..
PROJECT MANAGEMENT JOURNAL
Sep 1993; P. 23
CONSTRUCTION PLANNING: Revising the Paradigm
..
Alexander Laufer, National Building Research Institute,
Technion, Isreal Institute of Technology, Haifa, Israel
Gregory A. Howell, University of New Mexico,
Albuquerque. New Mexico
..
The recommended DIDA approach to the project planning process and the
typology of plans:
- Is more difficult to understand--it is not neat and straightforward; and
..
- It makes planning more difficult--the planning process involves
more judgment and collaboration;
But...
..
- It results in plans which are more implementable--the planning
process produces models which reflect the real, imperfect world.
..
It is time to change the theory of project planning. Approaches that are widely
taught (and routinely required by company procedure) have been criticized
because they fail to measure up to the demands of reality [5] [22] [23] [27].
Discontent, particularly among line managers, has been evident for some time.
This discontent first became public in the Business Roundtable report "More
Construction for the Money," which says that modern management tools, including
planning systems, seldom realize their potential or live up to their promise
[4] 15].
..
Academic theory offers little help to practitioners who must make decisions in
the face of uncertain environments. Thus instead of relying on methods
described in textbooks, literature, and corporate manuals, experienced
practitioners apply unique, often intuitive ways to cope with difficulties as
they plan and complete projects. Lacking a theoretical basis, these managers
cannot systematize, expand or improve their practice. Absent a theory for
planning grounded in practice, academics find the methods used by managers
difficult to explain or teach. Practice has outrun theory, but cannot develop
further without adequate theory.
..
Instead of complaining that the current dissatisfaction arises from inadequate
implementation of "proven modern management techniques" by the untrained or
unmotivated, this article argues that dissatisfaction flows from more basic
defects in the paradigm that supports these "modern management techniques."
Exposing these defects requires that assumptions and conceptual foundations
inherent in the current paradigm be made explicit. As I.I. Mitroff says in
Break-Away Thinking, "The new thinking we require cannot result unless we first
understand the deeper patterns or road maps of reality that lie beneath the old
way of thinking" [40].
..
When change occurs at the paradigm level, formerly unquestioned assumptions and
supposedly obvious procedures come into question as they become explicit.
Principles once intuitively clear become unacceptable in the new light cast by
the emerging paradigm. In turn, new empirical findings possible within the new
paradigm add strength, supporting the emerging understanding. This article,
based on conclusions from a five-year research project into construction
planning, moves toward defining a new paradigm by showing that current
academically defined techniques are internally inconsistent, rest on false
assumptions about project circumstances, and fail to inform "best practices."
Simply put, current approaches are inadequate in the face of the reality of
uncertainty as experienced by practitioners.
..
The article is broken into seven parts;
..
- "Ten Currently Accepted Ground Rules" summarizes the prevailing
academic theory or paradigm of planning.
..
- "Major Premises--The Practitioner's Reality" provides the basis
for a critical challenge to the Accepted Ground Rules.
..
- "From Premises Toward a New Paradigm" presents important aspects
of the new approach that also serve to sharpen the distinction between
the old and new paradigm.
..
- A"Typology of Plans."
..
- "Why the Accepted Ground Rules are Faulty" critically reviews and
chal- lenges ground rules presented in the first section.
..
- "Where Do We Go from Here?" points new directions.
..
- "Summary" closes the discussion.
..
TEN CURRENTLY-ACCEPTED GROUND RULES FOR
CONSTRUCTION PLANNING
Ten ground rules for effective construction planning represent the commonly-
taught and -accepted approach. These rules, while not formally published, have
been compiled, tested and verified by three sources. Note that these rules will
be subsequently disavowed after being challenged by the premises drawn from the
practitioners' reality.
..
These ground rules were first compiled in discussions with more than 120
project managers at 11 U.S. construction-related firms L21]. There was strong
agreement during in-house lectures and discussions with these managers that
almost all formal education channels, including corporate and continuing
education programs, stress these ground rules as the basis for planning.
"Plans are nothing; planning is everything."
..
Dwight Eisenhower
..
Second, project management procedure manuals of 20 large owners or purchasers
of construction services and 13 large contractors were examined with respect to
their approach to overall project preparation and the development of project
teams in a research project conducted for the Construction Industry Institute
[14].
..
Third was a series of studies performed over the past five years at the
University of New Mexico, North Carolina State University, and Texas A&M
University. More than 60 graduate students in construction, with an average of
3-5 years of practical experience, were polled. They fully agreed that the
ground rules embody the material taught in their civil engineering and
construction management curricula.
..
The Ten Ground Rules
..
- Construction plans should be prepared before the onset of
construction, and as early as possible.
..
- A plan's time horizon should be maximal and a plan should be
comprehensive, detailed and complete.
..
- Information input for planning is received primarily from the
owner, the design engineers, the information system, subcontractors,
suppliers, and the experience accumulated within the construction
company.
..
- The major functional areas of the plan are scheduling, site
layout, production means, resources, and cash flow.
..
- A progressive construction company has a planning and
scheduling department responsible for preparing construction plans in
cooperation with site management, which then approves them. Planning
is performed with the aid of sophisticated methods, hardware, and
software.
..
- A progressive construction company will use the same formal
standardized planning techniques and procedures for planning all
projects.
..
- In those particular cases in which project uncertainty is high,
the use of statistically-based decision-making models is necessary.
Investment in an advanced and fast information system is also
required.
..
- Following the start of construction, one proceeds from planning
to the control stage. The essence of control is measurement and
evaluation of output and performance, followed by corrective steps
that adjust performance to the plan.
..
- The timely and frequent updating of the construction plan of
execution during construction is an indicator of effective control.
Advanced means are needed in order to distribute revised plans to
the site.
..
- Managers competently trained in the techniques of project
planning the techniques of project planning would eliminate most
planning problems encountered.
..
While these ground rules have been widely accepted as guidelines for teaching
planning in academia and are embodied in corporate planning procedures, they do
not ensure effective planning or useful plans. The group of 120 U.S. project
managers mentioned above agreed that the demands of the industry and the way
managers plan in practice are neither captured by nor reflected in these accep-
ted rules. Nonetheless, these managers were unable to articulate a clear alter
A fresh look at the assumptions or premises which underlie planning in the real
world begins to explain the gap between the promise offered by the accepted
rules and the limited achievement experienced in the field. These premises will
show that the practitioner's reality is inconsistent with academic theory,
i.e., the old planning paradigm. The premises describe that reality and serve
as a basis for the new paradigm.
..
MAJOR PREMISES--
THE PRACTITIONER'S REALITY
Premises or mindsets, although seldom consciously identified, guide the plan-
ning process and underlie the planners' activities. Argyris defines a premise
as " . . . a fact or proposition that is proved or assumed to be proved. A
premise is grounds for an inference and conclusion" [3]. Outlined below are
five commonsense premises culled from the reflections and observed practice of
18 mature and competent practitioners. All were senior project managers with
major well-established regional companies in the Western United States having
had significant experience in managing large construction projects. They were
selected by their companies to participate in research projects on the basis of
senior management assessment of their competence, experience and accomplish-
ments as demonstrated by their consistent high level of performance on a
variety of challenging projects [31]. These premises shape the planners' work
and form the basis for a revision of the accepted view.
..
- Uncertainty is not an exceptional state in an otherwise predictable
process of construction work. In fact it is a permanent feature in
the realm of construction.
..
Uncertainty, as defined by Galbraith, is the gap between the information
required to perform a task and that already possessed by the organization [12].
The planner, therefore, experiences uncertainty when there is a gap between the
two. As Duncan points out, uncertainty will be highest when the components of
the environment in which planning and implementation take place are numerous,
dissimilar, and in a continual process of change [9].
..
According to Duncan [9], the environmental factors of construction (e.g., wea-
ther, market conditions, clients, designers, suppliers, subcontractors, and
resource inputs) result in high levels of uncertainty. With the super position
of internal factors (e.g., fluctuations in the flow of new projects, labor
turnover, and the high degree of interdependence among project tasks) it is not
surprising to find that even projects considered to be "routine" are charact-
erized by high levels of uncertainty in comparison with many other industries.
Lawrence, who pioneered research on organizational environment and uncertainty,
divided uncertainty into two types, which he termed "information domain" and
"resource tension" [32]. He ranked various industries according to this classi-
fication, listing construction first in the information domain.
..
This ranking has also been demonstrated by a recent research study in which 93
managers of U.S. owner and construction companies from across the U.S. were
asked to characterize their "typical project" [15]. The projects ranged from $1
million to more than $200 million, with an average of about $40 million. About
two-thirds of the contracts were lump-sum and half of the projects were
industrial facilities. They were then asked to rate the degree of uncertainty
of the project's objectives at the beginning of construction. Project
objectives were defined to include scope definition, performance requirements
for the completed facility, and engineering design criteria. The managers used
a four-point scale from "very low uncertainty" (meaning that the project's
objectives were complete, clear, and stable) to "very high uncertainty" to rate
the degree of uncertainty of their project.
..
One would expect uncertainty about the project's objectives (determined
primarily by the owner and design engineers) to be very low in almost every
case at the start of construction. Nonetheless, the managers indicated that
very low uncertainty existed on only 22 percent of their typical projects. In
other words, high levels of uncertainty are the rule rather than the exception
even under what appears to be the circumstances most conducive to low
uncertainty.
..
The evidence shows that uncertainty is and should be regarded as a "fact of
life" in most construction projects rather than an isolated problem of limited
importance. Hence, our goal is to assist project management teams by providing
explicit strategies for diagnosing and coping with uncertainty rather than
ignoring or denying it.
..
- The longer the time interval between planning and
implementation, the higher the uncertainty concerning the planned
activity.
..
Since the future is inherently uncertain, the gap between available information
and needed information widens as the time span between planning and action
extends [l0] [33]. Some information regarding future activities can be acquired
only after other preparatory activities have been completed. Harrison brings it
ad absurdum by claiming that complete information is really available only
after the project is more or less finished [13].
..
- The higher the uncertainty in a project, the more difficult it
is to plan, and the less effective the plans will be at articulating
specific actions and outcomes.
..
Planning is always an inherently difficult and time-consuming process and
becomes even more so under conditions of high uncertainty [21]. As the
information gap widens, uncertainty increases and the planner's ability to get
a handle on the project's "big picture" is impaired. Information is in a state
of flu the instability often forcing revi- sion of whatever plans have been
prepared. Then, in turn, a new set of questions arises with respect to
reliability and/or sufficiency of the gathered information. These additional
issues further aggravate the difficulty faced by the planner.
..
The higher the uncertainty at the time of planning, the lower the accuracy of a
plan; that is, the lower the likelihood that the plan will materialize as it
stands. This relationship is well-established in cost estimation: namely, the
more information one has, the higher the accuracy of the cost estimate [5] [6].
This has been shown to be equally valid for schedule variance: an increase in
project uncertainty, such as that caused by a premature construction launch,
may result in greatly reduced accuracy in the realization of the plan. In a
recent study, managers overwhelmingly predicted a fivefold increase in schedule
variance if a $20 million construction project scheduled to take 18 months is
started at 20 percent rather than 50 percent design completion [25].
..
The increase in planning difficulty and the decrease in planning effectiveness
that result from high uncertainty often lead to problems related to planning
motivation and accountability. The motivation to invest in planning will suffer
as the planner does not perceive a strong linkage between the effort exerted in
planning and the effectiveness of the plan. Accountability may also suffer, as
it is difficult to pinpoint reasons for the plan's low effectiveness.
..
- Sophisticated models for planning are inadequate for coping
with highly uncertain situations.
..
Used mainly for scheduling and resource allocation, these models incorporate
probability analyses. Dean and Chaudhuri, summarizing developments in project-
scheduling techniques, state:
Computer simulation is an excellent method for finding the expected
completion time.... In this case activity times randomly selected for
each activity from the corresponding frequency distribution. The
project path length, duration and critical path are then calculated in
the usual (CPM) way. The procedure is repeated thousands of times,
using a computer program, and a record is kept of each run. An average
project duration and standard deviation are calculated on the basis of
the simulation. The resulting simulation estimates are usually more
reliable than the PERT calculation ... [8].
..
We agree that models may be useful to forecast the overall duration of a
construction process [44]. However, they will hold true only when the logic of
the network is firm (and changes are expected only in the duration of
activities). When construction begins with uncertain objectives, one should
expect significant changes in the scope and logic of the network itself. This
expectation has been substantiated by Mason [36]. The planner can rarely
construct a frequency distribution that describes changes in the relationships
between network activities; therefore, statistical models are of little help
when network logic itself is undergoing significant and unpredictable change.
..
While there is great interest in developing artificial intelligence techniques,
Kartam and Levitt, in a discussion on the limitations of progressive artificial
intelligence techniques, concluded that they were useful "... only for
addressing uncertainties for which discrete outcomes can be specified in
advance ...," [18] a condition which rarely exists.
..
Jans and Christensen negate the use of sophisticated planning models even under
this restriction. "... planning times based on Monte Carlo simulations or other
mathematical models may be dangerous because up to 80 percent of the total risk
in a project are of a human or society-associated nature. These factors are not
built into the traditional risk analysis method, which is an error of the
highest order." When it comes to planning rather than forecasting, they
conclude that "... such a model is not able to give the decision maker the real
cause for uncertainty. Consequently, the project manager will be in doubt about
the proper actions to be taken to counter the risks [16].
..
In addition to the various specific drawbacks, forecasting models suffer from a
fundamental flaw. Implicit in forecasting models is the assumption that the
future is determined more by what has happened between the past and the present
than by what will happen between now and the future. The more distant the point
in time to which the plan refers, the less valid is this assumption [2].
Construction projects do not conform to this assumption because of changes in
environment and in the objectives of the project, which render even the most
sophisticated models useless .
..
- The earlier in the life of a project the planner comes to
grips with all the relevant functional areas, the greater the
influence on the project.
..
Initiation of planning should allow sufficient time for the decision-making
process to run its course. This process includes gathering and processing
information, developing alternatives, analyzing and evaluating the
alternaatives, and finally selecting one alternative as a plan of action.
..
Moreover, early project planning prior to the launch of construction yields
significant savings in time and cost. It is widely accepted [13] [43] that
management's influence on cost performance declines dramatically as the project
progresses. The critical reasor for the potential impact of early project
plamnning lies in management's ability tc influence the broad parameters that
will establish the essential nature of the project. These decisions strongly
affect final costs prior to construction. As time passes the ability to
influence costs falls off sharply. This influence can be exercised only if
pre-project planning includes comprehensive thinking and examination of all
relevant areas covering the breadth of the project, including construction
methods, schedule, cost, organization and systems, site layout, logistics,
production means, work methods, and resources.
..
lt is, however, very difficult to estimate and prove the potential for saving
or to determine responsibility for the success or failure of a project because
of the inherent difficulty of evaluating the effectiveness of planning. The
results might have been due to a wide number of variables other than planning,
such as implementation and/or the environment. This applies particularly to
early planning, when much effort is spent on eliminating many possible
alternatives to prevent future undesirable states that would otherwise occur.
It is always difficult to identify and estimate the economic impact of
prevented states.
..
Taken together these premises highlight the importance of coping with
uncertainty in the planning process. Current academic theory offers no guidance
to the manager who is always faced with the uncertainty-related dilemma of
determining the timing of planning: should one plan and make decisions well
ahead of implementation and benefit from wielding greater influence (premise
#5); or postpone making the decisions until closer to implementation and secure
higher planning accuracy (premises #2 and #3)?
..
FROM PREMISES TOWARD A NEW PARADIGM
In the face of these premises, construction planning must be reconsidered. In
the new paradigm, the concept and process of planning is expanded by
identifying the stages of planning and different types of plans. These were
drawn from observations of experts in the field at work and were subsequently
verified through extensive review by experienced practitioners, and at a number
of leading construction companies in the United States [31].
..
The new approach to construction planning includes four stages:
..
- Diagnosis
..
- Information-gathering
..
- Deferring and splitting decisions
..
- Absorbing uncertainty
(the "DIDA" approach).
..
Since it is easier to explain the diagnostic activity after the other three are
understood, the discussion begins with the information-gathering step. A new
typology related to the DIDA approach is provided after discussing the four
steps.
..
Information-Gathering
As previously stated, missing information means uncertainty. Therefore, this
step is intended to reduce uncertainty through an active, dynamic approach to
information-gathering. But first it is necessary to put the
information-gathering stage to a systematic, formal procedure of scrutiny. The
role of the planning coordinator (e.g., the project engineer) should be to
collect, verify and exchange information.
..
The planning coordinator, guided and assisted by site management, should
determine whether all relevant information has been collected. Items on the
information-gathering checklist should include:
..
- Topics: items for decision, factors affecting the items for
decision, items alfected by the decision;
..
- Sources: internal, external, primary, secondary;
..
- Type: quantitative, qualitative, formal, informal;
..
- Time setting of events studied: historical, current, future;
and
..
- Method of acquiring information: collecting (e.g., documents,
telephone calls, meetings, tours), purchasing, generating.
..
Planning coordinators should continually judge the relevance, timeliness, and
cost-effectiveness of their information-gathering efforts. If additional
information is required, are the necessary information gathering resources
cost-effective? Will the process of further information-gathering leave
sufficient time for its exploitation? These issues are particularly significant
for schedule-driven projects. A recent study [24] indicates that when
uncertainty is high, practitioners intensify their information-gathering
efforts.
..
Deferring and Splitting Decisions
This step refers to the timing of decisions and an intervening splitting cess.
Once all possible measures have been taken to reduce uncertainty in the
information-gathering step, it is recommended that explicit efforts be made to
adapt to uncertain situations by splitting and deferring more detailed
decisions. The degree of detail involved is a key determinant of the
effectiveness of the plan, as shown to be the case by Harrison [13],
Lichtenberg [34], and Mason l36l. The most common error is probably that of too
much detail, too soon [41]. Some of the negative ramifications of an
overly-detailed plan, prepared too early, include:
..
- Costliness;
..
- Clutter, which obscures the overview of the project;
..
- Heavy updating requirements, which are time-consuming in the
monitoring and re-planning phases and make for long response times
and a concomi- tant loss of opportunity for corrective action in the
field; and
..
- Rapid obsolescence, because some decisions are based on
information provided by intelligent guesses rather than on reliable,
stable data.
..
We recommend that plans be prepared initially to a low level of detail. The
detail should vary inversely with the planning horizon, i.e., decisions about
details should be made as late as possible in the planning process. This does
not imply that decisions should be arbitrarily "put off until later." Rather,
it is a stage of deliberately splitting off those planning aspects which can be
acted upon more opportunely in the future. In essence, the authors recommend
that planners deliberately and formally decide to postpone making certain
decisions.
..
This concept is intended to help site management to make firmer decisions. As
both the situation and the related information become more stable with the
progress of time, the level of detail should be raised. This automatically
reduces the plan's level of uncertainty because decisions about highly unstable
situations are made closer to the point of implementation.
..
The authors do not advocate delaying the start of planning. To the contrary,
when partial solutions (i.e., initial planning with a low degree of detail) are
given their rightful place, early planning becomes meaningful. This planning
style provides for planning in greater detail at the appropriate stage of the
project.
..
A note of caution should be added: Splitting decisions may impair the
consistency of a plan (itself a system of integrated decisions), causing
confusion and leading to later possible overlooking of details.
..
The level of detail should, therefore, be considered on merit. It is advisable
to work to a level commensurate with the project's degree of certainty. If
uncertainty is low either because the technology is well-established by past
experience, or because the project objectives are not problematical and
environmental conditions are stable -- the plan should provide a
correspondingly higher degree of detail. (The relationship between the time
horizon and the degree of detail, and ways to maintain the consistency of the
plan, are illuminated by a new typology of project plans presented later.)
..
Absorbing Uncertainty
The final step in the DIDA approach is to help the planner absorb uncertainty
by facilitating flexible responses to future changes. By increasing the
flexibility in the content of the decisions (e.g., incorporating redundancy and
duplicating resources), the planner allows for unexpected contingencies.
..
One example of this flexibility is seen in the erection of a building skeleton.
For identical work packages (e.g., successive floors in a high-rise building),
it is usually beneficial to select a uniform pace of progress for the basic
skeleton activities (e.g., form work, bricklaying, and concrete placing). This
can be achieved by adjusting crew sizes. In order to overcome uncertainty, time
buffers can be added between activities to avoid an overly rigid schedule.
..
Flexible planning that allows for various contingencies is expensive. In
essence, it calls for a deliberate, careful, and confined slackening of
resources. However, when done properly, flexible planning reduces actual
incurred cost. As an illustration, the benefit of the additional time buffers
in the above example lies not only in reducing the time required for planning
and re-planning but also, more importantly, in absorbing many unexpected delays
(which might be caused by a single trade), without altering the timetable for
the entire work force. Budgetary contingency does not change the production
process. Flexible planning does. That's why it can lower project costs.
..
Diagnosis
The diagnostic activity, the first step in the DIDA approach, is part of
charting the planning process [27]. It focuses on the methodology of planning
and on ways to cope with uncertainty. Analysis of the project's unique
environment and its technological and organizational aspects is carried out
prior to the launch of, and periodically throughout, the project. The diagnosis
is intended to find the correct balance between: (I) information-gathering
efforts: too much will be too expensive and possibly too late for use, while
too little will lead to baseless decisions; (2) splitting and deferring
decisions: too late will not be useful, while too early will produce unfounded
decisions; and (3) absorbing uncertainty: overly-flexible decisions will lead
to undue expense, while overly-rigid ones will be vulnerable to frequent
changes and their attendant expenses.
..
Examining DIDA with regard to the decision-making process shows that each step
is associated with a different aspect of decision-making:
..
- The diagnosis is a pre-decision phase
..
- Information-gathering constitutes preparation of the process;
..
- Splitting plans and deferring decisions affects the entire
process (e.g., timing and pace); and
..
- The uncertainty-absorption phase affects the content of the
decisions themselves.
..
TYPOLOGY OF PROJECT PLANS
Types of Plans
A plan can be viewed as an attempt to model reality. The act of planning is an
attempt to create maps for the future that capture the essential ingredients
and dynamics of achieving goals and objectives. The process of planning is
characterized by a series of interrelated decisions, while the documented
results are referred to as "plans." In this regard we shall analyze the second
stage of DIDA, that of deferring and splitting declsions .
..
The typology presented in Table I describes three basic types of plans
categorized by their purposes. They can be identified and grouped into action
plans, decision guides, and forecasts, ranging from process-oriented project
decisions (first category) to preliminary action models (second category) to
final, product-oriented plans (third category).
Action plan
..
In an action plan, an implementor converts a model into reality. This plan
is characterized by its short time horizon, typically less than one month,
and by its very high level of detail. Action plans are intended to provide
the basis for a commitment to implement activities. They result in direct
assignment of work to be done, including decisions about who will do what,
when, and how. Action plans are typically used by their preparers (e.g.,
the site managers) and/or the preparers' subordinates. One example of an
action plan is a foreman's weekly work schedule [30].
..
Decision guides
These are models used for creating expanded and more-detailed plans, and
they eventually result in action plans. Decision guides involve only a
moderate degree of detail and typically have a time frame of one to six
months. They can be thought of as preparation and study for action plans,
and they ensure the continuity and consistency of decisions throughout the
project's implementation phase.
..
An important additional feature of the decision guide is that it offers a
global--rather than local--perspective. The superintendent who prepares a
decision guide (for example, a rolling three-month plan) for use by five
subordinate general foremen takes into account the scarcity of shared
resources (e.g., a crane) which would otherwise not be part of the
individual general foreman's action plan.
..
Decision guides, primarily used by subordinates of the preparer, provide
some or all of the following: (I) criteria for their later decisions (i.e.,
intermediate objectives and constraints); (2) a methodology for finding
solutions to problems at their respective levels (e.g., lower management
must follow given procedures in order to address their own detailed
decisions); and (3) a constraining framework for the possible alternatives.
..
The decision guide is also used as an outline and basis for coordination by
managers of peer units, both inside the organization (e.g., purchasing and
plant managers) and outside (e.g., subcontractors, suppliers). Whether used
as a guide or as an outline, the preparer should expect -- and even
encourage -- modifications of the plan based on feedback from users.
..
In more stable organizational environments the guidelines are provided by
standing organizational policies. Examples of the various evolutionary
stages of decision guides are found in the engineering design profession
(e.g., conceptual, preliminary, and detailed designs). The unique feature
of decision guides in construction planning is the need to make and revise
them throughout the life of the project.
..
While both action plans and decision guides document decisions with limited
time horizons, it is recommended that the planning process leading up to
these documents consider a long, or even a very long, period of time.
Plans should cover a limited time horizon, but planning should not;
planning should start very early and encompass a long time horizon. One
should always be peering into the long-term future while making concrete
action plans and decision guides for the near-term. Studying long-term
implica- tions, opportunities, changes of assumptions, objectives, and
constraints -- while not requiring full documentation in a plan -- may
exact rigorous analysis and cause major strategic decisions and new
directions.
..
Forecast
Forecasting involves estimating project performance at specified future
milestones. It is essentially an information process based primarily on
past performance, serving the needs of the preparer and upper management.
Its time horizon stretches one or more years into the future (to the end of
the project) and includes few details. A forecast initially determines
objectives and is used later to evaluate and control project performance.
It serves as a basis for the assumptions needed for decision-making and
upper-management planning, (e.g., based on the forecast, the owner may set
or revise the tenant occupancy schedule). "Project CPM" and "PERT charts"
are common forms for these plans.
..
A forecast is not really a plan of documented decisions, even though it is
based on high management directives (i.e., objectives and sub-objectives as
milestones). Currently, a forecast is often the only formal planning
document serving as a "master plan" or "project plan," while the fact that
it is only an estimate is overlooked [27l [28]. Since forecasts are
considered to be plans, the authors will use the term here; they qualify
it, however, by adding the modifier "initial."
..
We stress that an early start of the decision-making process gives
management greater influence (e.g., on project cost) in two respects:
..
- a decision guide permits an early study of the general
setting and the specific elements, and
..
- a forecast provides managers and/or planners with early
knowledge of where to expect problems and holdups. An early start
gives the planner a chance first to influence upcoming developments
early and, second, to direct attention to areas requiring early
intervention (i.e., to set up priority listings for
decision-making).
..
Multiple Time Horizons
It would be a mistake to assume that the three plans are distinguishable only
by a technical, hierarchical order or planning horizon, and that they are
easily expanded downward or summarized upward. This is not so. The plans differ
greatly and none of them lends itself to extrapolation, particularly not the
decision guides.
..
Every project plan should contain all three of the distinct "sub-plans"
discussed above: action plans for the short-term, decision guides for the
medium- term, and forecasts for the long term. The three sub-plans together
will enable us to do continuous mixed scanning, to identify broad fundamental
choices and to make incremental decisions. This method is highly compatible
with the recommendations of Etzioni in the Harvard Business Review [11].
..
While many construction organizations develop and maintain plans with different
time horizons and levels of detail, neither the planning process nor the
resultant plans demonstrate the differences of purpose, focus and product shown
in Table 1. The three types in Table 1 represent the result of a different
planning approach and process.
..
WHY THE ACCEPTED GROUND RULES ARE FAULTY
Clearly, the ten ground rules outlined in the first section of this article,
upon which construction has relied in classroom and in practice, require closer
critical evaluation, as in varying degrees they fall short of addressing the
real issues and problems faced by planners. As a system, the current paradigm
glosses over vital items and fails altogether in providing guidance for
planning. It has led construction planners to produce ineffective and partly
useless plans. Below, the ten points are presented again, this time with
elaboration on critical elements shown in bold typeface.
..
- Construction plans should be prepared before the onset of
construction, and as early as possible.
..
- A plan's time horizon should be maximal and the plan itself
should be comprehensive, detailed and complete.
..
Plans, the time horizon they cover, the timing of their preparation, and the
degree of detail involved, should depend on and reflect:
..
- The type of plan intended,
..
- The organizational level it is designed to serve, and
..
- The degree of uncertainty involved or the size of the "information
gap" the planner faces.
..
Planning should start early, but should not be completed in full detail prior
to construction. Completing the plan is an iterative process that unfolds
throughout construction.
..
- Information input for planning is received primarily from the
owner, the design engineers, the information system, and the
experience accumulated within the construction company, as well as
from subcontractors and suppliers.
..
Planning is inherently active, not passive. Unlike the design engineer who is
handed most of the necessary information, the planner is not a passive
recipient. The planner relies little on the information provided, but plays a
proactive, dynamic role in the information-gathering process -- collecting,
trading, and verifying information, and tailoring information-gathering tools
and sources to the project's specific needs.
..
- The major functional areas of the plan are scheduling, site
layout, production means, resources, and cash flow.
..
In addition to developing the plans mentioned, the planner must consider issues
of construction technology (constructability), organizational structure,
staffing, contractual strategies, work methods, and quality and safety systems.
Emphasis should be shifted from preparing product-oriented plans for control
(mainly scheduling for forecasting) to making process-oriented plans (mainly
logistics, production means, and work methods for implementation) [31].
..
- A progressive construction company has a planning and
scheduling department responsible for preparing construction plans
in cooperation with site management, which approves them after their
input is completed. Planning is performed with the aid of
sophisticated methods, hardware, and software.
..
The separation between planners and plan implementors is always fraught with
difficulty [28] [29l. Often when planning and doing are separated, "planning is
given the kiss of death" [ 1 ] . Under uncertain conditions, such as those
common at the construction site, plans must be updated frequently. This demands
speedy communication between, first, the source of the changes in the plan (the
site); second, the locus of decision-making (the planning department); and
third, the implementation arena (again, the site). With its ongoing operations
the site cannot normally wait for this process, even when it is expedited;
placing the planner on site may help. Even when planners are located on-site,
however, active participation by the site manager is essential in crucial
stages of the planning process [29]. It is a fallacy to assume that a manager
can approve plans comprising hundreds of interrelated decisions if the plans
were prepared by a staff specialist without the manager's active, significant
involvement.
..
Sophisticated methods are helpful mainly in the processing of information.
Uncertainty, however, makes planning difficult both at the
information-gathering stage and at the point of communicating the plans'
contents in usable form. Situations of high uncertainty demand uncomplicated,
flexible planning organization and methods.
..
- A progressive construction company will use the same formal
standardized techniques and procedures for the planning of all
projects.
..
An organized body such as a construction company must, of course, adopt formal
procedures, but the "one best way" approach to problem-solving should be
avoided. A situational analysis approach should be adopted that tailors the
problem-solving method to the particular situation at hand [22]. The diagnosis
section of the DIDA approach addresses precisely this issue.
..
- In particular cases where project uncertainty is high,
recourse to statistically-based decision-making models is dictated.
Investment in an advanced, fast information system is required for
this purpose.
-
..
The usefulness of statistical models in forecasting is limited to conditions
where available data describe the stochastic behavior of the variables
involved, and only if the future can be reasonably deduced from the past. The
constantly-changing information and dynamic environment of most construction
projects, especially with project objectives in flux, calls into question the
value of such models in making and integrating numerous highly interrelated
decisions.
..
- After construction has started, emphasis shifts from planning
to control. The essence of control is measurement and evaluation of
output and performance followed by corrective steps that adjust
performance to the plan.
..
Under the typically uncertain conditions prevalent at most construction sites,
planning is not a one-shot operation performed prior to the beginning of
construction. Instead, planning is an ongoing effort that continues throughout
the life of the project. As project uncertainty in- creases, control is less a
"governor" of execution (ensuring that implementation conforms to plans), and
more a data collection and analysis function [26].
..
- Timely and frequent updating of the construction plan in the
course of the construction process is an indicator of effective
control. Advanced means are needed for distribution of revised plans
to the site.
..
"Capacious planning" assumes predictability of the future. "Good planning"
assumes that the future is not predictable. Since models and software are
designed for revisions, and paper accommodates them, planners often see
updating as one of their primary purposes. But site management in real life
situations finds it difficult to cope with frequent changes in the master plan.
By adapting the plan's time horizons, degree of detail, and frequency of
revisions both to prevailing uncertainty and to management level, effective
planning and comntrol may be realized. Thus, updating means preparing a
detailed short-term plan based on new information gathered. The action plan is
the one that should be updated frequently.
..
Advanced means may be an asset for redistributing plans, but it is unlikely
that anything less than site management's real involvement in re-planning will
bring about a genuine attempt to adhere to the revisions. It is worth
reitering that when uncertainty is high, communications should be as speedy as
possible and most of the planning must be done on-site.
..
- Managers competently trained in project-planning techniques
would eliminate most planning problems encountered.
..
The current problems in planning are beyond the responsibility of the managers
alone. The problems lie in the qualification, orientation, and motivation of
all parties involved. This holds true throughout the life cycle of planning --
beginning with research and development of planning techniques by academics and
system analysts, to application of these techniques by engineers, and lastly,
to utilization of the plans by managers [27] .
..
Training in techniques is only one factor affecting the ability of managers to
plan effectively. Emphasis should be shifted from training managers in
techniques (and certainly not in currently recommended techniques), but rather
to addressing the numerous issues presented in this article. Mintzberg
castigates U.S. business schools' overemphasis on techniques. He demonstrates,
with research findings by professors of Harvard Business School, and others,
that business schools were partly responsible for the current misguided
obsession with technical analysis [38l. No correlation was found between grades
at Harvard Business School and subsequent success in management jobs.
..
WHERE DO WE GO FROM HERE?
In the last few years, the ideas expressed here have been brought before two
different groups. First are various assemblies of people charged with "doing"
in the world of action, e.g., managers attending construction companies'
in-house seminars. Second are professors and consultants attending academic
gatherings at a number of European and American universities. The reactions of
the two groups differed sharply.
..
Practical people, largely experienced construction project managers either in
forums sponsored by the Construction Industry Institute and in-house seminars
or those who have reviewed drafts of this article (and are acknowledged at the
end of the article) received the ideas enthusiastically. Some already operate
on this basis, at least partly if not fully. Others have asked to incorporate
the recommendations into company practices and training sessions. Generally
they clamor for more specificity in this direction. In the words of one project
manager who expresses the thoughts of many: "Now, at last, I feel I have useful
working principles at my disposal which will protect me from an overdose of
management science techniques on the one hand, and from relying too much on my
intuitive gut feeling on the other. What we need now are a large number of
examples of problems and good solutions taken from real life situations from
which more detailed procedures could be developed." Even more encouraging are
the positive reports from managers (some mentioned in the acknowledgment)
actively employing the DIDA process and the typology of plans.
..
The response from academicians was quite the opposite. Most rejected the ideas
out of hand, perhaps feeling that hallowed (if unspoken) principles are in
jeopardy. This corresponds closely to what Thomas Kuhn ll9] [20] had to say
(primarily about academicians from the exact sciences): The scientific communi-
ty is held together by bonds and commitments, the most fundamental of these
being the bonds which concern the world view scientists share and which
underwrites their approach to scientific enquiry (paraphrased).
..
In recent years, paradigms quite different from those long-established at the
universities have been introduced by Kanter [117], Pascale [42], and Senge
[46]. All three are members of reputable establishments (Harvard, Stanford and
the Massachusetts Institute of Technology, respectively). Unlike their
colleagues, these academicians have worked closely with practitioners.
Hopefully, the discipline of construction management will follow a similar
process.
..
Clearly, much of the current methodology does not measure up to reality,
because some rules were conceived in error and because important aspects --
particularly uncertainty -- were given insufficient weight. In this article the
authors point to new concepts as a basis for future work on development of a
comprehensive system to cope with construction planning under uncertain
conditions. The research and academic communities have a responsibility to
clarify, define, and express rules for effective project planning under
conditions of uncertainty. They owe the industry an attempt at developing rules
that mirror the successful experience of the practitioner. Schon wrote: "...
competent practitioners usu- ally know more than they can say.... They know
about coping intuitively with uncertainty and effective project planning" [45].
The practitioners must, in collaboration with the research and academic
communities, reflect on their work to promote the development of a realistic
theory of planning that will contribute to more effective execution of
construction projects.
..
SUMMARY
As a group, the ten commonly accepted ground rules for construction planning:
- Are simple to understand--they make for a compact orderly theory; and
..
- Are convenient as a basis for planning--the planning process is
relatively simple if followed;
But...
..
- They result in plans which are impossible to implement ment --
they ignore the issues of turbulence and uncertainty that are inherent
characteristics of the industry.
..
The recommended DIDA approach to the project planning process and the
typology of plans:
- Is more difficult to understand--it is not neat and straighfforward;
and
..
- It makes planning more difficult--the planning process involves
more judgment and collaboration;
But...
..
- It results in plans which are more implementable -- the planning
process produces models which reflect the real, imperfect world.
..
That planning problems can be eliminated altogether is wishful thinking.
Preparing and implementing operational plans entails many difficulties, and
explanations for the final outcome should be tempered with moderation. For
planning to become effective, methods should be changed (e.g., gathering and
communicating of information), policies modified (e.g., roles of planning and
control), assumptions adjusted (e.g., attitudes toward uncertainty), and the
overll philosophy and underlying paradigm of project management re-examined.
..
ACKNOWLEDGMENTS
The nature of the research reported here requires verification by leaders in
the field. The following persons have provided vital feedback and assistance:
..
Glenn Ballard - Ballard Management
Services
Len Harris - Brown & Root Braun
Dr. Richard Tucker; Chuck McGinnis -
Construction Industry Institute
..
H.R. Benton; Pete Richmond - E.I. duPont de Nemours
Steve Braunstein; Neil L. MacFarlane;
Ed McGuire; Edward Wynant -
Exxon Research and EngineeringAlvin Burkhart - Hensel-Phelps
Theodore J. Kratt - J.A. Jones Construction Co.
..
James R. Carroll; Dr. James M. Neil - Morrison-Knudsen
Hasan Hammami - Procter & Gamble
C.H. Oglesby; Henry Parker - Stanford
University
Donald T. Killilea - Stone and Webster
Engineering
..
Lauri Koskela - Technical Research
Center of Finland
Dr. George Stukhart - Texas A&M University
Vally N. Kovary - The International
Foundation Center
..
Dr. John Borcherding - University of
Texas
..
REFERENCES
- Ackoff, R.L. 1970. A Concept of
Corporate Planning. New York: John
Wiley & Sons.
..
- Ackoff, R.L. 1983. Beyond Predic-
tion and Preparation. Journal of Manage-
ment Studies, 20(1), 59-69.
- Argyris, C.1982. Reasoning, Lea~ning
and Action. San Francisco: Jossey-Bass.
- Business Roundtable. 1983a. More
Construction for the Money. Summary
Report of the Construction Industry Cost
Effectiveness Project. The Business
Roundtable. New York.
..
- Business Roundtable. 1983b. Re-
port on Planning and Scheduling. Ap-
pendix A-6.1--Report A-6--Modern
Management Systems. The Business
Roundtable. New York.
- Clark, F.D. and Lorenzoni, A.B.
1978. Applied Cost and Engineering.
New York: Marcel Dekker.
..
- Cohenca, D., Laufer, A. and Ledbet-
ter, W.B. 1989. Factors Affecting Con-
struction Planning Efforts. Construction
Engineering and Management, ASCE,
115(1), 70-89.
- Dean, B.V. and Chaudhuri, A.K.
1980. Project Scheduling: A Critical Review.
In B.V. Dean and J.L. Goldhar
(eds.), Management of Research and Innovation.
New York: North-Holland Publishing Co.
..
- Duncan, B. 1972. Characteristics of
Organizational Environment and Per-
ceived Environmental Uncertainty. Ad-
ministrative Science Quarterly, 17,
313-327.
- Emery, J.C. 1969. Organizational
Planning and Control Systems, Theory
and Technology. London: The MacMil-
lan Company.
..
- Etzioni, A.1989. Humble Decision
Making. Harvard Business Review.
(July/August), pg . 122 - 126.
- Galbraith, J.R. 1977. Organizatian
Design. Reading, Massachusetts: Ad-
dison-Wesley Publishing.
..
- Harrison, F.L.1981. Adlunced Pro-
ject Management. New York: John Wiley
and Sons.
- Howell,G.1990.HowOwnersand
Contractors Organize Project Teams. Re-
port to the Construction Industry Insti-
tute. Austin, Texas.
..
- Howell, G., Laufer, A., Ballard, G.
1993. Uncertainty and Project Objec-
tives. In Press, Project Appraisal.
- Jans, A. and Christensen, P.J.1988.
New Generation of Risk Management in
Off-Shore Projects. Presented at the 9th
Intemet World Congress on Project Man-
agement. Glasgow, Scotland.
..
- Kanter, R.M. 1989. When Giants
Learn To Dance. New York: Simon and
Schuster, Inc.
- Kartam, N.A. and Levitt, R.E.1991.
An Artificial Intelligence Approach to Pro-
ject Planning Under Uncertainty. Project
Management Joumal, XXII(2), pp.7-11.
..
- Kuhn, T.S. 1962. The Structure of
Scientific Revolutions. Chicago: Univer-
sity of Chicago Press.
- Kuhn, T.S.1974. Second Thoughts
on Paradigms. In Frederick Suppe (ed.),
The Structure of Scientific Theories: 459-
482. Urbana, Illinois: University of Illi-
nois Press.
..
- Laufer, A. 1989. Owner's Project
Planning: The Process Approach. Report
to the Construction Industry Institute.
Austin, Texas.
- Laufer, A. 1991a. Construction
Planning in Uncertain Environments. In-
tern. Journal of Project Management,
vol. 9, no. #1, (February),53-6.
..
- Laufer, A.199lb. ProjectPlanning:
Timing Issues and Path of Progress. Pro-
ject Management Journal, XXII(2), pp.
29-45.
- Laufer, A.1992. A Micro View of the
Project Planning Process. In press, Con-
struction Management and Economics.
..
- Laufer, A. and Cohenca, D. 1990.
Factors Affecting Construction Planning
Outcomes. Construction Engineering and
Management, ASCE,116(1),135-156.
- Laufer, A. and Koskela, L. 1992.
Anticipatory Project Control. Working
paper. National Building Research Insti-
tute, Technion, Haifa.
..
- Laufer, A. and Tucker, R.L. 1987.
Is Construction Project Planning Really
Doing Its Job? A Critical Examination of
Focus, Role and Process. Construction
Management and Economics, 5(3), 243-
266.
- Laufer, A. and Tucker, R.L. 1988.
Competence and Timing Dilemma in Con-
struction Planning. Construction Manage-
ment and Economics, 6(4),339-355.
..
- Laufer, A., Cohenca-Zall, D. and
Howell, G.A. 1992. The Planner-Man-
ager Competence Dilemma Revisited. In-
ternet Conference, Florence Italy. (July).
- Laufer, A., Howell, G.A. and Rosen-
feld, Y. 1992. Three Modes of Short-Term
Construction Planning. In press, Construc-
tion Management and Economics.
..
- Laufer, A., Shapira, A., Cohenca-
Zall, D., and Howell, G.A. 1992. The Pre-
Bid and Pre-Construction Planning
Process. Journal of Construction Engineer -
ing and Management. ASCE In press.
- Lawrence, P.R. 1981. Organization:
An environmental perspective. In A.H. Van
de Ven and W.F. Joyce (eds.). Perspectives
on Organizational Design and Behavior.
New York: John Wiley and Sons.
..
- Le Breton, P.P. and Henning, D.A.
1961. Planning Theory. Englewood
Cliffs, NJ: Prentice-Hall.
- Lichtenberg, S. 1986. Medieval
Remains in Modem Project Management
and Successor Principles for the Nineties.
In M.C. Grool et al. (eds.), Project Man-
agement in Progress. Amsterdam: El-
sevier Science Publishers.
..
- March, James.1988. Decisions and
Organizations. Blackwell: Basic Books.
- Mason, D. 1984. The CPM Tech-
nique in Construction: A Critique. Trans-
actions of the American Association of
Cost Engineers. E.2.1 -E.2.10, Montreal.
..
- Mintzberg, H. 1973. The Nature of
Managerial Work. New York: Harper and
Kow.
- Mintzberg, H. 1981. Research
Notes and Communications--What is
Planning Anyway? Strategi( Manage-
ment.loul-nal, 2, 319-324.
..
- Mintzberg, H. 1989. Mintzberg on
Management. New York: The Free Press.
- Mitroff, I.I. 1988. Break-Away
Thinking: How to Challenge Yow- Business
Assumptions (and Why You Should). New
York: John Wiley and Sons.
..
- Morton, G.H.A. 1983. Human Dy-
namics in Project Planning. In D.I.
Cleland and W.R. King (eds.), Project
Management Handbook. New York: Van
Nostrand Reinhold.
- Pascale, R.T. 1990. Managing on
the Edge: How the Smartest Companies
Use Con~lict to Stay Ahead. New York:
Simon and Schuster, Inc.
..
- Paulson, B.C. 1976. Designing to
Reduce Construction Costs. Journal of
the Construction Division, ASCE,
102(4),587-592.
- Ragsdale, C. 1989. The Current
State of Network Simulation in Project
Management Theory and Practice.
OME~GA International Journal of Man-
agement, Science, 17 (1), 21 -25.
..
- Schon, D.A. 1983. The Reflective
Practitioner. New York: Basic Books.
- Senge, P.M. 1990. The Fifth Disci-
pline: The Art and Practice of the Learn-
ing Organization. New York: Doubleday/
Currency.
..
AUTHORS
Alexander Lauler is a faculty member in the Department of Civil Engineering at
the Technion, Israel Institute of Technology. He is also head of the Department
of Construction and Economics at the National Building Research Institute. Dr.
Laufer obtained his Ph.D. in construction engineering and management at the
University of Texas, following which he taught at Texas A&M University. He also
spent a year at North Carolina State University as a visiting associate
professor.
..
Gregory A. Howell received his B.S.C.E. and M.S.C.E. in construction management
from Stanford University. He is a registered Professional Civil Engineer in
California, and a member of the American Society of Civil Engineers and the
Project Management Institute. He is an associate professor in civil engineer-
ing at the University of New Mexico, where he was honored with the College of
Engineering Teaching Excellence Award in 1991.