
value plus a contingency of, say, 6
months to allow for some activities to
fall over the “Most Likely” value.
Now the example can be readily
modelled in @Risk. The mean dura-
tion for the project is found to be 90.5
months, and the 90% point (i.e., the
time by which you would be 90% sure
of finishing the project) is 103 months,
that is, 18 months late. The probabili-
ty distribution of the duration of the
project is shown in the @Risk printout
given in Figure 2, and this is what a
standard Monte Carlo simulation
analysis would give.
The analysis can also provide the
most critical and crucial activities (i.e.,
Activity Criticality Cruciality
General design 1 0.53
Engine design 0.95
D/b engine manufacture 0.48
Assemble d/b aircraft 0.51
Engine development 0.49
Engine production 0.49
Engine/frame flight trials 0.51
Avionics production 0.50
Table 2. Each activity bears its own criticality
and cruciality ratings
56 • Project Management Journal September 2004
X 05.??
9%
100 120
Months
respectively, the probability of activi-
ties falling on the critical path, and the
correlation between activity duration
and project duration (see Williams,
1993)). This gives management guid-
ance as to the activities most likely to
be on the critical path (and, thus,
whose probable duration needs most
careful management), and those activ-
ities whose risk most affects the project
duration (and, thus, whose uncertainty
in duration needs most careful man-
agement) (see Table 2).
But let us imagine being in this
project. If the project started to fall
behind, what would management do?
And most importantly, what would
the effects of the actions be? It would
not be sensible to say “if the project
was late, then management would
expedite and thus stop it being late” –
otherwise, the simulation would sim-
ply model the project finishing on
time. But, clearly, management would-
n’t sit back and allow the project to be
18 months late without doing some-
thing. Some of the high-criticality
0.63
activities in Table 2 might not be
0.11
amenable to acceleration (e.g., the tri-
0.09
als activity). But typical management
0.36
actions that could be taken, which we
0.05
can model easily, could be the follow-
0.20
ing illustrative examples, where we
0.15
will be looking at some of the high-
criticality or high-cruciality activities
shown in Table 2.
First action: expedite an activity.
Let us consider the activity which is
95% critical and the most crucial (i.e.,
risk): the “engine design” activity. As is
often typical in design activities, this
has a significant tail in distribution
(although note that Bowers describes
this as “Medium Uncertainty”), with a
maximum duration of 55 months. But
the activity was planned to take 32
months (the “Most Likely” value) and
it is on the critical path; thus, if the
activity goes beyond 33 months, then
management will certainly be aware
that the project is already running late.
Certainly they would make every
effort to avoid a duration of 55
months, which would put in a 23-
month delay right at the start of the
project!
So, let us suppose that if the length
of the activity is going to be beyond 33
months, then management will expedite
the activity, probably by adding extra
resources to the activity. For simplicity,
let us suppose that management can
reduce the overrun beyond 34 months
by 2/3 (so a duration of 43 months
without management action can be
reduced to 34 + 0.33*(43-34) = 37
months after management action).
But this type of action will have disad-
vantages as well as benefits. As suggested
by Eden, Williams, Ackermann, &
Howick (2000), typical disadvantages
would include:
• A considerable increase in cost:
Let us suppose, for example,
that the additional cost would
be zero if no mitigation takes
place (i.e., expected duration is
< 34 months), increasing linear-
ly to $10 million if the duration
without mitigation would have
been the maximum, 55 months.
• Increased downstream quality
issues, which would increase the
duration of later activities: For
example, let us suppose that the
“d/b engine manufacture” and
the “engine development” activ-
ities are each increased by a
duration which will be zero if
no mitigation takes place (i.e.,
expected duration is < 34
months), increasing linearly to
 a maximum of two months.
Second action: increased paral-
lelism. Again, the network shows that
“engine production” should not begin
until “engine development” is com-
plete. But, if delays have occurred, the
reaction of management would, in all
probability, be to initiate production
before the development has complet-
ed, with early production and the final
stages of development occurring con-
currently. For this example, suppose
that engine production would normal-
ly wait for engine development to
complete, but if the flight trials have
been completed for at least a month,
and engine development is not expect-
ed (without mitigation) to complete
before time 80 (when management
would already be predicting nine
months lateness in the project), then
parallelism will be initiated to allow
the period of time before engine pro-
duction can be started, to halve. This is
a very common type of action,
although well-known downstream
problems can arise (Eden et al., 2000;
Williams, Eden, Ackermann, & Tait,
1995). Such problems typically
include:
• Extra cost to production: Let us
suppose, for this example, that
the extra cost in millions of dol-
lars is equal to the number of
months the engine production
start is brought forward.
• Extra duration to production as
the production becomes more
difficult: Let us suppose, for this
example, that the extra duration is
10% of the period the engine pro-
duction start is brought forward.
Third action: temporary work-
arounds. The network shows that the
first production aircraft cannot be
assembled until engine, airframe and
avionics production are complete. But
the avionics and engine production are
both around 50% critical—and while
we have done something in our first
two actions about the engine develop-
ment line, the avionics production is,
in fact, more crucial (and, thus, has
more uncertainty affecting the net-
work) than engine production.
However, it is sometimes the case that
in large projects, the first production
item might be produced with a tempo-
rary electronic system that is not fully
functional. In this case, let us suppose
that if “avionics production” is going
to finish more than one month
beyond the completion of “engine pro-
duction” and “airframe production,”
and that this will take the final ready-
to-assemble point beyond 85 months,
then management will decide to use a
temporary avionics kit, which will cut
in half the extra duration beyond the
completion of engine and airframe.
The disadvantages of this action do not
appear within this part of the network,
but the likelihood of this action needs
to be estimated.
The foregoing paragraphs illus-
trate three typical types of action that
management could take. As can be
seen, each can be modelled easily in an
@Risk / Excel format. But does this
extra layer of complexity change the
results? Well, running the amended
example gives a mean duration for the
project of 87.2 months (only three
months earlier than above), but a 90%
point (i.e., the time by which you
would be 90% sure of finishing the
project) of 95 months, that is, some 10
months late, but almost cutting in half
the 18 months’ lateness of the original
example. The probability distribution
of the duration of the project as com-
pared to the non-expedited version is
1.2
1
0.8
Cumulative probability
0.6
0.4
0.2
0
65 70
Figure 3. Duration of project (comparing the two simulation runs)
shown in Figure 3, which clearly shows
the effect of the mitigation in bringing
forward the “tail” in distribution as
management reacts to the project run-
ning late.
However, as we have said, this
extra control comes at a cost, which for
this example is twofold:
• There are extra costs quoted in
the above, which result in a
mean extra cost of $2.7 million,
with a distribution shown in
Figure 4 (note that there is a
considerable probability of zero
extra cost) and a 90 percentile
extra cost of $8.2 million.
• The temporary fix for avionics
was utilized on 28% of occa-
sions, so this is the frequency
with which the first production
aircraft will have non-fully-func-
tional avionics; this might be
too high, in which case the
management decision-rule
would need to be adjusted in
the analysis above, which itself
will give useful guidance to
project management.
Discussion of First Conclusion
We have seen that time-risk analyses
that assume no mitigation actions
from project management are not real-
istic, that modelling some such actions
can be done straightforwardly, and
that the results are significantly differ-
Time to completion of project Without mitigation
With mitigation
75 80 85 90 95 100 105 110 115 120 125 130
Months
September 2004 Project Management Journal • 57
ent when these actions are included.
We have also noted that both manage-
ment actions and the consequences of
these actions must be modelled. In
particular, probability distributions
that are often significantly narrowed as
very long project overruns would not
be tolerated by management. So, sim-
ple Monte Carlo simulations can be
overly pessimistic, in terms of dura-
tion. Is that, then, the end of the story?
Clearly many simulations do give out-
puts that are distributionally too
broad, but we also know that simula-
tions rarely capture the catastrophic
overruns that occasionally overwhelm
a project. Sometimes, of course, these
are due to unforeseen and unforesee-
able circumstances. But, management
actions taken in response to lateness
can magnify these catastrophic over-
runs, as shall be seen in the following
section.
Management Action and Feedback
As discussed above, management
actions designed to recover slippage in
a project generally have disadvantages
as well as benefits. In the previous
paragraphs, we have discussed how to
model the immediate effects of each
action. We can extend this to consider
the secondary effects of those immedi-
ate effects, and in principle we can con-
tinue with tertiary effects and so on.
However, we start to set up causal
1
0.8
0.6
Probability
0.4
0.2
0 gives other effects, in particular
0 4
$M extra cost
Figure 4. Distribution of extra cost
58 • Project Management Journal September 2004
chains of effects, which are sometimes
difficult to foresee when taking the
actions. A number of researchers have
been studying these explanations for
project behavior, in particular those at
Strathclyde University (e.g., Eden et al.,
2000, Ackermann, Eden & Williams,
1997, Williams et al., 1995) and PA
Consulting (e.g., Cooper, 1980 and
1993, Lyneis, Cooper, & Els, 2001).
This body of work explains the behav-
ior of large complex projects by the
dynamic behavior resulting from
causal chains of effects following per-
turbations to the original project plan,
and, in particular, when the dynamics
which are set up turn into positive
feedback loops, or “vicious circles,”
which feed upon themselves to pro-
duce spiralling costs. This is when the
overspent or overrun project runs away
and becomes a catastrophic overspend.
And a major contributor to these feed-
back effects (and thus, badly failed
projects) is often the management
action taken to accelerate a time-con-
strained project after a delay.
Let us look at our example project
above for potential occurrences of
feedback. The feedback is perhaps sim-
pler to see in the second action, so we
will look at that first.
• Second action: increased paral-
lelism. As indicated in the exam-
ple above, this is an action that
can have considerable chains of
8 12 16
ramified consequences (see e.g.,
Williams et al., 1995). Parallel
design of cross-related items can
increase difficulty in providing a
system freeze. In a situation
where the project is behind
schedule anyway, this forces
work on items for which the sur-
rounding system is not yet
frozen, and the design of such
items will have to be reworked
when there are changes in the as-
yet-unfrozen surrounding sys-
tem. Thus, rework increases and
delay is exacerbated, so that
management is forced to utilize
even more parallelism, and a
feedback loop is set up, as
shown in Figure 5. There are, of
course, other feedback loops
also caused (Williams et al.,
1995). When parallelism is
between design and manufactur-
ing, design activities finish later
and thus increase concurrency,
and also items begin manufac-
ture and are then changed,
which leads to retrofit, degrada-
tion of manufacture learning
and products are no longer
Designed For Manufacture (a
key element of Concurrent
Engineering).
• First action: expediting an activity,
say by increasing manpower. This
is another action with well-
known feedback consequences as
expounded in Cooper’s “$2,000
hour” (Cooper, 1994). First of all,
of course, having more workers
than the natural limit produces
overcrowding. But beyond this,
there is generally a
communication overhead as the
team size increases, which is the
basis of Brooks (1978):
“oversimplifying outrageously, we
state Brooks’ Law: adding
manpower to a late software
project makes it later.”
Furthermore, Eden et al. (2000)
first, that new staff will be less
productive until they are
acquainted with the company
and with the project – and
indeed, during this time existing

Increase
Delay
parallelism
Rework
Figure 5. One feedback loop from second action
staff will become less productive
as they take time to train the new
staff – and second, that as the
workforce changes, the project
team is “churned”: staff are
reallocated from one job to
another, they lose the value of the
learning they have gained and
begin a task with less expertise
and thus lower productivity.
Figure 6 shows these effects and
some of their relationships.
• Third action: temporary work-
arounds. Producing temporary
production items will enable
first-off items to be produced
earlier. However, this will again
have downstream effects, some
fairly obvious – such as the need
to later retrofit the full-specifica-
tion equipment into the earlier
Difficulty providing
a system freeze
Work on items for
which system
not frozen
partially fitted items (thus using
up precious resources and
maybe making other items even
later) – but some maybe less
obvious, such as the loss of com-
monality between the first few
items produced and thus the
loss of worker learning (so again
using more resources), or the
inability to prove the production
process so the normal industrial
engineering process cannot pro-
ceed to reduce standard hours.
The body of work from
Strathclyde and PA discussed above
show how such effects can set up
causal chains which loop back upon
themselves and set up positive feed-
back, causing manpower spend to spi-
ral out of control. Furthermore, it is
well known that the human brain
finds feedback difficult to estimate or
model intuitively, so these effects are
often difficult to foresee for the project
manager in the middle of a late-run-
ning project. But more than this,
investigating the effect of these feed-
back chains on the project model
becomes very difficult, because when
these occur the network itself and its
underlying assumptions start to break
down (Eden et al., 2000). CPM, PERT
and similar network-methods assume
a forward flow of activity and are not
amenable to analyzing systems with
return flows. There is an analysis
method, known as System Dynamics
(SD), and it is this which underlies
nearly all of the Strathclyde and PA
work, and indeed a lot of analysis of
project behavior (see a review in
Rodrigues and Bowers, 1996). By
including systems of causal chains
within the modelling, including the
“softer” effects which are sometimes
less amenable to quantification
(morale, communication, etc), SD can
be used to give interesting, and some-
times somewhat counter-intuitive,
insights into the behavior of large
projects. However, while SD model-
ling takes an overall, systemic view of

Overcrowding

Churning staff, so Lower productivity

reallocating them per person

If effect of extra
Behind Increase Untrained staff manpower is less than
schedule manpower less productive loss of effective
effort, then more delay

Higher Lower proportion of

communication workers’ time spent
overhead on productive work

Trained workers
diverted to train
new workers

Figure 6. Feedback loops from first action

September 2004 Project Management Journal • 59

the whole project, as normally used it
puts aside operational information
such as that contained in the network
and work breakdown structure. Thus
while it has shown interesting insights
in post mortem analyses, SD has been
virtually ignored at the operational
project management level which mon-
itors and controls on-going projects
(although some attempts have been
made at hybrid SD/classical project-
management methods – see e.g.,
Rodrigues & Williams, 1997).
But this misses one of the main
messages of the Strathclyde / PA work.
This body of work analyzes projects
post mortem and establishes that set-
ting up feedback has in projects led to
huge overspends. The object of much
of that work was to explain in detail
and quantitatively why projects had
behaved this way in the past. Our
objective in risk analysis (for this
paper) is to plan what we are going to
do in the future. Now if we under-
stand that a particular set of effects
and management responses would
lead to excessive overspends, then our
reaction is clearly not simply to model
that, but to establish those manage-
ment policies or actions which would
break the feedback loops or turn them
into balancing control loops (rather
than positive feedback loops or
“vicious circles”). For example, we
know that resorting to uncontrolled
parallelism in our design and produc-
tion cycle will lead to the types of
effects described in the “second
action” above. But knowing this, and
understanding the underlying “vicious
circles” (Williams et al., 1995),
enables management to establish
mechanisms and a decision-rule that
will ensure that the project does not
get into such cycles. So the simple
decision-rule that we modelled initial-
ly would not be sufficient, rather we
would need to include operational
considerations such as “parallel pro-
duction will not be initiated until after
design-freeze of subsystems A, B and
C” – but more importantly, we would
need to ensure that these operational
considerations were adhered to if these
circumstances arose in the actual proj-
ect, to avoid feedback being set up. Or
60 • Project Management Journal September 2004
for the first action, we might decide
that “no more than X extra staff will be
employed….key design managers will
be retained as designers not supervising
new staff…the rate of new staff joining
must be less than Z,” and so on. Again,
we need to include this in our model,
but more importantly, the lessons on
the dangers of overmanning must be
learned by operational management.
Rigorous qualitative analysis of the
actions management would be likely to
take in the case of a late-running project
would enable decision-rules to be
defined that would indicate what would
be likely to happen in particular scenar-
ios, both the accelerating action (e.g.,
changing the project network to
increase parallelism) and the disadvan-
tages (e.g., longer integration activities
downstream and increased cost). Where
this analysis suggests a positive feed-
back loop (or “vicious circle”) would be
set up, the project team must take
action to ensure that in practice this
does not happen, then it need not be
modelled. Otherwise, the decision-rule
would need to be quantified, and a
Monte Carlo simulation can be modi-
fied to take this into account. System
Dynamics modelling is complex and
does not fit well with operational proj-
ect management; however, in this way,
the general results of the System
Dynamics project analyses can con-
tribute to project management without
the need for the technique itself to be
used in practice.
Conclusions
We can, therefore, draw a set of con-
clusions as follows:
(i) Conventional Monte Carlo net-
work simulations are inade-
quate, as they do not model the
actions management would
take in response to a late-run-
ning project.
(ii) Results given by conventional
Monte Carlo network simula-
tions will usually, therefore,
give unreasonably wide proba-
bility distributions, which will
have reduced credibility.
(iii) Modelling management
actions within the context of a
Monte Carlo simulation of a
network is quite feasible,
requiring the establishment
and then coding of the deci-
sion-rules, and modelling of
the consequences of these
decisions.
(iv) Management actions have dis-
advantages as well as benefits,
and both must be modelled.
Not only the immediate
effects, but any secondary
effects and so on, must be
modelled.
(v) If a chain of causality found in
(iv) “circles around” to
become a positive feedback
loop, the potential of major
overspends is generated. While
this cannot be modelled easily
within the context of a net-
work model, this is neverthe-
less an unacceptable scenario,
and policies or action-plans
must be established to ensure
that such feedback is not set
up. Flagging this is up to man-
agement and ensuring such
planning is done can, in itself,
be valuable in avoiding major
overspends.
(vi) Conclusion (v) means that
rigorous analysis of the
actions management would
be likely to take in the case of
a late-running project should
enable decision-rules to be
defined that would indicate
what would be likely to hap-
pen in particular scenarios,
and thus a simulation can be
modified to include these.
(vii) Conclusion (ii) above implies
that in general simulations
have been overly pessimistic in
neglecting management recov-
ery actions. However, this is to
some extent offset by the
overoptimism implied by
neglecting chains of causality
and particularly feedback,
which can lead to larger-than-
expected overspends.
(viii) Starting to model the behavior
of project managers can bring
realism, and both usefulness
and credibility, to our Monte
Carlo simulations.
References
Ackermann, F., Eden, C., &
Williams, T. (1997). Modelling for
litigation: Mixing qualitative and
quantitative approaches. Interfaces,
27(2), 48–65.
Bowers, J.A. (1994). Data for
project risk analyses. International
Journal of Project Management, 12(1),
9–16.
Brooks, F.P., Jr. (1978). The
mythical man-month. Reading, MA:
Addison-Wesley.
Cook, M.S. (2001). Real-world
Monte Carlo Analysis. Proceedings of
PMI® Annual Seminars and
Symposium, Nashville, 2001.
Newtown Square, PA: Project
Management Institute.
Cooper, K.G. (1980). Naval ship
production: a claim settled and a
framework built. Interfaces, 10(6),
20–36.
Cooper, K.G. (1993). The rework
cycle: benchmarks for the project
manager. Project Management Journal,
24(1), 17–21.
Cooper, K.G. (1994). The $2,000
hour: how managers influence proj-
ect performance through the rework
cycle. Project Management Journal,
25(1), 11–24.
Eden, C.E., Williams, T.M.,
Ackermann, F.A., & Howick, S.
(2000). On the Nature of Disruption
and Delay (D&D) in major projects.
Journal of the Operational Research
Society, 51(3), 291–300.
Houghton, F.K., Kindingerm, J.P.,
Gemeinhart, D., & Whyte, G. (2001).
LANL’s development of schedule con-
tingency based on probabilistic risk
results. Proceedings of PMI® Annual
Seminars and Symposium, Nashville,
2001. Newtown Square, PA: Project
Management Institute.
Lyneis, J.M., Cooper, K.G., & Els,
S.A. (2001). Strategic Management of
Complex Projects: A Case Study
Using System Dynamics. System
Dynamics Review, 17(3), 237–260.
Moder, J. (1988). Network tech-
niques in project management. In
D.I. Cleland & W.R. King (Eds.),
Project Management Handbook, 2nd
Edition. New York: Van Nostrand
Reinhold, pp. 324–373.
Palisade Corporation (multiple
years, a). @Risk for Project.
Newfield, NY: Palisade Corporation.
Palisade Corporation (multiple
years, b). @Risk for Excel. Newfield,
NY: Palisade Corporation.
Primavera Systems, Inc. (multi-
ple years). Monte Carlo for
Primavera: Project risk analysis soft-
ware. See www.primavera.com/prod-
ucts
Pritchard, C.L. (1997). Risk
Management: Concepts and Guidance.
Arlington, VA: ESI International.
Project Management Institute
(2000). A Guide to the Project
Management Body of Knowledge.
Newtown Square, PA: Project
Management Institute.
ProjectGear Inc. (multiple years).
Risk+ for Microsoft Project. Tacoma,
WA: ProjectGear Inc.
Rodrigues, A. & Bowers, J.
(1996). The role of system dynamics
in project management. International
Journal of Project Management, 14(4),
213–220.
Rodrigues, A. & Williams, T.M.
(1997). Systems Dynamics in soft-
ware project management: towards
the development of a formal integrat-
ed framework. European Journal of
Information Systems, 6(1), 51–66
Schuyler J. (Sept. 2000) Project
Risk Management by the numbers.
PM Network, 69–73.
Simon P., Hillson, D., &
Newland, K. (1997). PRAM: Project
Risk Analysis and Management Guide.
Norwich, UK: APM Group Ltd.
Van Dorp, J.R., & Duffey, M.R.
(1999). Statistical dependence in risk
analysis for project networks using
Monte Carlo methods. International
Journal of Production Economics,
58(1), 17–29.
Welcom Software Technology (multiple
years). Guide to OPERA (Open Plan
Extension for Risk Analysis). See
http://www.wst.com/products/opera.thml
Wideman, R.M. (1992). (Ed.)
Project and Program Risk Management.
Newtown Square, PA: Project
Management Institute.
Williams, T.M. (1990). Risk
Analysis using an embedded CPA pack-
age. International Journal of Project
Management, 8(2), 84–88.
Williams, T.M., (1993). What is
critical? International Journal of Project
Management, 11(4), 197–200.
Williams, T.M., Eden, C.L.,
Ackermann, F.R., & Tait, A. (1995).
Vicious circles of parallelism.
International Journal of Project
Management, 13(3), 151–155.

TERRY WILLIAMS worked as an operational researcher for nine years at Engineering Consultants YARD (now
within BAE SYSTEMS), developing project risk management work, and later acting as Risk Manager for major
projects. He rejoined Strathclyde University in 1992 and is now professor of operational research and head
of the Management Science Department. He continues research and independent consultancy modelling
the behavior of major projects. He is one of a team that has supported major post-project claims, particularly
delay and disruption, totalling over $1.5 billion in Europe and North America. He continues to advise on
project risk, and feels it important to use real post-mortem analysis to help guide future projects.

Terry is a speaker on project management, and has written many articles and a book on modelling complex
projects. He is a PMP, PhD, Chartered Mathematician, Fellow of two Institutes, and edits a learned journal.

September 2004 Project Management Journal • 61

 COVER TO COVER
Kenneth H. Rose, PMP, Book Review Editor
EFFECTIVE RISK MANAGEMENT: SOME KEYS TO SUCCESS,
SECOND EDITION
BY EDMUND H. CONROW
dmund Conrow’s Effective Risk
E Management, Second Edition is a highly
original work. As you read it, you have
a strong sense that the author really knows
what he is talking about. He is not afraid to
express his views on different approaches to
risk management, even when these views go
against today’s common wisdom. For exam-
ple, he is firm in his belief that risk manage-
ment should focus on untoward events and
their consequences, and is critical of current attempts by standard-
setting bodies (including PMI) to define risk both in terms of
untoward events and opportunities. Whether you agree or dis-
agree with his perspective, he argues his points well.
As Conrow acknowledges, this book is not for the risk man-
agement novice. While it is filled with hundreds of tips on good
risk management practice, it is not a primer. Many of the tips –
particularly those that deal with the technical aspects of risk man-
agement – are valuable to the experienced risk management prac-
titioner and address issues not covered in other risk management
books. For example, he proffers substantial useful advice on how
to deal with subjective and ordinal approaches to handling prob-
abilities when making risk assessments. These are practical matters
that all experienced risk managers encounter, yet they are rarely
discussed in risk management texts.
Conrow’s book is most useful for risk management profes-
sionals working in the defense sector. His risk management expe-
rience comes primarily from dealing with large defense programs,
such as projects associated with the Strategic Defense Initiative. He
acknowledges that the risk management approach that has influ-
enced him most heavily is the approach found in the Defense
Acquisition University publication titled Risk Management Guide
for DoD Acquisition. The downside of this focus on defense-orient-
ed risk issues is that people working on smaller, private sector
projects may find much of the discussion to be over their heads or
of limited relevance to them on their projects.
I found the eleven appendices to be among the best I have
encountered in any book. Appendix I, “Example of Technical Risk
Analysis,” does a great job of demonstrating how risk manage-
ment principles can be applied in a real world example. Appendix
G, “Sample Language for Including Risk Management in an RFP”
offers practical advice on how RFPs can include risk management
provisions. The other appendices are as informative and practical
62 • Project Management Journal September 2004
as these. Each is a must read.
Regrettably, the book has a number of shortcomings that can
be laid at the feet of its editors. One of the most important func-
tions of book editors is to keep things in perspective for authors
and to make sure that writing style and grammar are flawless.
Sadly, the editors fell short in their duties. I will mention three
major problems that arose as a consequence.
One set of problems is tied to an abundance of grammatical
errors in the text. In particular, there are many instances where
grammatical/spelling errors proliferate. Consider the following
example: “Implemented the risk management process at the start
of the development phase and update it as needed during the
course of the program.”
A second set of problems is reflected in the lack of balance
in chapter length and focus. Most chapters are quite short, rang-
ing from ten to twenty pages in length. Then there are two giant
chapters, one 140 pages long and the other 160 pages long. This
wild inconsistency in chapter length and focus reflects a design
flaw in the book that the editors could easily have identified and
remedied.
The final shortcoming is that the book is too closely tied to
a set of Defense Department processes that have experienced
two major modifications since the book was written. The most
recent major modification of the process occurred in May 2003,
just after the second edition of Effective Risk Management was
published. In his Preface, Conrow anticipates ongoing changes
to the acquisition process and rightly points out that the basic
risk management principles raised in his book are as valid under
the new system as the old. Still, in order to extend the shelf life
of the book, the editors should have insisted that the book high-
light general acquisition management principles and their risk
implications, rather than focus on specific government regula-
tions and instructions that were bound to change momentarily.
Overall, however, the book is thoughtful and filled with
original insights that you will not encounter in many risk man-
agement texts.
AIAA, 2003, ISBN: 1-56347-581-2, hardcover, 554 pp.,
$71.96 Member, $79.95 Nonmember
Reviewed by J. Davidson Frame, PhD, PMP, Academic
Dean, University of Management and Technology, Rosslyn,
Virginia, USA
PROJECT MANAGER’S PORTABLE HANDBOOK
BY DAVID I. CLELAND AND LEWIS R. IRELAND
aizen is the Japanese word for
K incremental, continuous
improvement. It is a well-
known term describing a concept
critical to substantive quality effort.
With Project Manager’s Portable
Handbook, Second Edition, authors
David I. Cleland and Lewis R. Ireland
show that kaizen is more than a con-
cept. It is a method and approach
demonstrable in practice.
This second edition is a book different and better than
the first. But the authors have not made massive revisions.
Instead, they have fine-tuned an already excellent item, much
as one would a high-performance racecar after a period of
experience on the track. Changes are subtle. Many may be
apparent only after careful, page-by-page comparison—some-
thing readers need not do. Readers benefit from the results of
the update: a better book with reduced redundancies and
tighter logical linkage.
Two improvements are immediately obvious. First, the
book is now consecutively page-numbered, making it more
comfortable for readers and emphasizing the unity of the sep-
arate chapters. Second, the book now contains numerous
summary points highlighted in boxes for emphasis through-
out the text. Professional aphorisms such as, “Working proj-
ects is working with change!” and “Project priority is based on
the contribution to the organization’s goals.” greet readers at
the turn of almost every page. They facilitate learning by
breaking up the prose into more retainable pieces and pro-
viding a memorable means for indexing concepts that are
rather densely packaged.
Other improvements are not so obvious because of the
overall natural, even flow of the text. In Section 1, The
Discipline of Project Management, the previously existent
description of the Project Management Institute has been
replaced with a description of project management compe-
tency, a matter of far more importance to those who man-
age projects.
The authors have also deleted from Section 1 a separate
paragraph on the strategic context of projects by weaving
this information into a description of project management
evolution for a tighter, smoother presentation to readers.
The essential admonition that “Projects are the building
blocks in the design and execution of organizational strate-
gies.” is still there, but now it is part of a more comprehen-
sive, integrated view.
The strategic context of projects remains the subject of
Section 4, which is now expanded to include discussion of
program management and portfolio management, both
emerging issues of importance to project managers. Neither is
a matter to be settled by a simple definition. Cleland and
Ireland describe programs as groupings of related projects
that have a common purpose in supporting the goals of an
organization, eventually supporting organizational strate-
gies. Programs evolve in different ways. The authors discuss
alternatives and hearken back to discussion of the project
management office in Section 2 as a structure for control
and support.
Portfolios are even broader in scope. The authors
emphasize that a project portfolio is the future of enterprise
strategy. And that’s the key: an enterprise view. Portfolio
management involves selecting a mix of projects by subject,
size, and risk—all supporting the strategic objectives and
competencies of the organization. This single issue may be
the greatest added value of the second edition. The authors
provide a model for project mix and examples of project
selection criteria and variables. A closing example of a
notional portfolio of projects shows readers just how the
issue might play out in the real world.
Project leadership was well covered in the previous edi-
tion. The authors have added a subsection on team leadership
that offers a new and illuminating discussion of decisions and
modus operandi of successful leaders.
Section 7, Project Planning and Control, the longest in
the book, has been augmented with new information on
establishing priorities and earned value management. In a
world of “do everything now,” where urgency has almost lost
its exceptional meaning, the authors’ model for a project pri-
ority system is a welcome addition. The discussion of earned
value management is a top-level review of the concept and
process, not—thankfully—a detailed delineation of acronyms
and formulae.
Beyond the technical improvements, this second edi-
tion delivers the same strengths of the first: convenient size
and durability—a truly portable handbook—“bullet” for-
mat that allows quick scanning and assimilation of infor-
mation, section summaries, questions for discussion, and
annotated bibliographies with entries as recent as 2004.
Project Manager’s Portable Handbook is not the El
Dorado of project management. It is not the ever-elusive
one best method that need only be plugged into every
project and applied without deviation for consistent, per-
fect results. It is perhaps the one book that best summa-
rizes a complete knowledge set to be applied in
successfully managing projects. It is the one book that
project managers—as stated in the book’s preface—should
not “go to work” without.
McGraw-Hill, 2004, ISBN: 007143774-6, paperback, 525 pp.,
$56.95 Member, $59.95 Nonmember
Reviewed by Kenneth H. Rose, PMP, Director, Peninsula
Center for Project Management in Hampton, VA, USA
September 2004 Project Management Journal • 63
EFFECTIVE OPPORTUNITY MANAGEMENT FOR PROJECTS:
EXPLOITING POSITIVE RISK
BY DAVID HILLSON
pportunity Management
O has become the topic du
jour under intense debate in
the risk management field, ever
since PMI introduced opportunity
in the definition of risk with release
of A Guide to the Project Management
Body of Knowledge (PMBOK®
Guide)—2000 Edition. To date, no
text has done as good a job of
addressing the understanding of
risk, both threat and opportunity,
as Effective Opportunity Management for Projects, by Dr.
David Hillson.
Early on, Hillson goes to great lengths discussing the
nature of risk, and provides practically every definition of
risk ever published to show the trend toward the definition
of risk encompassing opportunities as well as threats.
The probability-impact image on the cover of the book
clearly illustrates the point – that opportunities are the
opposite of threats, and Hillson advocates that they should
be captured and exploited just the same for maximum
project benefit. The image has been extended to mirror the
typical threat-impact grid shown in most risk management
plans today, whereby allowing the top opportunities to
appear together with the top threats, creating what the
author accurately describes as a “management attention
zone.” For risk practitioners and project managers who
wish to adopt opportunity management, this should be a
relatively straightforward process change. For students who
plan further study in the area, the balanced treatment of
both subjects provides a broad introduction to the most
recent changes in the field of risk management.
In part II, Hillson provides a bridge from the theoreti-
cal to the practical, and clears the fog between threat and
opportunity when he walks through the steps of managing
risk on projects. The combined “mirror probability-impact
grid” shown (the same as the one shown on the cover) may
be improved by somehow incorporating a way to show the
project baseline on the grid.
Another notable contribution to the field of risk man-
agement appears in Hillson’s section on risk response
planning. For organizations that seek to employ the
opportunity aspect of risk management in their risk
process, the book provides tangible applications of incor-
porating opportunity management, a term often used, but
rarely made relevant. The typical threat-based risk respons-
es of avoid, mitigate, transfer, and accept have been
defined as exploit, enhance, share, and accept in planning
opportunity-based risk responses.
64 • Project Management Journal September 2004
The latter part of the book, entitled “Concluding
Considerations,” focuses on key implementation issues
and also explores future developments. Hillson clearly
communicates the many challenges of effective risk man-
agement in projects and provides simple, proven methods
for addressing them. The author’s experience with project
risk management is clearly evident in his practical treat-
ment of the subject.
In addressing future developments, Hillson defines
four levels of risk management capability through an
Organizational Risk Management Maturity Model. Like
any conscientiously implemented process improvement
program, the benefits of operating under a maturity model
come with time. Following the approach presented, along
with a dedicated commitment, should show results as
organizations begin to understand where they are and
where they want to be in terms of managing project risk.
In Chapter 10, Hillson discusses future opportunities
for process improvements, updates to standards and
guidelines, and even a vendor call for tools. He advises
software providers to be proactive in adopting opportu-
nity management within their current threat-based
toolsets, and suggests that those who do will be better
positioned for the future than those who respond later,
or not at all.
If the book has an area for improvement, it is in the
area of risk process integration. To maximize the effec-
tiveness of risk management, the risk management
process should be integrated with other processes within
the project manager’s toolkit. When initiated as a stand-
alone process, it is often abandoned. Perhaps he plans to
do this in a second edition.
Effective Opportunity Management for Projects is an
easy read and a highly useful desk reference for any proj-
ect manager concerned about planning for and mitigat-
ing risk on projects, and has done a great job in
advancing the knowledge base on risk. It provides readers
at all levels of experience with an authoritative, up-to-
date reference on effectively managing risk on any proj-
ect. If your budget limits you to the purchase of one book
on risk management, purchase this one.
Marcel Dekker, 2003, ISBN: 0824748085, Hardcover,
316 pages, $118.75 Member, $125.00 Nonmember
Reviewed by J. Steven Waddell, PMP, Project Manager,
Northrop Grumman, Hampton Roads, VA. Mr. Waddell also
teaches project management courses for the University of
California at Irvine (UCI), and graduate level courses in
project management for The Florida Institute of Technology
MANAGING RISK IN ORGANIZATIONS
BY J. DAVIDSON FRAME
J. Davidson Frame has written an
excellent high-level analysis of
methods, tools, and techniques that
can be used for effective manage-
ment of risks within organizations of
all sizes. The need to actively manage
risk at all levels of the organization is
supported by both case studies and
longitudinal analysis that was con-
ducted by the author. This is an
excellent book both for managers who are new to imple-
menting formal risk management processes within the
organization, and for those seeking confirmation and
direction for existing risk management practices.
The focus of Managing Risk in Organizations is Frame’s
five-step risk management process that is based on the six
risk management processes, as described in A Guide to the
Project Management Body of Knowledge (PMBOK® Guide)—
2000 Edition. The five steps are: identifying risk, qualita-
tive risk analysis, quantitative risk analysis, planning to
handle risk, and monitoring and controlling risk. Frame
notes that although the first allusion to managing risk is
the Code of Hammurabi (18th century B.C.), the use of
mathematics to estimate expectancies is as recent as the
early 1700’s.
Several circumstances lead to poor risk management
processes in organizations. Organizations such as financial
institutions, whose primary product focus requires effec-
tive risk management, often fail to transfer this knowledge
and practice into internal operations. Also, risk manage-
ment is not likely to realize the success of total quality
management (TQM), since only one of the three aspects of
the success of TQM—the existence of an external threat—
is present for risk management. Frame cites innumeracy, or
quantitative illiteracy, as a reason for lack of effective risk
management. The fact that educational systems in indus-
trialized countries no longer emphasize math and science
leads Frame to state that, “many people [including those
with advanced degrees] charged with monitoring and han-
dling risk lack the quantitative skills to engage in quantita-
tive analysis.”
The basic tenets of effective risk management include
the establishment of good operating procedures, organiza-
tional structure, and the use of contracts. An organization’s
ability to function effectively leads to implicit risk man-
agement. Risk can be designed into or out of centralized
and decentralized organizations, depending upon the
players and effective use of the structure.
Explicit measures can be taken to build risk manage-
ment processes into the organization’s business activities,
including project management. The author asserts that a
best practice is to adopt good risk management practices
into the business processes at the organizational level.
These practices can be adapted by the divisions of the com-
pany and within projects. A list of six best practices for
organizational risk management is given.
The author uses clear language and case studies to sug-
gest tools that can be used throughout the organization to
identify risks. Tools include checklists, process/environ-
ment diagrams, brainstorming, fishbone diagrams, and
other techniques. All techniques can be applied to any part
of the business process, including risks in research and
development, product development, manufacturing, and
sales and marketing. A similar approach is used to present
tools and techniques for assessing value and ratings, as
well as methods for managing risks.
Frame points out that much of the process of moni-
toring risk is informal. Frequently, risk monitoring is per-
formed as a part of a person’s job. Three conditions are
presented that must be met for successful surfacing of risk
events. When responding to risk events, flexibility is
imperative, since neither the risk event nor the response
generally occurs as planned.
Although business risk, operational risk, and project
risk are each treated in separate chapters, they are shown to
be integrated into overall risk management processes with-
in an organization. Frequent sources of risk are described
as incompetence, poorly trained workers, inattention, and
poorly maintained or obsolete equipment or software.
When discussing business risk, Frame takes the reader
into interesting territories such as real options, cost of cap-
ital, and risk in the product development cycle. The book
reduces the complex concept of real options into a presen-
tation that may be clearly understood by the student who
is new to the concept. A tight bond is developed between
operational risk and quality in business processes and
product production. Frame demonstrates that merely
focusing on the standard triple constraints of project man-
agement can lead to risk in project success. He shows that
the focus must be on customer satisfaction.
Risk management requires understanding of the environ-
mental context of each situation. Risk managers must accept
that although scenarios rarely transpire as anticipated, a con-
scious process approach to managing risk is necessary in all
organizations. Using the processes and case studies in Managing
Risk in Organizations will assist any risk manager to make better
risk management decisions and implement better processes.
Jossey-Bass, 2003, ISBN: 0-7879-6518-9, hardcover,
288 pp., $45.00
Reviewed by Mike Wakshull, PMP, CISA, Manager of
Information Systems Policies, Standards, and Metrics at
Amgen Inc. He is an editor of the Project Risk Management
chapter in the PMBOK® Guide—2000 Edition and president
of the PMI Los Padres Chapter
September 2004 Project Management Journal • 65
 Guidelines for Project Management Journal Book Reviews
Selecting Books for Review
Project Management Journal welcomes recommendations from project managers and others regarding
books that may be of professional value to fellow PMI associates. Areas of potential interest include:
new ideas about the theory, concepts, and techniques of project management; new approaches to
technology and management; getting business results; competing in today’s complex workplace; and
global changes. Recommendations should include the title, author, and publisher, and a brief state-
ment as to why the book should be considered for review. The Journal will select books for review and
identify a reviewer. Individuals recommending books for review may also volunteer to write the review.
However, individuals should not submit a review before the Journal has selected the book. The Journal
receives many books from publishers and authors and cannot review them all.

Guidelines for Writers

Reviews should begin with a strong, brief opening paragraph that identifies the book and author, and tells
the reader why the book is important. The review should not only describe the content of the book, but also
what the content means; that is, why it is a contribution to the project management body of knowledge.
Reviewers may include the following elements:
■ A summary of key or unique concepts
■ Favorite quote, graphic, chart, etc.
■ Important tips or guidelines
■ New terms or phrases, such as “knowbots” or “teamocracy”
■ Message from the book that should be remembered for future use, or should have been disclosed
years ago.

Reviews should include the book’s strong points and any weak points if this information will be useful
to the reader. Reviews should be written in a conversational style that maintains academic rigor.
Reviewers should avoid use of the first person (“I”) and focus on the book and its contents. Reviewers
should also avoid use of extensive lists as a means of describing or duplicating content. Instead, focus
on what the content means to readers. Reviews should be no longer than 750 words (please use your
computer word count to verify length of the review).

Reviews should include complete publishing information, if possible: title, author(s), publisher (city and
state), year published, ISBN number, total pages, and price in U.S. dollars. The Journal will add any infor-
mation that is not available to reviewers.

Reviews should be prepared using MS Word and may be submitted by e-mail (preferred) or on CD. Submissions
should include the name, title, company, address, phone/fax/e-mail, and brief (one or two sentence) biosketch
of the reviewer. Reviews should be submitted to Natasha Pollard at natasha.pollard@pmi.org

or via mail to:

Project Management Journal
Book Review Editor
c/o PMI Publishing Department
Four Campus Boulevard
Newtown Square, PA 19073

PMI reserves the right to edit all material submitted for publication.

66 • Project Management Journal September 2004

Notes for Authors
SCOPE
Project Management Journal is the professional
journal of the Project Management Institute. The
mission of the Journal is to advance the state of
the art of the knowledge of project and program
management. The Journal presents useful infor-
mation on both theory and practice in the field
of project management. Authors are encouraged
to submit the following types of original manu-
scripts: descriptions of innovative practices; sum-
maries of research results; reviews of current
literature; surveys of current practices; critical
analyses of concepts, theories, or practices; devel-
opments of concepts, theories, or practices;
analyses of failure. Manuscript length should not
exceed 12,000 words. The selection of manu-
scripts for publication is based on the extent to
which they advance the knowledge and under-
standing of project management. PMI neither
approves nor disapproves of any data, claims,
opinions, or conclusions presented.
MANUSCRIPT REVIEW
Project Management Journal uses a double-blind
review process. The first review of every manu-
script is performed by two anonymous referees
(usually members of the Editorial Review
Board). The manuscript is then either accepted,
rejected, or returned to the author for revision
(with reviewer comments furnished to the
author). Revised manuscripts are sent to the
Editor, who makes a final disposition in con-
sultation with the Publisher. The Journal strives
to respond to all authors within three months
of the date the manuscript is received at the
PMI Publishing Department. Accepted manu-
scripts are subject to editorial changes. The
author is solely responsible for all statements
made in the manuscript, including editorial
changes.
ORIGINAL PUBLICATION
It is the policy of PMI to be the sole, original pub-
lisher of manuscripts. Manuscripts that have been
submitted simultaneously to other magazines or
journals will be rejected outright and will not be
reconsidered. Republication of a manuscript,
possibly revised, that has been disseminated via
conference proceedings or newsletter is permitted
if the Editor judges there are significant benefits
to be gained from publication.
COPYRIGHT
Upon acceptance of a manuscript, authors will
be asked to transfer copyright of the article to the
publisher. This transfer will ensure the widest
possible dissemination of information. This
transfer of copyright enables PMI to protect the
copyrighted material for the authors, but does
not relinquish the author’s proprietary rights.
The copyright transfer gives PMI the exclusive
rights to republish or reprint the manuscript in
any form or medium as well as to grant or refuse
permission to third parties to republish all or
parts of the manuscript.
SHORT ITEMS
Short items do not need rigorous academic
scrutiny and are not refereed. Upon receipt, how-
ever, these items become the copyright property
of PMI.
■ Opinion presents thoughtful discussion of
project management issues.
■ Correspondence pertains to the project and
program management profession, including ref-
erences to literature, practice, and scholarship as
well as discussion and replies related to articles
published in the Journal.
■ Book Reviews express opinions about books
related to the project management profession, or
about general management or technical books
that cover topics of particular value to the project
manager.
■ Calendar of Events offers notices of forth-
coming meetings and calls for papers.
SUBMISSIONS
All manuscripts must be submitted electronical-
ly either by e-mail to natasha.pollard@pmi.org
or on CD and sent to: Project Management
Journal, Attn: Natasha Pollard, 4 Campus Blvd.,
Newtown Square, PA 19073 USA. If you submit
your manuscript on CD, please include a print-
out of the manuscript, including all tables and
figures, on 8-1/2 x 11 inch paper, double spaced
throughout, and printed on one side only.
Manuscripts should include the following in the
order listed:
■ A title page that includes the title of the
manuscript and each author’s name, affiliation,
mailing address, and phone, fax, and e-mail
address. Correspondence will be directed only to
the first author listed.
■ An abstract of 100 words or less that out-
lines the purpose, scope and conclusions of the
manuscript, and selected keywords.
■ Text (use headings and no more than two
levels of subheadings). To permit objective
reviews by two referees, the abstract and first
page of the text should not reveal the authors
and/or affiliations, but only the manuscript title.
■ References.
■ Illustrations and Tables. These should be titled,
numbered (in arabic numerals and captions), and
each on a separate sheet, and the preferred location
indicated within the body of the text.
■ Biographical details of each author.
Upon manuscript acceptance, authors must also
provide a black-and-white passport-style photo-
graph and a signed copyright agreement.
COMPUTER-GENERATED
TEXT AND ILLUSTRATIONS
Authors are requested to submit the final text
and illustrations via e-mail or on CD. As with the
requirements for manuscript submission, the
main text, list of references, table and illustration
captions, and author biographies should be
stored in separate text files with clearly identifi-
able file names. Keep the layout of the text as
simple as possible and save text in its original
September 2004 Project Management Journal • 67
applications and/or Rich Text Format (RTF). It is
essential that the name and version of the word
processing program and format of the text files are
clearly indicated (example: Word for Windows
2000 doc). The electronic version should only be
sent with the final accepted version of the paper to
the Editor. NOTE: The hard copy and electronic
files must match exactly.
Upon acceptance of the manuscript for
publishing, authors will also be asked to provide
illustrations placed or embedded within their
chosen word processing program. If this isn’t
possible, please submit illustrations in their
native programs. Be sure to include a hard copy
as well. PMI now recreates all illustrations, fig-
ures and tables electronically for publication. By
doing so, the publication is ensured a consistent
look thoughout. Although this makes electronic
versions of illustrations less important, a hard
copy becomes crucial for re-creation purposes.
Contact PMI’s Publishing Department for fur-
ther details.
STYLE OF TEXT
You should write in clear and concise English.
Spelling should follow Webster’s New World
Dictionary. Authors whose native language is not
English are assured that in-house editorial atten-
tion to their manuscript will improve clarity and
acceptability to readers. For questions regarding
style and format of text, refer to the Publication
Manual of the American Psychological Association,
Fifth Edition.
REFERENCES
For questions regarding reference format, refer
to the Publication Manual of the American
Psychological Association, Fifth Edition.
References used in the text should be identified
by author name and publication date in paren-
theses, e.g., (Cleland & King, 1983), and listed
alphabetically at the end of the manuscript.
Page numbers should be cited for all quota-
tions. Follow the format example shown below:
Baker, Bud. (1993). The project
manager and the media: Some les-
sons from the stealth bomber pro-
gram. Project Management Journal, 24
(3), 11–14.
Cleland, David I., & King,
William R. (1983). Systems analysis
and project management. New York:
McGraw-Hill.
Hartley, John R. (1992).
Concurrent engineering. Cambridge,
MA: Productivity Press.
Please ensure that references are complete, that
they include, where relevant, author’s name,
article or book title, volume and issue number,
publisher, date and page reference.
The use of page footnotes should be kept
to a minimum. Footnotes should be numbered
consecutively and listed at the end of the text
as endnotes.
 KEYWORDS
Keywords categorize your manuscript. They cover project management methodologies and processes, tools and techniques, PMBOK® Guide
knowledge areas, industries, types of projects, geography. Please list three or four keywords that best categorize your manuscript. Choose
from the following list of suggested keywords (this is not a comprehensive list) or you may use your own.

Accounting Multiproject Planning CHECKLIST

Activity Duration Estimating Negotiating ■ Manuscript via e-mail or on CD
Agriculture Networking ■ 100-word abstract
Arrow Diagramming Method New Product Development ■ Illustrations
Baselines Organizational Planning ■ Author biographies
Benchmarking Organizational Structure ■ Black and white author photographs
Benefit/Cost Analysis Parametric Modeling (upon acceptance)
Budgeting Performance Reporting ■ Signed copyright agreement
Change Control Pharmaceuticals (upon acceptance)
Communications Management Procurement Management
REVISIONS
Concurrent Engineering Productivity
Correspondence and files for revision will
Configuration Management Project Life Cycle
be sent to the first-named author unless
Conflict Resolution Project Management Software otherwise indicated. Copyediting of
Constraints Project Plan Development manuscripts is performed by PMI staff.
Construction Quality Assurance The authors are asked to check edited files for
Contingency Planning Quality Management typographical errors and to answer queries
Contract Closeout Reengineering from editors. To improve publication times it
Cost Estimating Resource Planning is important that revised files be returned
Cost Management Responsibility within three days. Excessive turnaround time
Critical Path Risk Management may jeopardize publication of papers.
Delegation Risk Response Development
Deliverables Schedule Development COPIES AND REPRINTS
Design Schedule Control Authors will receive 10 copies of the Journal
Documentation Scope Management free of charge. Additional copies of the Journal
Earned Value Scope Definition and/or article reprints can be ordered at any
Engineering Scope Change Control time from PMI.
Environment Simulation Project Management Institute
Estimating Staff Acquisition Publishing Department
Fast-Tracking Stakeholders 4 Campus Blvd.
Feedback Standards Newtown Square, PA 19073 USA
Finance Statistical Sampling Tel: 610/356-4600 ext. 1135
Float Team Development Fax: 610/355-1633
Funding Time Management E-mail: natasha.pollard@pmi.org
Human Resource Management Tools
Information Systems Training
Integration Management Transportation Variance
Large Project Utilities
Leadership Virtual Organization
Life-cycle Costing Work Breakdown Structure
Manufacturing Work Packages
Management Skills
Matrix Organization
Milestones
Mitigation
Monte Carlo Analysis

Index of Advertisers

PMBOK® Guide–Third Edition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .C2

PMI Leadership/Anniversary Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .C4

68 • Project Management Journal September 2004

CALENDAR OF EVENTS
SEPTEMBER 2004
6-9 September
PMI SeminarsWorld®. London, UK. For more information, visit
SeminarsWorld Homepage
9-11 September
Project Management Technology Forum & Exhibition 2004. Beijing, China.
Managed by the State Administration of Foreign Experts Affairs Centre of
Training. A summary of the development of project management technology
products and education programs in China. The forum and exhibition will
also be a platform to show the needs of better international project manage-
ment technology products from many Chinese enterprises and working
units. For more information, please contact Li Xiao Peng at lixi-
aopeng2104@163.com; xpli1968@sohu.com;
li_xiao_peng145@hotmail.com or telephone +86-10-8855-2995, ext. 2006.
14 September
Special meeting of PMI Olympia, Washington Potential Chapter. Topic:
Procurement in Government and the Benefits of PMI to Government
Project Managers. Speaker: Linda Salac, Deputy Director, Government
SIG. Olympia, Wash., USA. For more information, contact Glenn Briskin,
g.briskin@comcast.net
15 September
IT & Telecom Brazilian Meeting. Seminar about technology and telecom-
munication project experiences. Sponsored by PMI Rio de Janeiro
Chapter. Rio de Janeiro, Brazil. For more information, visit
www.pmirio.org.br
16-17 September
Project Management—Driving for Success. Project management regional
symposium presented by PMI Northeast Florida Chapter. Jacksonville, Fla.,
USA. Tracks include project management project overviews, project man-
agement risk management, and preparing for the Project Management
Professional (PMP®) exam. For more information, visit pmi-nefl.org
18 September
PMI Delaware Valley Chapter presents a Project Management Tools
Exhibition and Seminars on the latest tools & techniques to help project
managers improve cost, schedule and quality. Speakers, demonstra-
tions, vendors & multiple PDU’s available. Philadelphia, Pa., USA. For
more information, visit gomcg.com/pmi/
21-22 September
Second Annual Student Forum on Project Management. Ottawa, Ontario,
Canada. Sponsored by the PMI Ottawa Valley Outaouais Chapter. For more
information, contact Jovica R. Riznic at studentforum@sympatico.ca
21-23 September
2nd Latin America International Government Forum and Project
Management Meeting. Brasilia, Brazil. Sponsored by the PMI Brasilia -
Distrito Federal Chapter and the PMI Government SIG. Special
presentation by the PMI Financial Services SIG. 6 keynote addresses,
18 presentations, 9 workshops and 3 panels. For more information,
contact Vilca Damiani at contato@pmidf
21-24 September
PMI SeminarsWorld®. Hong Kong. For more information, visit
SeminarsWorld Homepage
24 September
Professional Development Day 2004. Sponsored by the PMI Central Iowa
Chapter. Four tracks, three keynote speakers. For more information, visit
www.pmi-centraliowa.org
27-28 September
2004 Project Management Development Days, sponsored by PMI KC Mid
America Chapter. First day is a one-day workshop presented by Dr.
Harold Kerzner. Overland Park, Kan., USA. For more information, visit
www.kcpmichapter.org
28 September – 1 October
PMI SeminarsWorld®. Cincinnati, Ohio, USA. For more information, visit
SeminarsWorld Homepage
OCTOBER 2004
5-8 October
PMI SeminarsWorld®. Denver, Colo., USA. For more information,
visit SeminarsWorld Homepage
12-15 October
PMI SeminarsWorld®. Detroit, Mich., USA. For more information,
visit SeminarsWorld Homepage
13-15 October
“Back to the future”. 10th Anniversary of PMI in New Zealand. Hosted in
Wellington by the PMI New Zealand Chapter, 35 speakers and 350 par-
ticipants are expected will attend. For more information, visit
www.pmi.org.nz/conference/default.asp or contact Warwick Jones at
conference@pmi.org.nz
15 October
PMI Puerto Rico Chapter Seventh Annual Symposium. Theme: “Project
Management: Obtaining Results.” San Juan, Puerto Rico.
For more information, visit www.pmipr.org
18 October
Principles and Practices of Project Management for Engineers and
Technical Professionals. Course for engineers, scientists and project
leaders to learn to manage technological projects more effectively.
Offered by University of Wisconsin-Madison, Department of Engineering
Professional Development, Madison, Wis., USA. For more information,
contact Bruce D. Kieffer at +1-608-262-2624,
kieffer@epd.engr.wisc.edu, or visit
http://epdweb.engr.wisc.edu/WEBF299
20 October
Brazilian Construction Meeting. Seminar about construction project
experiences. Sponsored by PMI Rio de Janeiro Chapter. Rio de Janeiro,
Brazil. For more information, visit www.pmirio.org.br
23-26 October
PMI Global Congress 2004—North America. Anaheim, Calif., USA.
For more information, visit www.pmi.org
27-29 October
3rd International Symposium—Project Management Kazakhstan 2004.
Co-organized by the PMI Kazakhstan, Almaty Potential Chapter. Almaty,
Kazakhstan. For more information, visit www.pmc.kz or contact Vladimir
Zatolokin, Vice President, PMI Almaty, Kazakstan Chapter, at
bmdleasing@nursat.kz
NOVEMBER 2004
9-12 November
PMI SeminarsWorld®. Orlando, Fla., USA. For more information, visit
SeminarsWorld Homepage
10 November
Brazilian Information Systems Meeting. Seminar about systems devel-
opment project experiences. Sponsored by PMI Rio de Janeiro Chapter.
Rio de Janeiro, Brazil. For more information, visit www.pmirio.org.br
12 November
Central Ohio Chapter Professional Development Day. Columbus, Ohio,
USA. Opening Keynote Speaker: internationally acclaimed communica-
tions expert Leil Lowndes. Closing Keynote Speaker: Salt Lake City
Olympics Organizing Committee, winner of the 2003 PMI Project of the
Year Award. Multiple Seminar Tracks. Vendor Exhibits.
For more information, visit www.pmicentralohio.com
14-17 November
16th Annual International Integrated Program Management Conference,
sponsored by PMI College of Performance Management. Tysons Corner,
Va., USA. For more information, visit www.pmi-cpm.org or call
+1-703-370-7885
16-19 November
PMI SeminarsWorld®. Scottsdale, Ariz., USA. For more information, visit
SeminarsWorld Homepage
16-19 November
PMI SeminarsWorld®. New York, N.Y., USA. For more information, visit
SeminarsWorld Homepage
29 November–1 December
PMI Global Congress 2004–Latin America
(5° Congreso Ibero Americano de Gerencia de Proyectos)
Sheraton Buenos Aires Hotel
Buenos Aires, Argentina
PMI 35th Ann.qxd 8/6/04 3:23 PM Page 1

DON’T MISS
C O M I N G I N 4 T H Q U A RT E R
PMI’S BONUS
PUBLICATIONS!

BONUS
35th Anniversary Special
Commemorative Section
This year, the Project Management Institute celebrates its 35th

anniversary. To mark this important milestone, PMI is prepar-

ing a 35th Anniversary Special Commemorative Section that

will be distributed to every PMI member with the October

2004 issue of PM Network. Watch for this special section, which

details efforts to further the discipline of project management.

PMI’s Leadership in Project

Management Annual
The PMI® Leadership in Project Management Annual is a

specially themed publication that will premiere in December

2004. This annual issue will assist project practitioners in

building and sharpening their leadership and people skills.

The PMI Leadership in Project Management Annual will

prove a valuable resource for project practitioners through-

out the world as well as a handsome “keepsake” addition to

members’ libraries. It contains a wealth of ideas, tools and

techniques that project managers can draw upon to articu-

late project management concepts to business, government

and academic officials.