J Syst Sci Syst Eng (Jun 2012) 21(2): 129-143 ISSN: 1004-3756 (Paper) 1861-9576 (Online)

DOI: 10.1007/s11518-012-5190-5 CN11-2983/N

PROJECT MANAGEMENT: RECENT DEVELOPMENTS AND RESEARCH

OPPORTUNITIES

Nicholas G. HALL
Department of Management Sciences Fisher College of Business, The Ohio State University, USA
hall_33@fisher.osu.edu ()

Abstract
This paper studies the business process known as project management. This process has exhibited a
remarkable growth in business interest over the last 15 years, as demonstrated by a 1000% increase in
membership in the Project Management Institute since 1996. This growth is largely attributable to the
emergence of many new diverse business applications that can be successfully managed as projects.
The new applications for project management include IT implementations, research and development,
new product and service development, corporate change management, and software development. The
characteristics of modern projects are typically very different from those of traditional projects such as
construction and engineering, which necessitates the development of new project management
techniques. We discuss these recent practical developments. The history of project management
methodology is reviewed, from CPM and PERT to the influential modern directions of critical chain
project management and agile methods. We identify one important application area for future
methodological change as new product and service development. A list of specific research topics
within project management is discussed. The conclusions suggest the existence of significant research
opportunities within project management.
Keywords: Project management, overview, recent practical developments, opportunities for research

1. Introduction the tasks. These relations typically define

A project is conventionally defined as a constraints that require one task to be completed
“temporary endeavor undertaken to create a before another starts.
unique product or service” (Project Management Compared to many business processes,
Institute 2008). Alternatively, a project can be project management appears to be particularly
thought of as a well defined set of tasks that difficult, from both theoretical and practical
must all be completed in order to meet the perspectives. From a theoretical perspective, the
project's goals (Klastorin 2004). In a typical fundamental planning problem of resource
project, many tasks are performed concurrently constrained scheduling is highly intractable.
with each other. Another key feature of projects From a practical perspective, the two standard
is the existence of precedence relations between objectives in project management are defined to

 Systems Engineering Society of China & Springer-Verlag Berlin Heidelberg 2012

 Hall: Project Management: Recent Developments and Research Opportunities
130 J Syst Sci Syst Eng

be completion of the project on time and on bridges and skyscrapers. This was still the case
budget. Yet, many projects fail to meet these two when project management became formalized in
criteria, despite detailed planning before the 1960s with the help of new computing power.
execution begins and the use of modern project A particularly impressive project management
management software. Further, the failure rate of achievement at that time was the Apollo moon
projects is higher in many modern applications landing project (1961-1969), which required the
than in traditional ones, due to less reliable data coordination of about 410,000 workers at a cost
and the more challenging characteristics that are of $25 billion in 1961 dollars, or $154 billion in
discussed in Section 2 below. Indeed, it can be 2011. Another impressive achievement is the
said that, despite its recent massive growth in organization of the Olympic Games using
use, project management is a difficult to manage project management. This is an example of an
business process. As we discuss, this is creating event project, where the project deadline is fixed
extremely interesting research opportunities. The and cannot be violated.
purposes of this work are to outline what those These successes were achieved for a fairly
opportunities are, and to provide some specific narrow range of applications. However, the
examples. potential for project management to be applied
This paper is organized as follows. In to a much wider set of applications gradually
Section 2, we provide a brief history of project became apparent. Important modern applications
management, document its dramatic recent include implementing a new IT system, research
growth, and discuss the new challenges and development (for example, for
presented by its greatly expanded set of pharmaceuticals), the management of strategic
applications. Section 3 summarizes the organizational change, new product and service
development of methodology for project development, and software development.
management over time, from familiar techniques However, traditional and modern projects
developed in the 1950s to more recent and often have very different characteristics. First,
newly influential ideas. In Section 4, we identify the eventual configuration of traditional projects
many specific research opportunities in project is much more transparent than for modern
management. Section 5 provides a conclusion. projects. For example, construction of a bridge
or skyscraper typically does not start until very
2. History and Growth detailed blueprints have been drawn up. We say
The use of project management as a business that such project management applications are
process goes back a long time. Indeed, the deterministic. By contrast, the exact drug
building of the Egyptian pyramids is believed by formula to be used in a new pharmaceutical is
many to have been assisted by the use of simple typically not known until late in the project, as a
project management principles. For much of the result of testing and regulatory approval.
history of project management, the predominant Similarly, the exact configuration of a software
application type was engineering and code is not known until its last line is written.
construction projects - for example, roads, We say that such project management
Hall: Project Management: Recent Developments and Research Opportunities
J Syst Sci Syst Eng 131

applications are nondeterministic. Not 1. Project management effectively controls

surprisingly, the processes of scheduling and change, allowing organizations to
budgeting the project are considerably more introduce new products, processes and
difficult for nondeterministic projects than for programs.
deterministic ones. A second distinction lies in 2. Projects are becoming more complex,
the difficulty of estimating the amount of work making them more difficult to control
that has been completed so far. While a rough without a formal management structure.
estimate is visually available in the case of a 3. Projects with substantially different
skyscraper, it is typically not available in the characteristics, especially in IT, are
case of a software program. This lack of emerging.
transparency about project progress makes it 4. Project management helps cross-functional
difficult to estimate time and cost variance teams to become more effective.
relative to project progress, and without this 5. Companies are using project management
information is it difficult to allocate resources to develop and test their future leaders.
that protect the performance of the project The dominant professional organization that
relative to its overall schedule and budget. A supports project management is the Project
third defining difference lies in the time pressure Management Institute, which operates branches
under which projects are completed. By their in 180 countries. This organization had a
nature, traditional projects are often of lengthy worldwide membership of less than 10,000 in
duration, whereas modern projects can be much 1980, but this grew substantially to about 50,000
shorter, especially for new products and services. by 1996. Today, PMI has about 550,000
With short product and service life cycles, for members. There are few, if any, other business
example as in the consumer electronics industry, processes that can demonstrate three consecutive
a delay in project completion can mean that a decades of exponential growth in interest, or
product is simply uncompetitive. such a large professional interest level today.
In all the three examples considered above, The rapid growth in project management
the characteristics of modern projects make applications has also created bottlenecks in the
them substantially harder to manage, and job market. For example, in summer 2010, there
especially to complete on time and on budget, were 10,000 unfilled jobs in IT project
than traditional projects. As discussed in Section management in Asia alone.
3 below, this has resulted in the development of
new, alternative project management 3. Development of Methodology
methodologies that can more effectively deal Although project management was not yet
with the difficulties of modern projects. formalized as a business process at the time, a
Several reasons can be identified for the significant development occurred in 1917 when
increasing importance of project management as Henry L. Gantt (1861-1919) invented the Gantt
a business process. Principal among these chart. This chart keeps track of the progress of
reasons are the following. tasks and the allocation of resources to them
 Hall: Project Management: Recent Developments and Research Opportunities
132 J Syst Sci Syst Eng

over time, and is the central tool for visualizing management, the uniqueness of projects implies
project progress when using project that such distributions are rarely available in
management software. practice. Although they can be estimated, it is
The critical path method was developed by difficult to predict the effect of choosing an
DuPont Company in the late 1950s. This method incorrect distribution on the overall project
plans projects without considering either duration estimate. A further problem is that
resources or uncertainty in task times. These companies have been reluctant to implement
simplifications permit the use of a simple Monte Carlo simulation, apparently due to
algorithm that delivers optimal solutions. unfamiliarity with its statistical justifications and
Several project management software packages, a possibly naive satisfaction with their existing
including the market leader Microsoft Project, methodology.
also use this method as a key step in developing In recent years, there have been two
solutions. significant methodological innovations in
In order to model the effect of uncertainty in project management. Critical chain project
task times, the consulting firm Booz Allen management (Goldratt 1997) was developed by
Hamilton developed the Program Evaluation and the influential consultant and business writer
Review Technique (PERT), also in the late Eliyahu M. Goldratt (1947-2011). In order to
1950s. This technique enables estimation of the prevent the dispersion of slack time around the
impact of uncertainty in individual task times on project, where it may become lost due to
the uncertain duration of the overall project. Parkinson’s Law (Parkinson 1958), slack time is
PERT is still used today in many companies. collected into specific buffers. This converts the
However, PERT relies on several strong project management scheduling problem into
statistical assumptions that are difficult to justify one of buffer maintenance and management.
for most projects. As a consequence, the project Since the availability of software, for example
duration estimates obtained from PERT are often ProChain, to support critical chain project
unreliable and on average substantially biased management, many companies have reported
towards the low side. Schonberger (1981) significantly improved project performance as a
documents these problems. Many companies result of using it.
develop their own adjustment factors to account Finally, the notion of agile project
for the bias in PERT, but these factors are management was first popularized by the Agile
difficult to estimate robustly. Manifesto (agilemanifesto.org 2001). Agile
An alternative to PERT is Monte Carlo principles include minimal planning and
simulation. This methodology has been applied documentation, the submission of deliverables in
to project management since the 1960s, and small increments to obtain user feedback, and
avoids the worst problems of PERT. However, it quick response. Agile project management has
makes stringent requirements on available data, so far been influential mainly for the
since it requires knowledge of a probability nondeterministic applications discussed in
distribution for each task time. In project Section 2, especially research and development
Hall: Project Management: Recent Developments and Research Opportunities
J Syst Sci Syst Eng 133

and software development. However, because it have an exact, or even approximate, probability
has some of the features of nondeterministic distribution for the times of the individual tasks.
projects, the application area of new product and Another difficulty is that the uncertain task times
service development offers significant potential may be correlated. For example, if the same
for future expansion (Smith 2007). A challenge resources are used for two different tasks, it is
that remains to be solved is the difficulty of likely that their performance on those tasks will
scaling agile methodology for large projects show a positive correlation. Therefore, crashing
outside of the software development domain. decisions need to be made against a background
of substantial and complex uncertainty. Two
4. Opportunities for Research standard approaches that are widely used to
In this section, we describe a series of make crashing decisions are PERT and Monte
important research topics for project Carlo simulation. However, PERT requires
management, along with various associated problematic and potentially biasing assumptions
research questions. These topics are grouped as (Schonberger 1981), whereas there is
follows. Section 4.1 considers project considerable resistance among project managers
management under uncertainty, Section 4.2 to the use of Monte Carlo simulation (White &
considers contractual issues in project Fortune 2002).
management, and Section 4.3 considers other In these situations, modern robust
emerging topics. optimization techniques (Bertsimas & Sim 2004,
Goh & Sim 2010) may be useful. Such
4.1 Project Management under techniques can be used to develop linear
Uncertainty decision rules that convert information that is
revealed over time into decisions. In this case,
4.1.1 Robust Optimization for Project task times are revealed over time, and that
Scheduling information is converted into crashing decisions.
In projects with tight deadlines, or those that Because these techniques do not make strong
experience delays at the execution stage, a distributional assumptions, their performance is
standard technique is crashing, i.e. expediting, typically much more robust against unusual or
tasks. In practice, crashing can be accomplished extreme task time realizations. A related concept
by a variety of means, for example by overtime, that may be useful is satisficing (Brown & Sim
by the application of additional resources, or by 2009). In evaluating the complex tradeoffs
the subcontracting of tasks. A crashing decision between time, cost and risk that are required for
typically represents a tradeoff between time and effective crashing decisions, satisficing models
cost, one that can be modeled as a simple linear provide the advantage that they do not require
program where task times are known. Crashing the decision maker to specify, a priori, a risk
decisions become much more complex, however, aversion parameter. Moreover, some satisficing
when task times are uncertain. A frequent measures have desirable properties that
difficulty is that the project manager does not generalize standard objectives, for example the
 Hall: Project Management: Recent Developments and Research Opportunities
134 J Syst Sci Syst Eng

maximization of the probability of a project project returns, and (c) correlation between
meeting its total cost, which includes both project returns, for example between projects in
crashing cost and overhead based on project the same industry or locality.
completion time. Goh & Hall (2012) develop Several classical approaches exist for
linear and piecewise linear decision rules, and a financial portfolio selection. These include (a)
rolling horizon approach, for a total cost the maximization of expected return, (b) the
satisficing model. minimization of the probability of
underperformance relative to a given target, (c)
4.1.2 Robust Optimization for Project the minimization of variance subject to meeting
Selection a target expected return (Markowitz 1959), and
Both statistical and informal evidence (d) the maximization of the safety-first ratio
suggest that a big factor in successful project (Roy 1952). The work of Aumann & Serrano
management is selection of the right projects (2008) on riskiness indices suggests a new
(Cooper et al. 2000). A major justification for direction related to target based choice.
this statement is that well chosen projects Intuitively, it should be possible to identify the
typically run smoothly and are easy to manage, least risky project portfolio that meets a given
whereas poorly chosen projects not only target in a certainty-equivalent sense. However,
underperform but may hurt other projects by modeling and identifying this portfolio present
absorbing their resources. There are two formidable computational challenges. For this
fundamentally different approaches to project reason, the design of a complex methodology for
selection. First, projects can be selected identifying an optimal portfolio needs to be
individually, usually based on a quantitative and supported by the design of a simpler
qualitative evaluation of their characteristics methodology that can reliably identify good
relative to predefined benchmarks. Alternatively, quality heuristic portfolios. Hall et al. (2011)
a project portfolio planning approach can be minimize the underperformance risk of the
used to evaluate multiple projects and make project portfolio, using absolute risk aversion as
simultaneous decisions about whether to accept a criterion, and considering both correlation and
them. The portfolio planning approach is both interaction effects between project returns.
more precise and more powerful, but presents
modeling and computational challenges. If 4.1.3 Earned Value Analysis
project returns are known, these can be Earned Value Analysis (EVA), also known as
addressed using the knapsack (Kellerer et al. earned value management, is an accounting and
2004) and related models (Fox et al. 1984, control system for projects that was developed
Dickinson et al. 2001). However, these in 1962 as part of a U.S. Department of Defense
challenges become much greater in the presence study. It is based on metrics that monitor the
of (a) uncertain project returns without well progress of the project, including the actual cost
specified probability distributions, (b) of work performed, the budgeted cost of work
interactions, such as synergistic value, between scheduled (i.e., the budget), and the budgeted
Hall: Project Management: Recent Developments and Research Opportunities
J Syst Sci Syst Eng 135

cost of work performed, which is also known as (Klastorin 2004, Kerzner 2009). In traditional
earned value. The central idea in EVA is that project management, these are usually defined
cost and time variances should be computed as Early Start (ES) and Late Finish (LF) times.
relative to the progress actually completed on The ES time serves as a wake-up call to the task
the project. Despite this apparently sensible operator, since it represents the earliest time at
justification, EVA is not widely used in industry which work on the task should begin. The LF
and has significant weaknesses. time serves as a deadline by which the task
One major weakness is that it does not operator should deliver the task to the project
distinguish between critical and noncritical manager; any delay beyond this point potentially
activities, although time variance is clearly a results in delay of the overall project.
more serious concern for critical activities. A During project execution, however, it often
related weakness is that EVA assumes that tasks becomes necessary to adjust the ES and LS
are independent, whereas in practice they are times, either earlier or later, although later is
often dependent, and consequently variance in more common. However, two problems arise.
one task affects the performance of another. First, notifying a task operator of new ES/LF
Behavioral issues also arise from the use of EVA. times has administrative and disruption costs.
For example, in order to make cost variance Second, the information based on which the
positive, managers may decrease the actual cost notification is made may not be final, and hence
of work performed (Kim & Ballard 2000). EVA may change again. In this case, decisions that
has no capability to measure quality, hence a are made on the basis of a notification that
project may be delivered on time and on budget subsequently changes may be costly. For
but not meet client expectations (Wikipedia example, they may result in resource
2012). A further problem is that EVA imposes reallocations that will lead to resources being
substantial information requirements on projects, underutilized if a task is not available as
for example precise measurement of project predicted. As another example, they may result
progress against a very detailed work breakdown in a loss of opportunity to improve the overall
structure, that projects often fail to meet (Lukas project completion time, as would occur if late
2008). These weaknesses and limitations of ES/LF times were given but the preceding tasks
earned value analysis imply the need for a new were actually completed early. A policy for task
approach that can be used more generally and notification is a set of rules for determining
more flexibly. This is particularly important in when one or more task operators should be
view of the increasing use of agile methodology, notified about a change in their ES/LF times.
which is not well served by EVA (Wikipedia Policies can range from the simplest one of
2012). giving notification only when information is
final, to more complicated ones that consider
4.1.4 Policies for Task Notification both the current best estimate of the time when a
In a typical project, the start and end times of task could start and the reliability of that
tasks are set from an initially planned schedule estimate. In this case, reliability can be
 Hall: Project Management: Recent Developments and Research Opportunities
136 J Syst Sci Syst Eng

measured using a variety of statistical metrics, project is not delayed. There are various
for example variance. The development of a task definitions of information availability that need
notification system that effectively balances the to be considered, for example whether or not the
costs of undernotification and overnotification project manager has full information about the
can greatly assist project companies in managing costs and resource availabilities of the task
their resources and controlling their project operators or subcontractors. The full information
costs. case leads to the issue of contract design. It is
necessary to specify incentives within the
4.2 Contractual Issues in Project contract that ensure cooperation, and this can be
Management achieved using either a common contract or one
that is customized for each task operator or
4.2.1 Cooperation in Project Management subcontractor. If the project manager does not
Successful completion of a complex project have full information, then it is necessary to
requires the cooperation of many task operators estimate costs and resource availabilities, and
or subcontractors. In either case, they need to be what is needed is a contract design that performs
motivated to crash, or expedite, their activities robustly against poor estimates. The possible use
where doing so benefits the project as a whole. of strategic information by the task operators or
However, such crashing activities incur subcontractors needs to be studied, for example
additional cost for the task operator or whether they will report false information in
subcontractor. This additional cost requires order to obtain additional compensation.
compensation from the project. The need for Conditions under which strategic reporting will
cooperation also arises in a second way. By not occur need to be developed. Cai et al. (2012)
sharing their resources, the task operators or model the coordination problem as a cooperative
subcontractors can achieve synergies that game, which is shown to be balanced, and
improve efficiency. However, even though the consider various contract design issues.
overall project benefits, not all of them will
necessarily benefit, hence once again 4.2.2 Real Options Analysis in Project
compensation is required. The structure of this Evaluation
problem suggests that it can be modeled as a For projects with predictable cash flows, a
cooperative game (Peleg & Sudhölter 2003). standard financial approach for evaluation is Net
The more important cooperative game concepts, Present Value (NPV). For projects that include
such as balancedness, can be applied. The unpredictable outcomes, for example the results
literature (Brânzei et al. 2002, Castro et al. 2008) of mining exploration or of an application for
applies cooperative game theory to the problem regulatory approval, net present value analysis
of setting compensation in a project that has can be extended into decision tree analysis.
already been delayed. However, both these methods ignore the effect
A potentially more useful application of of management options on project risk.
game theory is attempting to ensure that the Management options can include, for example,
Hall: Project Management: Recent Developments and Research Opportunities
J Syst Sci Syst Eng 137

delaying a decision until more information exist, however, further research is needed to take
becomes available, abandoning a project, or the important tool of ROA more deeply into the
selling rights that have been acquired. These evaluation of projects. There are mathematical
options substantially change the risk profile of a issues that need to be overcome to do this, and a
project, and hence the discount factor that greater understanding of the components of
should be applied in evaluating it. Yet neither of project risk needs to be obtained. Some
the above approaches accomplishes this. More counterintuitive results, to the effect that
specifically, while it is possible to use different variability in outcomes may reduce the
discount rates at different stages of a project in probability of management options ever being
NPV or decision tree analysis, the rates chosen exercised, and thereby reduce their value, are
would be arbitrary. However, Real Options obtained by Huchzermeier & Loch (2001).
Analysis (ROA) provides a way to calculate the
value of management options and impute 4.2.3 Design of Early Completion Incentives
discount rates that correctly adjust for project Task times in projects are typically uncertain,
risk. An introduction to the use of real options and may be either longer or shorter than their
for investment decisions, including project original estimate. Problematically, however, the
selection, is given by Amram & Kulatilaka effect on project completion time is not
(1999). symmetric. As a result, task times that are longer
ROA is widely used for financial asset than expected typically increase project
evaluation, but here the analysis is easier completion time. Whereas, task times that are
because of the existence of multiple assets that shorter than expected fail to reduce project
replicate the performance of an option. By completion time. The latter effect is primarily
contrast, the impact of ROA on risk management due to the behavioral phenomenon known as
in projects has been limited. A leader in the Parkinson’s Law (1958). This well known
application of ROA is Hewlett-Packard, principle states that “Work expands to fit the
however the company apparently uses it mainly time available” (Parkinson 1955). An equivalent
for procurement and other low risk, statement in the context of project management
contract-protected decisions. ROA has been little is that the time which it takes to complete a task
used in high risk industries, such as is the amount of time that is made available for
pharmaceuticals. ROA is not useful if a it (Wikipedia 2011). Goldratt (1997) proposes a
company lacks the discipline to implement a project management structure, Critical Chain
management option. For example, ROA author Project Management (CCPM), that claims to
N. Kulatilaka says, “Although you can make any resolve the problem of Parkinson’s Law. Patrick
project look good if you build in enough options, (1998) provides a detailed and supportive
a real world approach must address two discussion of CCPM. However, as Raz et al.
questions: when exactly do you shut it down, (2003) describe, there are several potential
and is there a good mechanism in sight to do problems with CCPM. These include concerns
that?” For situations where such discipline does about the behavioral effects of reducing safety
 Hall: Project Management: Recent Developments and Research Opportunities
138 J Syst Sci Syst Eng

margins, and also about the relative lack of or even identical to those in other projects. This
progress control that arises from the removal of similarity makes it worthwhile for companies to
task deadlines. consider investing in learning activities between
One possible resolution to Parkinson's Law projects (Kotnour 2000). Some companies use a
is a mechanism that is incentive compatible Project Management Office partly for this
(Hurwicz 1972) for the agents, i.e. the task purpose. Several project learning methods have
operators. A mechanism is incentive compatible been developed, and their success rates
if it is each agent's optimal choice, irrespective investigated (Schindler & Eppler 2003).
of what other agents do. Myerson (1979) shows Different learning strategies are discussed by
that any equilibrium outcome of an arbitrary Pich et al. (2002). However, business pressures,
mechanism can be replicated by an especially the need for project managers to start
incentive-compatible direct mechanism. This is work on a new project, tend to limit investment
the revelation principle, which simplifies the in learning from completed projects. It is indeed
search for an efficient mechanism. What is the case that investment in learning may delay
needed to resolve Parkinson's Law is an the start of a later project. Therefore, there exists
incentive system that encourages task operators a significant tradeoff in finding the amount of
to report early completion of their tasks, and learning between projects that is optimal, or
allocates rewards for doing so with sufficient even reasonably appropriate.
flexibility that various considerations, such as Research is needed to evaluate this tradeoff.
incentives for the project manager and meeting The main variable to be considered is how much
the available budget, are respected. The design investment, measured by time and resources,
of such a scheme becomes substantially more should be made in learning on completion of a
complex if a single task operator has control given project. This amount necessarily depends
over multiple tasks, and moreover those tasks on the similarity of the completed project to
are mutually dependent, for example if some are future planned or anticipated projects; the
successors of others. An interesting extension greater the similarity, the more investment is
would be to consider whether there is a worthwhile. Diminishing returns to time and
difference between short-run and long-run resources invested in learning need to be
incentives. For example, in the long run, early considered. The cost of learning should be
completion could result in a reduction in the modeled. Because time spent in learning may
time allowed for a task. delay the start of the next project, the time value
of money should be considered. The completion
4.3 Other Emerging Topics times of future projects will depend on their
makespan values, which need to be modeled as a
4.3.1 Learning between Projects function of previous learning and project
A popular view of projects is that each is similarity. Finally, depending on contract design,
unique. However, when considered at the task these values may have revenue and/or penalty
level, many projects have tasks that are similar implications.
Hall: Project Management: Recent Developments and Research Opportunities
J Syst Sci Syst Eng 139

4.3.2 Scalability of Agile Methodologies components, because of the importance that

The Agile Manifesto (agilemanifesto.org agile methodology places on interactions and
2001), proposes a new methodology for communications. An interesting potential
software development projects, but has since extension would be to develop a model that
become more widely applicable. Its principles could be used, for a given project management
include valuing: individuals and interactions application, to inform a choice between
over processes and tools, working software over traditional project management and agile
comprehensive documentation, customer methodology. The dynamic, synergistic aspects
collaboration over contract negotiation, and of agile methodology would need to be included
responding to change over following a plan. in such a model.
Agile project planning is especially useful for
nondeterministic projects, i.e. those where the 4.3.3 Sustainable Project Management
final configuration of the product or service The footprint of the world's economic
being developed is not known at the start of the activity exceeded the earth’s biocapacity by 50%
execution stage and only reveals itself as a result in 2007 (WWF 2010). In the U.K., for example,
of subsequent developments. Examples of the main source of pollution is construction and
nondeterministic projects include research and demolition of buildings (Royal Commission on
development, software development, and Environmental Pollution 2007); these activities
pharmaceutical drug development. There are are almost always organized as projects. An
many well documented success stories for agile interesting research question is how a regulator
methodology in projects involving small project can incentivize multiple projects to pollute less
teams (objectmentor.com 2012). However, the without unnecessarily increasing their costs.
implementation of agile methods for large This problem has been studied under voluntary
projects is more problematic. A generic idea, standards, especially the Leadership in Energy
forming teams of teams, has been tested but not and Environmental Design (LEED) program
extensively researched. For software (Corbett & Muthulingam 2007). However, it has
development applications specifically, there are apparently not been studied under explicit
some solutions related to appropriate environmental regulatory constraints.
architecture design (MSDN Blogs 2009). A typical regulatory constraint model
However, these solutions apparently do not involves two levels. At the higher level is a
extend to the broad and expanding range of regulator who acts as the decision leader. The
project management applications where agile regulator sets limits on pollution of different
methodology is now being used. types. The regulator optimizes a social welfare
An important issue, therefore, is to what function that may include project costs,
extent it is possible to model and evaluate the environmental benefits, and some investment in
scalability of agile project management green technology. At the lower level are multiple
methodology. Such models would necessarily project managers, who optimize their project
include both economic and behavioral costs, given the limits set by the regulator. Since
 Hall: Project Management: Recent Developments and Research Opportunities
140 J Syst Sci Syst Eng

the project managers act independently, the the growing range of applications.
regulator may need to provide a subsidy to 4. Important recent developments on the
ensure optimization of its social welfare function. business innovation side of project
Among the issues to be overcome is the design management are not yet well supported by
of incentives for truth telling. For example, the research.
project managers may exaggerate some of their 5. Leading journals in operations research and
costs, in order to receive less strict regulation. operations management have published
Ultimately, the most successful design of the few articles on project management in the
system will (a) encourage truthful reporting, and last 10 years, compared to many other
(b) coordinate the decisions of the regulator and topics of comparable practical importance.
the project managers. Because of both the 6. Practice and research have diverged, and
growing problem of industrial pollution, and the few new researchers have entered the
increasing use of project management in a wide project management field.
variety of business applications, the Our overall conclusion is that a confluence
development of an effective environmental of these factors has resulted in numerous
regulation system for projects is an important interesting research opportunities in project
issue. management for at least the next 10 years.

5. Conclusions Acknowledgments
Several conclusions can be drawn from this This research is supported in part by the
work. Summer Fellowship Program of the Fisher
1. We identify the following trends making College of Business, The Ohio State University.
project management harder: increased Preliminary versions of this paper were
competition, shorter product and service presented as plenary addresses to ICSSSM'2011,
life cycles, tighter budgets, unfamiliar and Tianjin, China (June 2011) and to Taishan
more complex applications, globally Academic Forum, Rizhao, China (July 2011).
distributed and multicultural project teams.
2. In contrast, several trends are making References
project management easier: better project [1] agilemanifesto.org. (2001). Manifesto for
management training, publication of best Agile Software Development. Available via
practices information, and better software DIALOG. http://agilemanifesto.org/
support. The relative impact of these two [2] Amram, M. & Kulatilaka, N. (1999). Real
effects varies from application to Options: Managing Strategic Investment in
application. an Uncertain World. Harvard Business
3. Underestimation of the value of project School Press, Boston, MA
management as a planning methodology [3] Aumann, R. & Serrano, R. (2008). An
over the last 20 years has led research to economic index of riskiness. Journal of
fall behind recent business innovation and Political Economy, 116: 810-836
Hall: Project Management: Recent Developments and Research Opportunities
J Syst Sci Syst Eng 141

[4] Bertsimas, D. & Sim, M. (2004). The price [12] Fox, G.E., Baker, N.R. & Bryant, L.J.
of robustness. Operations Research, 52: (1984). Economic models for R&D project
35-53 selection in the presence of project
[5] Brânzei, R., Ferrari, G., Fragnelli, V. & Tijs, interactions. Management Science, 30:
S. (2002). Two approaches to the problem of 890-902
sharing delay costs in joint projects. Annals [13] Goh, J. & Hall, N.G. (2012). Total cost
of Operations Research, 109: 359-374 control in project management via satisficing.
[6] Brown, D.B. & Sim, M. (2009). Satisficing Working paper, revised for publication,
measures for analysis of risky positions. Fisher College of Business, The Ohio State
Management Science, 55: 71-84 University, Columbus, Ohio
[7] Cai, X., Hall, N.G. & Zhang, F. (2012). [14] Goh, J. & Sim, M. (2010). Distributionally
Cooperation and contract design in project robust optimization and its tractable
management. Working paper, Department of approximations. Operations Research, 58:
Systems Engineering and Engineering 902-917
Management, Chinese University of Hong [15] Goldratt, E.M. (1997). The Critical Chain.
Kong North River Press, Great Barrington, MA
[8] Castro, J., Gόmez, D. & Tejada, J. (2008). A [16] Hall, N.G., Long, Z., Qi, J. & Sim, M.
polynomial rule for the problem of sharing (2011). Managing underperformance risk in
delay costs in PERT networks. Computers & project portfolio selection. Working paper,
Operations Research, 35: 2376-2387 Fisher College of Business, The Ohio State
[9] Cooper, R.G., Edgett, S.J. & Kleinschmidt, University, Columbus, Ohio
E.J. (2000). New problems, new solutions: [17] Huchzermeier, A. & Loch, C.H. (2001).
making portfolio management more effective. Project management under risk: using the
Research-Technology Management, 43: real options approach to evaluate exibility in
18-33 R&D. Management Science, 47: 85-101
[10] Corbett, C. & Muthulingam, S. (2007). [18] Hurwicz, L. (1972). On informationally
Adoption of voluntary environmental decentralized systems. In: McGuire, C.B.,
standards: the role of signaling and intrinsic Radner, R. (eds.), Decisions and
benets in the diffusion of the LEED green Organization. North-Holland, Amsterdam,
building standards. Working paper, Netherlands
Anderson Graduate School of Management, [19] Kellerer, H., Pferschy, U. & Pisinger, D.
UCLA (2004). Knapsack Problems. Springer, Berlin,
[11] Dickinson, M.W., Thornton, A.C. & Germany
Graves, S. (2001). Technology portfolio [20] Kerzner, H. (2009). Project Management:
management: optimizing interdependent A Systems Approach to Planning,
projects over multiple time periods. IEEE Scheduling, and Controlling, 10th edition.
Transactions on Engineering Management, Wiley, Hoboken, NJ
48: 518-527 [21] Kim, Y.W. & Ballard, G. (2000). Is the
 Hall: Project Management: Recent Developments and Research Opportunities
142 J Syst Sci Syst Eng

earned-value method an enemy of work flow? out from between Parkinson’s rock and
Working paper, Department of Civil and Murphy’s hard place. Available via
Environmental Enginering, University of DIALOG.
California at Berkeley, Berkeley, CA http://www.focusedperformance.com
[22] Klastorin, T.D. (2004). Project [32] Peleg, B. & Sudhölter, P. (2003).
Management: Tools and Trade-Offs, 1st Introduction to the Theory of Cooperative
edition. Wiley, Hoboken, NJ Games. Kluwer, Boston, MA
[23] Kotnour, T. (2000). Organizational [33] Pich, M.T., Loch, C.H. & De Meyer, A.
learning practices in the project management (2002). On uncertainty, ambiguity, and
environment. International Journal of Quality complexity in project management.
& Reliability Management, 17: 393-406 Management Science, 48: 1008-1023
[24] Lukas, J.A. (2008). Earned value analysis - [34] Project Management Institute. (2008). A
Why it doesn’t work. AACE International Guide to the Project Management Body of
Transactions, EVM.01.1-EVM.01.10 Knowledge, 4th edition
[25] Markowitz, H.M. (1959). Portfolio [35] Raz, T., Barnes, R. & Dvir, D. (2003). A
Selection: Efficient Diversification of critical look at critical chain project
Investments. Wiley, New York management. Project Management Journal,
[26] Myerson, R. (1979). Incentive 36: 24-32
compatibility and the bargaining problem. [36] Roy, A.D. (1952). Safety-first and the
Econometrica, 47: 61-73 holding of assets. Econometrica, 20: 431-449
[27] MSDN Blogs. (2009). Architecture makes [37] Royal Commission on Environmental
agile processes scalable. Available via Pollution. (2007). The urban environment.
DIALOG. Available via DIALOG
http://blogs.msdn.com/b/nickmalik/archive/2 http://www.official-documents.gov.uk/docu
009/05/19/architecture-makes-agile-processe ment/cm70/7009/7009.pdf
sscalable.aspx. Cited May 19, 2012 [38] Schindler, M. & Eppler, M.J. (2003).
[28] objectmentor.com. (2012). Agile/XP object Harvesting project knowledge: a review of
mentor success stories. Available via project learning methods and success factors.
DIALOG. International Journal of Project Management,
http://www.objectmentor.com/omSolutions/a 21: 219-228
gile_customers.html [39] Schonberger, R.J. (1981). Why projects are
[29] Parkinson, C.N. (1955). Parkinson’s law. always late: a rationale based on manual
Economist, November 19 simulation of a PERT/CPM network.
[30] Parkinson, C.N. (1958). Parkinson’s Law: Interfaces, 11: 66-70
The Pursuit of Progress. John Murray, [40] Smith, P.G. (2007). Flexible Product
London, UK Development: Building Agility for Changing
[31] Patrick, F.S. (1998). Critical chain Markets. John Wiley, Hoboken, NJ
scheduling and buffer management: getting [41] White, D. & Fortune, J. (2002). Current
Hall: Project Management: Recent Developments and Research Opportunities
J Syst Sci Syst Eng 143

practice in project management: an empirical degrees in economics from the University of

study. International Journal of Project Cambridge, a professional qualification in
Management, 20: 1-11 accounting, and a Ph.D. (1986) from the
[42] Wikipedia. (2011). Parkinson’s law. University of California at Berkeley. His main
Available via DIALOG. research interests are in tactical operations issues,
http://en.wikipedia.org/wiki/Parkinson’s_La especially project management, scheduling and
w pricing, public policy and sports management
[43] Wikipedia. (2012). Earned value problems. He is the author of over 70 refereed
management. Available via DIALOG. publications, and has given over 260 academic
http://en.wikipedia.org/wiki/Earned_value_ presentations, including 88 invited presentations
management in 20 countries, 6 conference keynote
[44] WWWF. (2010). Living planet report. presentations and 6 INFORMS tutorials. A 2008
Available via DIALOG. citation study ranked him 13th among 1,376
http://wwf.panda.org/about_our_earth/all_pu scholars in the operations management field. He
blications/living planet report is a Fellow of the Institute for Decision Making
under Uncertainty of the Chinese Academy of
Nicholas G. Hall is Professor of Management Sciences. He won the Faculty Outstanding
Sciences in the Fisher College of Business at Research Award of the Fisher College of
The Ohio State University. He holds B.A., M.A. Business in 1998 and 2005.