No creditforsolvingproblemsneverhappened

CALIFORNIA MANAGEMENT REVIEW VOL. 43, NO. 4 SUMMER 200164
Work reported here was supported by the MIT Center for Innovation in Product Development under
NSF Cooperative Agreement Number EEC-9529140. For more information on the research program
that generated this article, visit <http://web.mit.edu/nelsonr/www/>.
Nobody Ever Gets Credit
for Fixing Problems that
Never Happened:
CREATING AND SUSTAINING PROCESS
IMPROVEMENT
Nelson P. Repenning
John D. Sterman
H
ow much would your organization pay to develop manufacturing
capability equal to Toyota’s? How much would a world-class, six-
sigma quality program be worth to your company? How about
Harley-Davidson’s ability to tap into the hearts and minds of its cus-
tomers or Dell’s ability to manage its supply chain? Most firms are working
aggressively to develop these and similar capabilities through process improve-
ment. The combined expenditure of U.S. companies on management consul-
tants and training in 1997 was over $100 billion, and a sizeable fraction went
towards efforts to develop operational capabilities matching those of the best
firms in business. Whether it’s an advanced manufacturing system or the ability
to respond quickly to changing customer needs, the drive toward improvement
has become a way of life in corporations today. There is only one problem.
Despite these vast expenditures, and notwithstanding dramatic successes in a
few companies, few efforts to implement such programs actually produce signifi-
cant results.
Consider, for example, Total Quality Management (TQM). In the 1980s,
spurred by the success of many Japanese firms, TQM was all the rage among
U.S. firms. Consultants and business school faculty preached its virtues and
managers made pilgrimages to companies with award-winning quality pro-
grams. By the mid-1990s, however, TQM was considered passé. Academics had
moved on to other issues, TQM received rare mention in the popular business
press, and articles that did mention it usually did so in a negative context. TQM

had all the earmarks of a management fad: An initial burst of enthusiasm, a
flurry of activity, and then a steady decline as it was replaced by newer innova-
tions such as re-engineering. It would be easy to conclude that TQM’s underly-
ing value was minimal.
However, when one looks at the experience a little more carefully, a dif-
ferent picture emerges. A number of careful studies have now demonstrated that
companies making a serious commitment to the disciplines and methods associ-
ated with TQM outperform their competitors.1
There is now little doubt that
when used properly, TQM produces significant value to both organizations and
their customers. Yet paradoxically, it remains little used. A recent study found
that fewer than 10% of the Fortune 1000 had well-developed TQM programs;
and, in another study, TQM fell from the third most commonly used business
tool in 1993 to 14th
in 1999.2
The situation is similar for a wide range of other
administrative and technological innovations.3
Techniques touted as today’s
“core competencies” all too often become tomorrow’s failed programs. Once
an effort has failed, there is an almost irresistible temptation to label it a fad or
“flavor of the month.” However, digging a little deeper shows that many such
techniques have useful content. It should come as little surprise then that many
currently popular innovations are little more than old ideas with new acronyms.
The core disciplines associated with statistical process control and variance
reduction become six-sigma; what was once called a quality circle is now a
high-performance work team.
Thus, today’s managers face a paradox. On the one hand, the number of
tools and techniques available to improve performance is growing rapidly. Fur-
ther, with advances in information technology and the ever-growing legions of
management consultants, it is easier than ever to learn about these techniques
and to learn who else is using them. On the other hand, there has been little
improvement in the ability of organizations to incorporate these innovations in
their everyday activities. The ability to identify and learn about new improvement
methods no longer presents a significant barrier to most managers. Instead, suc-
cessfully implementing these innovations presents the biggest challenge. Put more
simply, you can’t buy a turnkey six-sigma quality program. It must be developed
from within.
To learn how firms can overcome this “improvement paradox,” we
have, over the past decade, studied process improvement and learning programs,
focusing on the dynamics of implementation and organizational change. We
conducted over a dozen in-depth case studies in industries including telecom-
munications, semiconductors, chemicals, oil, automobiles, and recreational
products.4
We gathered data through observations, extensive interviews with
participants, archival records, and quantitative metrics. We complemented
our field research with the development of a series of models capturing the
dynamics of implementation and improvement.5
Using system dynamics as the
basis for understanding implementation has yielded a number of insights into
the improvement paradox. In at least some cases, these insights have proven
Nobody Ever Gets Credit for Fixing Problems that Never Happened
CALIFORNIA MANAGEMENT REVIEW VOL. 43, NO. 4 SUMMER 2001 65

instrumental in helping firms benefit from the potential provided by available
improvement tools and techniques.
Most importantly, our research suggests that the inability of most organi-
zations to reap the full benefit of these innovations has little to do with the spe-
cific improvement tool they select. Instead, the problem has its roots in how
the introduction of a new improvement program interacts with the physical,
economic, social, and psychological structures in which implementation takes
place. In other words, it’s not just a tool problem, any more than it’s a human
resources problem or a leadership problem. Instead it is a systemic problem, one
that is created by the interaction of tools, equipment, workers, and managers.
The Structure of Improvement
We present the lessons that have emerged from our study in the form of
a causal loop diagram. Our model provides both a useful framework for thinking
about the challenges associated with implementing improvement programs and
practical suggestions to increase the chances that your next such effort will suc-
ceed. While the theory reported here initially emerged from the study of two
improvement initiatives in a major automaker,6
the resulting model is quite gen-
eral and can be applied to a range of situations. We have observed these dynam-
ics in almost every organization we have studied.
Figure 1 begins with the basic “physics” underlying process improvement.
The actual performance of any process depends on two factors: the amount of
Time Spent Working and the Capability of the process used to do that work. For
example, in manufacturing, net usable output is given by the product of labor
hours per day and productivity (usable units per labor hour).
The performance of any process can be increased by dedicating additional
effort to either work or improvement. However, the two activities do not pro-
duce equivalent results. Time spent on improving the capability of a process
typically yields the more enduring change. For example, boosting the workweek
20% might increase output 20%, but only for the duration of the overtime.
Gains in process capability, however, boost the output generated by every sub-
sequent hour of effort. Similarly, overtime devoted to reworking defective prod-
ucts can boost net usable output, but only as long as the overtime is continued,
while eliminating the root causes of those defects permanently reduces the need
for rework. We capture this persistence by representing Capability as a stock
(denoted by a rectangle), that is, as an asset that accumulates improvements
over time. Specifically, Time Spent on Improvement increases the flow of Investments
in Capability that augments process capability.
While it often yields the more permanent gain, time spent on improve-
ment does not immediately improve performance. It takes time to uncover the
root causes of process problems and then to discover, test, and implement solu-
tions, shown in the diagram as a delay between improvement activities and the

resulting change in process capability. Moreover, no improvement in capability
lasts forever. Machines wear, processes go out of control without regular atten-
tion, designs become obsolete, and procedures become outdated. Thus, we also
show an outﬂow from the stock capturing the inevitable decline of any capabil-
ity that is not regularly maintained. The lag in enhancing capability depends on
the technical and organizational complexity of the process. Studies show that
the delay in improving relatively simple processes such as the yield of machines
in a job shop is on the order of a few months, while the delay in improving
highly complex processes such as product development can be several years or
more.7
Similarly, the lifetime of improvements in capability will be shorter in
organizations with high rates of change in products and people.
Besides the physical and institutional structures that determine perfor-
mance, Figure 1 also shows the goal for process throughput set by senior man-
agers (labeled Desired Performance). The goal could be the number of products
demanded by customers each day, the rate at which claims need to be processed
by an insurance company, or the number of new products the ﬁrm seeks to
launch this quarter. People compare that goal to their actual performance to
determine the Performance Gap. Not surprisingly, in the organizations we studied
FIGURE 1. The “Physics” of Improvement
Note:Arrows indicate the direction of causality. Signs (‘+’ or ‘–’) at arrowheads indicate the polarity of relationships: a ‘+’ means
that an increase in the independent variable causes the dependent variable to increase, all else being equal (a decrease causes a
decrease); similarly, a ‘–’ indicates that an increase in the independent variable causes the dependent variable to decrease (a
decrease causes an increase). For more details, see J. Sterman, Business Dynamics: Systems Thinking and Modeling for a Complex World
(NewYork, NY: Irwin/McGraw-Hill, 2000).
Investment in
Capability
Capability
Capability
Erosion
Time Spent
Working Actual
Performance
Performance
Gap
Desired
Performance
Time Spent on
Improvement
DELAY
+
+
+
+
–

it was rare to ﬁnd a process performing above expectations. Instead, managers,
workers, and engineers usually faced high and rising demands, sometimes
despite downsizing and cuts in resources. They were constantly searching for
ways to improve and close the performance gap. Since most organizations are
reluctant to increase plant and equipment or hire more staff, managers hoping
to close a performance gap have only two basic options.
First, they can try to increase the amount of time people actually spend
working. Figure 2 shows this option, which forms a balancing feedback, the Work
Harder loop B1. The process represented by this loop works as follows: Managers
facing a performance gap are under pressure to increase performance. They
pressure people to spend more time and energy doing work. An increase in the
time spent working increases the performance of the process and closes the per-
formance gap. This structure is called a balancing feedback loop because it con-
stantly works to balance desired and actual performance.
Pressure to do Work includes, most obviously, direct measures such as
telling people to work faster or put in overtime, setting more aggressive targets
for throughput, and imposing more severe penalties for missing those targets.
Pressure also includes more subtle actions designed to extract greater effort from
employees. These include the frequency with which performance is reviewed,
the detail with which the reviews are conducted, and the seniority of those
doing the reviewing. At one company we studied, it was not unusual for senior
FIGURE 2. The Work Harder Balancing Loop
Note:The loop identiﬁer, B1, indicates a negative (balancing) feedback. See J. Sterman, op. cit.
Investment in
Capability
Capability
Capability
Erosion
Time Spent
Working Actual
Performance
Performance
Gap
Desired
Performance
Time Spent on
Improvement
DELAY
+
+
+
+
+
+
–
Pressure to
Do Work
B1
Work Harder

vice-presidents to review the performance of individual machines on the factory
floor. Not surprisingly, such attention sent a strong message to all involved: keep
the machines busy at all costs. Similarly, a project manager we interviewed
recalled that when a subsystem for which he was responsible fell behind sched-
ule, his boss required him to call in every hour with a status report until the pro-
totype met its specifications.
A second option to close a performance gap is to improve the capability of
the process. In Figure 3 we represent this option as another balancing feedback
process, the Work Smarter loop B2. Here, managers respond to a performance
shortfall by increasing the pressure on people to improve capability. They may
launch improvement programs, encourage people to experiment with new
ideas, and invest in training. If successful, these investments will, with time,
yield improvements in process capability, boost throughput, and close the per-
formance gap. Of course, everyone knows that it is better to work smarter than
to work harder: An hour spent working produces an extra hour’s worth of out-
put, while an hour spent on improvement may improve the productivity of
every subsequent hour dedicated to production. Yet, despite its obvious and
documented benefits, working smarter does have limitations. First, as shown
in the diagram, there is often a substantial delay between investing in improve-
ment activities and reaping the benefits. Further, the greater the complexity of
FIGURE 3. The Work Smarter Balancing Loop
Investment in
Capability
Capability
Capability
Erosion
Time Spent
Working Actual
Performance
Performance
Gap
Desired
Performance
Time Spent on
Improvement
+
+
+
+
+
+
+
+
–
Pressure to
Do Work
B1
Work Harder
B2
Work Smarter
Pressure to
Improve
Capability
DELAY

the process, the longer it takes to improve.8
Second, investments in capability
can be risky. Improvement efforts don’t always find the root cause of defects,
new tools sometimes don’t produce the desired gains, and experiments often
fail. While investments in capability might eventually yield large and enduring
improvements in productivity, they do little to solve the problems managers face
right now.
Thus, it is not surprising that managers frequently use the Work Harder
loop to both accommodate variations in daily workload and solve pressing prob-
lems created by unexpected breakdowns or defects. When a manufacturing line
serving an important customer goes down, a manager is unlikely to react by
sending the work team to training in reliability improvement. Instead, that
manager is going to get the line running and push for overtime until the ship-
ment is out the door. Of course when the line is back running and the product
has been shipped, the manager should return attention to the improvement
activities that will prevent future breakdowns, and make up for the improve-
ment time that was lost during the crunch. However, it doesn’t usually happen.
Instead, what we repeatedly observe, and what is more difficult to understand,
are organizations in which working harder is not merely a means to deal with
isolated incidents, but is instead standard operating procedure. Rather than
using the work harder loop to occasionally offset daily variations in workload,
managers, supervisors, and workers all come to rely constantly on working
harder to hit their targets and, consequently, never find the time to invest
in improvement activities. What starts as a temporary emphasis on working
harder quickly becomes routine.
The Reinvestment Loop
To understand why, it is helpful to consider how working smarter and
working harder are connected. The most important interconnection arises
because organizations rarely have excess resources. Increasing the pressure to
do work leads people to spend less time on non-work related activities like
breaks and to put in overtime (that is, they use the Work Harder loop). For
knowledge workers such overtime is often unpaid and spills into nights and
weekends, stealing time from family and community activities. There are, how-
ever, obvious limits to long hours. After a while there is simply no more time.
If the performance gap continues to rise, workers have no choice but to reduce
the time they spend on improvement as they strive to meet their ever-increasing
objectives. Figure 4 adds the connection between pressure to do work and the
amount of time spent on improvement.
The additional link creates the Reinvestment loop. Unlike those described
so far, the Reinvestment loop is a positive feedback that tends to reinforce which-
ever behavior currently dominates. An organization that successfully improves
its process capability will experience rising performance. As the performance
gap falls, workers have even more time to devote to improvement, creating a
virtuous cycle of improved capability and increasing attention to improvement.

Conversely, if managers respond to a throughput gap by increasing work pres-
sure, employees increase the amount of time spent working and cut the time
spent on improvement. Capability begins to decay. As capability erodes, the per-
formance gap grows still more, forcing a further shift towards working harder
and away from improvement. Here the reinvestment loop operates as a vicious
cycle, driving the organization to ever-higher degrees of work pressure and min-
imal levels of process capability. Not surprisingly, such a vicious cycle quickly
drives out meaningful improvement activity. Here, for example, is the way a
manager in an electronics assembly plant explained the persistent failure of the
organization to engage in process improvement:
“Supervisors never had time to make improvements or do preventative mainte-
nance on their lines . . . they had to spend all their time just trying to keep the
line going, but this meant it was always in a state of flux, which, in turn, caused
them to want to hold lots of protective inventory, because everything was so
unpredictable. A quality problem might not be discovered until we had produced
a pile of defective parts. This of course meant we didn’t have time to figure out
why the problem happened in the first place, since we were now really behind
our production schedule. It was a kind of snowball effect that just kept getting
worse.”
FIGURE 4. The Reinvestment Reinforcing Loop
Investment in
Capability
Capability
Capability
Erosion
Time Spent
Working Actual
Performance
Performance
Gap
Desired
Performance
Time Spent on
Improvement
+
+
+
+
+
+
+
+
–
–
Pressure to
Do Work
B1
Work Harder
B2
Work Smarter
R1
Reinvestment
Pressure to
Improve
Capability
DELAY
Note:The loop identifier, R1, indicates a positive (reinforcing) feedback. See J. Sterman, op. cit.

Shortcuts and the Capability Trap
The Reinvestment loop means a temporary emphasis on one option at the
expense of the other is likely to be reinforced and eventually become perma-
nent. Organizations that invest in improvement will experience increasing capa-
bility and find that they have more time to allocate to working smarter and less
need for heroic efforts to solve problems by working harder. In the successful
initiatives we studied, leadership often worked to strengthen the reinvestment
process by explicitly allocating the resources freed up by productivity gains to
further improvement. Unfortunately, however, these initiatives were the excep-
tion rather than the rule. In most of the organizations in our study the reinvest-
ment loop worked as a vicious cycle and prevented improvement programs from
getting off the ground. Even when improvement programs yielded initial results,
cost and schedule pressures soon tempted many organizations into downsizing
or higher performance goals that drained resources away from improvement,
weakening the reinvestment loop and causing capability to stall or even fall.9
Understanding why the reinvestment loop typically worked in the down-
ward, vicious direction rather than the upward, virtuous direction requires that
we add a final link to the model (see Figure 5). As discussed above, cutting
investments in maintenance and improvement in favor of working harder
erodes process capability and hurts performance. However, capability does not
drop right away. It takes time for process integrity to depreciate. In the mean-
time, the decision to skimp on improvement—skipping improvement team
meetings, neglecting to take machines down for scheduled maintenance, or
ignoring documentation requirements—boosts the time available to get work
done right now. We capture this interconnection by adding a negative link
between Time Spent on Improvement and Time Spent Working. When the perfor-
mance gap rises and managers resort to increased work pressure, overworked
people cut back improvement activity to free still more time for production. The
performance gap falls, closing a third feedback that works to balance desired and
actual performance. We label this the Shortcuts loop (B3) to capture the idea that
increased throughput comes at the cost of departing from standard routines and
processes, cutting corners, and reducing the time spent on learning and
improvement.
Shortcuts are tempting because there is often a substantial delay between
cutting corners and the consequent decline in capability. For example, supervi-
sors who defer preventive maintenance often experience a “grace period” in
which they reap the benefits of increased output (by avoiding scheduled down-
time) and save on maintenance costs. Only later, as equipment ages and wears
do they begin to experience lower yields and lower uptimes (see section 4). Sim-
ilarly, a software engineer who forgoes documentation in favor of completing a
project on time incurs few immediate costs; only later, when she returns to fix
bugs discovered in testing does she feel the full impact of a decision made weeks
or months earlier.10
Thus, the Shortcuts loop is effective in closing the throughput

gap only because capability does not change immediately when the time dedi-
cated to learning and improvement declines.
To illustrate these dynamics, Figure 6 shows two simulations of the model
in which we show how a hypothetical process reacts to working harder versus
working smarter. Both simulations begin in the same equilibrium state. The first
simulation shows the response to an increased emphasis on working harder. As
more effort is dedicated to work, gross throughput immediately rises. Time spent
improving falls immediately, but capability does not. Performance therefore
rises. The benefit of working harder is, however, short-lived. With less time
devoted to improvement, capability gradually erodes, eventually more than
offsetting the increased time spent working. Working harder creates a “better-
before-worse” situation. Conversely, as seen in the second simulation, increasing
the time spent on improvement reduces output in the short run. Eventually,
however, capability rises more than enough to offset the drop in work effort
and performance is permanently higher, a “worse-before-better” dynamic.
The interaction between the balancing Shortcuts loop and the reinforcing
reinvestment loop creates a phenomenon we call the Capability Trap and helps
explain why organizations often find themselves stuck in a vicious cycle of de-
clining capability. Managers and workers in need of an immediate performance
FIGURE 5. The Shortcuts Balancing Loop
Investment in
Capability
Capability
Capability
Erosion
Time Spent
Working Actual
Performance
Performance
Gap
Desired
Performance
Time Spent on
Improvement
+
+
+
+
+
+
+
+
–
–
–
Pressure to
Do Work
B1
Work Harder
B2
Work Smarter
R1
Reinvestment
B3
Shortcuts
Pressure to
Improve
Capability
DELAY

boost can get it by skimping on improvement and maintenance. However, capa-
bility eventually declines, causing the Reinvestment loop to work as a vicious
cycle. Managers who rely on working harder and shortcuts to meet immediate
throughput needs soon ﬁnd the process falling short of its objectives, requiring a
further shift towards working harder and away from improvement. To see the
capability trap in action, consider how a manufacturing supervisor in an auto
FIGURE 6. Simulations of the Working Harder and Working Smarter Strategies
SYSTEM RESPONSE TO:
Time Spent Improving
Time Spent Working
Time Spent Working
Time Spent Improving
→ Time → Time
→ Time → Time
→ Time → Time
WORKING HARDER WORKING SMARTER
Actual Performance Actual Performance
Effort Effort
Capability Capability
؄ ؄
‫؂‬ ‫؂‬

company explained the inability of her organization to make a commitment to
regular improvement activities:
“In the minds of the [operations team leaders] they had to hit their pack counts.
This meant if you were having a bad day and your yield had fallen . . . you had
to run like crazy to hit your target. You could say, ‘you are making 20% garbage,
stop the line and fix the problem,’ and they would say, ‘I can’t hit my pack count
without running like crazy.’ They could never get ahead of the game.”
By keeping the line going rather than stopping to fix the problem, these team
leaders relied on the Shortcuts loop to hit their throughput objectives. However,
by “running like crazy” they also caused the Reinvestment loop to operate as a
vicious cycle, driving the line to a minimal level of capability and forcing them
to run ever faster.
The capability trap is not limited to manufacturing—we have observed
it in firms ranging from financial services to construction. For example, the
capability trap prevented a product development organization we studied from
developing new processes that would have increased productivity. Like many
firms, they sought to create an engineering library or “bookshelf” of reusable
designs and software. However, as described by an engineering manager,
“An engineer might not take the time to document her steps or put the results of
a simulation on the bookshelf and because of that she saved engineering time and
did her project more efficiently. But in the long run it prevented us from being
able to deploy the reusability concepts that we were looking for.”
Just as machine operators and supervisors in the first example faced a basic
trade-off between producing and improving, development engineers were forced
to trade off getting their assigned tasks done against documenting what they
learned so that others might benefit. Engineers could make more rapid progress
towards their objectives by taking shortcuts and ignoring the bookshelf, but
doing so prevented them from initiating the self-reinforcing reinvestment loop
that would have led to improved process capability.
The Persistence of the CapabilityTrap
Because working harder and taking shortcuts produce more immediate
gains and help solve today’s problems, managers unaware of the inherent
“better before worse” trade-off are likely to choose them over working smarter.
Unfortunately, these temporary gains come at the expense of the long-run
health of the process. By pressuring people to work harder, managers often un-
wittingly force their organizations into the capability trap where ever-increasing
levels of effort are required to maintain performance. Of course, this phenome-
non is not limited to large organizations. Many readers will recognize this
dynamic in different aspects of their personal lives. In situations ranging from
learning how to use a new software package to committing to a new exercise

program, we often fail to do the things that will improve our long-run produc-
tivity and well-being due to the short-run stresses of other obligations.
A question naturally arising at this point is: “Wouldn’t managers eventu-
ally figure this out?” While it is understandable that, on occasion, people get
caught in the capability trap, wouldn’t they eventually realize the true source of
their problems and rebalance their efforts between working harder and working
smarter? Unfortunately, the data suggest that overcoming the capability trap is
rare. Managers often do not realize how deeply they are trapped in it. Instead,
the lessons that people learn when caught in the capability trap often lead to
actions that make the situation worse.
Faulty Attributions
Suppose you are a manager faced with inadequate performance. Your
operation is not meeting its objectives and you have to do something about it.
As we have outlined so far, you have two basic choices: get people to work
harder or get them to work smarter. To decide, you have to make a judgment
about the cause of the low performance. If you believe the system is underper-
forming due to low capability, then you should focus on working smarter. If, on
the other hand, you think that your workers or engineers are a little lazy, undis-
ciplined, or just shirking, you need to get them to work harder.
How do you decide? Research suggests that people generally assume
that cause and effect are closely related in time and space: To explain a puzzling
event, we look for another recent, nearby event that might have triggered it.
People also tend to assume each event has a single cause, underestimate time
delays, and fail to account for feedback processes. How do these causal attribu-
tions play out in a work setting? Consider a manager observing a machine oper-
ator who is producing an unusually high number of defects. The manager is
likely to assume that the worker is at fault: The worker is close in space and time
to the production of defects, and other operators have lower defect rates. The
true cause, however, may be distant in space and time from the defects it cre-
ates. Perhaps the defect is actually the result of an inadequate maintenance
procedure or the poor quality of the training program. In this case, the delay
between the true cause and the defective output is long, variable, and often
unobservable. As a result, managers are likely to conclude that the cause of low
throughput is inadequate worker effort or insufficient discipline, rather than
features of the process. The attribution of a problem to the characteristics—and
character flaws—of individuals in a system rather than to the system in which
they find themselves is so pervasive that psychologists call it the “fundamental
attribution error.”11
Suppose managers conclude that people, not the process, are the source
of low performance. Having made such an attribution it makes sense to increase
production pressure. As discussed above, an increase in production pressure
has two effects. Worker effort immediately rises, closing the performance gap
as the manager intended. However, workers are now less able to achieve their

objectives by increasing the time they spend working. To continue to hit their
ever-increasing targets, they eventually resort to shortcuts, cutting the time
spent on improvement. However, as highlighted above, the Shortcuts loop, while
having the desired effect in the short run, yields a long-run side effect. With less
effort dedicated to improvement, capability begins to decline. Performance falls,
offsetting the initial gains. By continually increasing throughput objectives in
the pursuit of better performance, managers who mistakenly attribute low per-
formance to the attitudes and dispositions of their workforce inadvertently force
the system into the capability trap.
Superstitious Learning
The bias towards blaming people rather than the system in which those
people are embedded means managers are prone to push their organizations
into the capability trap. As workers spend more and more of their time on
throughput and cut back on fundamental improvement, shouldn’t managers
realize that the true cause of sub-standard performance is low process capability
rather than unmotivated workers? Unfortunately, in many situations managers
learn the opposite lesson.
Managers cannot observe all the activities of the workers. Hence, after
they apply production pressure, they cannot easily determine how much of
the resulting rise in throughput is due to increased work effort (the Work Harder
loop) and how much to cutting back on training, improvement or maintenance
(the Shortcuts loop). For example, suppose there is a performance gap requiring
an additional six hours of productive effort per person per week. Managers,
believing employees are simply not working hard enough, increase production
pressure. Workers buckle down, cutting back on breaks, web-surfing and other
nonproductive time. Suppose these responses yield only two hours per person
per week. To close the remaining throughput gap, workers resort to shortcuts
and gradually reduce the time they spend on process improvement, training,
and experimentation until they free the needed four hours per week. Managers
observe that throughput rises by the equivalent of six hours of productive effort.
Because managers do not fully observe the reduction in training, experi-
mentation, and improvement effort (they fail to account for the Shortcuts loop),
they overestimate the impact of their get-tough policy, in our example by as
much as a factor of three. The feedback managers receive does not correct the
error. To the contrary, managers quickly learn that boosting production pressure
works—throughput rose when they turned up the pressure. The gains resulting
from production pressure provide powerful evidence confirming their suspicions
that workers were not giving their full effort.
We call this syndrome the Self-Confirming Attribution Error: Once managers
decide that the workforce is the source of their difficulties, they take actions
that provide convincing and immediate evidence confirming this erroneous
attribution. The cycle of self-confirming attributions drives the organization to
higher levels of production pressure and fewer resources dedicated to process

improvement. Far more importantly, however, it gradually changes the mental
models of the managers by providing them with increasingly compelling evi-
dence that the source of low throughput can be found in the poor attitudes and
weak character of the workforce. Recall the project manager discussed above
who was required to provide hourly status reports on a balky prototype. Soon
afterward the problem was solved, confirming the boss’s belief that he had acted
appropriately, indeed had decisively taken charge of the situation, even though
the team was already working around the clock and his interference drained
precious time from their efforts to solve the problem.
More subtly, the long-run effects of production pressure also reinforce
managers’ belief that workers are the problem. The delay between increased
production pressure and increased throughput (via the Work Harder and Shortcuts
loops) is short, and the connection between work effort and output is unam-
biguous. In contrast, the erosion of process capability caused by production
pressure is delayed, gradual, and diffuse. It is distant in time and space from its
cause. Managers are unlikely to attribute the cause of a throughput gap to the
pressure they placed on workers months or even years before. Instead, they
are likely to conclude that the workers have once again slacked off, requiring
another increase in production pressure.
Workers often unwittingly conspire in strengthening the managers’ attri-
butions. Faced with intense production pressure, people are naturally reluctant
to tell supervisors they can’t meet all their objectives. The more effectively work-
ers cover up the shortcuts they take to meet their throughput targets, the less
aware managers will be of the long-run costs of production pressure. Unaware
that improvement activity, maintenance, and problem solving have been cut,
throughput appears to rise without requiring any sacrifices, reinforcing manage-
ment’s attribution that the workers really were not working hard enough. When
managers eventually discover these shortcuts, their view of workers as untrust-
worthy is confirmed. Managers are then, as they see it, forced to monitor worker
effort even more closely (e.g., more frequent status reports, stiffer penalties for
missing targets, software for monitoring key-stroke rates of data entry opera-
tors). What starts as an erroneous attribution about the skills, effort, and char-
acter of the workers becomes true. Managers’ worst fears are realized as a
consequence of their own actions.
Consistent with our theory, we are not attributing these dynamics to
unskilled, inexperienced, or ill-intentioned managers. Rather, the structure of
the system inadvertently leads even many talented and dedicated managers into
the capability trap, while at the same time providing compelling evidence that
the sources of their difficulties lie in factors beyond their control, such as lazy
workers, a “difficult” union, faulty machinery, or fickle customers. Managers are
unlikely to escape the capability trap because they rarely realize they are in it.
Instead, as capability stagnates despite repeated attempts at improvement, they
slowly, perhaps reluctantly, but with increasing conviction, come to believe
that their problems lie in the attitudes and character of the people that work for

them. Having made such an attribution, the actions they take, while rational
from their perspective, make the situation worse.
How Superstitious Learning Thwarts Improvement Programs
What happens when an organization stuck in the capability trap attempts
to implement an improvement program? Performance is low and work pressure
intense. In such environments, improvement programs add to the workload—
the organization is so far behind that it cannot afford to cut back throughput.
Indeed, in many organizations, management imposes aggressive stretch objec-
tives for both throughput and improvement in the belief that aggressive goals
are needed to shake things up and motivate people. In one firm we studied,
the general manager laid out his goals for improving the product development
process by saying:
“We need a development process that is fast, is the best in the industry, and it
needs to increase throughput by 50% in two years. And everyone must adhere
to the same process.”
At the same time, they launched many new development projects in
anticipation of the expected productivity gains. Viewed through the lens of man-
agement’s mental model these decisions were entirely rational. However, that
mental model, conditioned by the self-confirming attribution error dynamics
discussed above, led them to the erroneous belief that the delay between
improvement effort and results was short and that their engineers were under-
utilized, undisciplined, unmotivated, and unwilling to adhere to the specified
process.
The company spent millions and invested countless person-hours to
create a new product development process. The new process included better
technical tools, such as improved CAD/CAE/CAM systems, but also increased
monitoring, including a structured stage-gate review process and mandated use
of project management software. While there were some pockets of success, in
most cases the effort had little impact. The leaders of the change effort often
attributed its failure to the engineers’ lack of discipline:
“Engineers—by trade, definition, and training—want to forever tweak things.
It’s a Wild West culture.”—Manager A
“We went through a period where we had so little discipline that we really had
the ‘process du jour.’ Get the job done and how you did it was up to you.”
—Manager B
“A lot of the engineers felt that [the new process] was no value-add and that
they should have spent all their time doing engineering and not filling out project
worksheets. It’s brushed off as bureaucratic.”—Manager A
“It was fair to say that a lot of engineers viewed this as a neat way to get some
fancy tools and to hell with process.”—Manager C

Yet, when we asked engineers why the effort failed, we got a different
story:
“We never had time to take the courses and get the equipment we needed to
really make this stuff work. . . . It was really exhausting trying to learn how to
use the tools and do the design at the same time. ”—Engineer A
“People had to do their normal work as well as [use the new project management
system]. There just weren’t enough hours in the day, and the work wasn’t going
to wait.”—Engineer B
“Under this system . . . the new workload was all increase. . . . In some cases your
workload could have doubled.”—Engineer C
“How did we catch up? We stayed late. Most of the team was working from 7:00
a.m. to 8:00 p.m. and on weekends. A lot of people worked right through the
Christmas vacation.”—Engineer D
“The new process is a good one. Someday I’d like to work on a project that actu-
ally uses it.”—Engineer E
While managers felt the engineers had little interest in following the
process, engineers became increasingly frustrated with leaders they felt had no
understanding of what was really required to develop new products. Faced with
the double bind of hitting aggressive performance targets and equally aggressive
improvement targets, they were forced to cut corners while still appearing to
follow the process. As one engineer remarked,
“In many ways we worked around the [new] system. Good, bad, or indifferent
that’s what happened. We had a due date and we did whatever it took to hit it.”
As management discovers the engineers’ shortcuts and workarounds, their view
that the engineers can’t be trusted is confirmed, and they are forced to step up
their monitoring. Faced with similar difficulties in its effort to implement a new
product development process, a different firm even created a cadre of “compli-
ance managers” whose sole job was to enforce adherence to their new develop-
ment process.
Workers in such organizations quickly learn to hide problems from oth-
ers. In all of the organizations we studied, engineers routinely neglected to
reveal the existence of serious design issues for fear of retribution from man-
agers. In one firm, the motto of the development engineers was “never reveal
you have a problem until you also have the solution.” In another, engineers
called the weekly progress review meetings the “liars’ club”—each participant
overstated the progress of his subsystem and hid known defects from others in
the hope that others would be discovered first, giving them time to catch up.12
The consequence is long delays in the discovery of needed rework, greatly
increasing costs, delaying launch, and, often, compromising quality.
The capability trap goes beyond low capability and high work pressure.
Eventually it gets embedded in deeper structures, including incentives and

corporate culture. As organizations grow more dependent on firefighting and
working harder to solve problems caused by low process capability, they reward
and promote those who, through heroic efforts, manage to save troubled pro-
jects or keep the line running. Consequently, most organizations reward last-
minute problem solving over the learning, training, and improvement activities
that prevent such crises in the first place. As an engineer at an auto company
told us, “Nobody ever gets credit for fixing problems that never happened.” Over
time, senior management will increasingly consist of these war heroes, who are
likely to groom and favor other can-do people like themselves. As described by
a project leader we interviewed,
“Our [company] culture rewards the heroes. Frankly, that’s how I got where
I’ve gotten. I’ve delivered programs under duress and difficult situations and
the reward that comes with that is that you are recognized as someone that can
deliver. Those are the opportunities for advancement.”
Thus incentives and culture not only reinforce the tendency toward short-run
thinking and working harder, but also are themselves shaped by that very short-
term focus and work-harder mentality, creating another reinforcing feedback
that intensifies the capability trap.13
An organization suffering from the self-confirming attribution error is
poorly positioned to escape the capability trap. Improvement programs add
stress to the organization, triggering greater work pressure that prevents people
from investing in improvement, and encourages shortcuts. In such organizations
many improvement programs never get off the ground. If, despite the work
pressure, people do succeed in allocating more time to improvement, the result
is a short-term drop in performance as time spent working falls before the
investments in improvement bear fruit. Observing that performance is not
improving, managers conclude the particular improvement method is not work-
ing and abandon it. Since the need to improve remains, they search for another,
more promising tool, only to find it too suffers a similar fate. The result is grow-
ing cynicism among employees about “flavor of the month” programs.
More insidiously, these dynamics strengthen stereotypes and conflicts
that not only hurt organizational performance but damage society. Consider,
for example, how a senior manager explained why the product development
improvement effort he ran had failed:
“Program management and the disciplines associated with it continue to be a
problem in my opinion in most western cultures. The people that are particularly
rigorous and disciplined, the Japanese and the Germans, tend to be so by cultural
norms. I can’t tell you if it’s hereditary or society or where it is they get it but the
best engineers are those that tend to be the most disciplined, not as individual
contributors but as team-based engineers. So there’s a strong push back from the
western type of engineer for much of this.”
There is no mention of the structural features of the system or the pres-
sure, felt throughout the organization, to deliver ambitious projects on time.

Instead, this manager blames the failure on the undisciplined character of
“Western” engineers. Such attributions, here generalized to entire national
groups, and invoking a disturbing racial and ethnic subtext, are typical of the
fundamental attribution error. As these attributions are shared and repeated
they become institutionalized. They become part of the corporate culture, and,
as suggested by the quote above, can strengthen pernicious stereotypes and prej-
udices in society at large.
Overcoming the CapabilityTrap
So what can be done? The most important implication of our analysis
is that our experiences often teach us exactly the wrong lessons about how to
maintain and improve the long-term health of the systems in which we work
and live. Successful improvement must include a significant shift in the mental
models of those both leading and participating in an improvement effort. This
insight was captured succinctly by one manager in a successful improvement
effort:
“There are two theories. One says, ‘there’s a problem, let’s fix it.’ The other says
‘we have a problem, someone is screwing up, let’s go beat them up.’ To make
improvement, we could no longer embrace the second theory, we had to use
the first.”
Once the cycle of self-confirming attributions is broken, any number of process
improvement tools and methods can help improve capability. Without this shift,
new tools and techniques, no matter how great their potential, are unlikely to
succeed.
Breaking the cycle of self-confirming attributions is not easy, but it can
be done. The following are examples from two organizations that overcame
these difficulties and introduced successful improvement efforts.14
Du Pont
In 1991, an in-house benchmarking study documented a gap between
Du Pont’s maintenance record and those of the best performing companies in
the chemicals industry. The benchmarking study revealed an apparent paradox:
Du Pont spent more on maintenance than industry leaders but got less for it.
Du Pont had the highest number of maintenance employees per dollar of plant
value, yet its mechanics worked more overtime. Spare parts inventories were
excessive, yet they relied heavily on costly expedited procurement of critical
components. Overall, Du Pont spent 10-30% more on maintenance per dollar
of plant value than the industry leaders, while overall plant uptime was some
10-15% lower.
An experienced manager, Winston Ledet, and a team charged with
improving maintenance operations, developed a system dynamics model of
these issues. The modeling process involved extensive hands-on workshops in

which the team, assisted by an experienced modeler, discussed, tested, and
changed the model as they identified areas needing improvement. Using the
model as a laboratory to design and test different policies, the team gradually
developed an appreciation for the capability trap and the paradox of high main-
tenance costs and low reliability.
To see how the capability trap arose in the chemicals industry, imagine
the effects of cost cuts on maintenance, such as those beginning with the oil
crisis of 1973 and subsequent recession. In chemical plants, when critical equip-
ment breaks down, it must be fixed. Hence maintenance managers required
to reduce costs must cut preventive maintenance, training, and investments in
equipment upgrades. The drop in planned maintenance eventually causes
breakdowns to increase, forcing management to reassign more mechanics from
planned maintenance to repair work. Breakdowns then rise even more. As
uptime falls, operators find it harder to meet demand and become less willing to
take equipment down for scheduled maintenance, leading to more breakdowns
and still lower uptime. More breakdowns simultaneously constrain revenue (by
lowering production) and increase costs (due to overtime, expedited parts pro-
curement, the nonroutine and often hazardous nature of outages, collateral
damage, and so forth). More subtly, lower uptime erodes a plant’s ability to meet
its delivery commitments. As it develops a reputation for poor delivery reliabil-
ity, business volume and margins fall further. The plant slowly slides into the
capability trap, with high breakdowns, low uptime, and high costs.
Policy analysis showed that escaping the capability trap necessarily meant
performance would deteriorate before it could improve: While continuing to
repair breakdowns, the organization has to invest additional resources in
planned maintenance, training and part quality, raising costs. Most importantly,
increasing planned maintenance reduces uptime in the short run because opera-
ble equipment must be taken off-line for the planned maintenance to be done.
Only later, as the Reinvestment loop begins to work in the virtuous direction, does
the breakdown rate drop. Fewer unplanned breakdowns give mechanics more
time for planned maintenance. As maintenance expenses drop the savings can
be reinvested in training, parts quality, reliability engineering, planning and
scheduling systems, and other activities that further reduce breakdowns. For
example, upgrading to a more durable pump seal improves reliability, allowing
maintenance intervals to be lengthened and inventories of replacement seals to
be cut. Higher uptime also yields more revenue and provides additional
resources for still more improvement. All the positive feedbacks that once acted
as vicious cycles dragging reliability down become virtuous cycles, progressively
and cumulatively boosting uptime and cutting costs.
Now the challenge facing the team was implementation. They knew
nothing could happen without the willing participation of thousands of people,
from the lowest-grade hourly mechanic to regional vice presidents. They also
realized that their views had changed because they had participated in the

modeling process. Somehow they had to facilitate a similar learning process
throughout the plants.
The team converted the maintenance model into an interactive role-
playing simulation they called the Manufacturing Game.15
The game is closely
based on the model and realistically captures the time delays, costs, and other
parameters characterizing typical plants. They embedded the game in an inter-
active workshop designed to create an environment for learning that addressed
emotional as well as cognitive issues. The process at Du Pont’s Washington
Works complex in Parkersburg, West Virginia, was typical:
The team was initiated with a two-day learning lab . . . learning the concepts of
defect elimination and experiencing the Manufacturing Game. . . . The material
is presented in the form of lectures, skits and participative exercises in an off-site
environment. Posters and music are used. The atmosphere is much different than
routine plant meetings or training, to open up their thinking. . . . Through inter-
active exercises, the team develops their personal aspirations for improving the
area where they have chosen to work. . . . [Then] they . . . develop an action
plan to immediately start working.16
Despite its many simplifications, the game quickly becomes in many ways a
real plant with real emotions and conflicts among players. Initialized with high
breakdowns and low uptime, the people playing the role of operations managers
face intense pressure to keep equipment running and often rebuff attempts to
increase planned maintenance, just as in the real world. Players who stick with
the prevailing cost-minimization, work-harder, reactive maintenance policies
can keep costs low for a while. However, as defects accumulate, uptime slowly
sinks while costs rise. Teams who follow a planned maintenance strategy first
find costs rise while uptime falls. Soon, however, costs begin to fall and uptime
rises. The game allows people to experience the worse-before-better dynamic in
a few hours instead of a few months. For many, the game was the first time in
their careers they experienced the possibility that improvement was actually
possible.
The game and learning laboratory proved popular. However, playing it
once was not enough. The team found that they had to run several workshops
for a given plant before a critical mass emerged to lead action teams and put
proactive maintenance policies into practice. Individual plants needed the capa-
bility to run the game so their own people, with their site-specific experience
and legitimacy, could run it on demand. By the end of 1992, some 1200 people
had participated in the workshop, and more than 50 facilitators had been
certified.
At plants that implemented the program by the end of 1993, the mean
time between failure (MTBF) for pumps (the focus of the program) rose by an
average of 12% each time cumulative operating experience doubled. Direct
maintenance costs fell an average of 20%. In 23 comparable plants not imple-
menting the program the learning rate averaged just 5% and costs were up an

average of 7%. Washington Works boosted production capability 20%,
improved customer service 90%, and cut delivery lead time by 50%, all with
minimal capital investment and a drop in maintenance costs. For the company
as a whole, conservative estimates exceed $350 million/year in avoided mainte-
nance costs alone.
However, success creates its own challenges. One issue related to the
persistence of the cost-saving mentality. A member of the modeling team com-
mented, “As soon as you get the problems down, people will be taken away
from the effort and the problems will go back up.” In fact, mandated corporate
cost-cutting programs did cause significant downsizing throughout the entire
company, weakening the reinvestment feedback and limiting their ability to
expand the program. Winston Ledet took early retirement and began working
with other companies interested in the game and learning lab. These firms
include other chemicals manufacturers along with firms in the energy, auto-
motive, and high-tech sectors.
British Petroleum
One of the organizations Ledet worked with after leaving Du Pont
was British Petroleum’s refinery in Lima, Ohio. Founded in 1886 by John D.
Rockefeller, and once “Queen of the Fleet,” the refinery engaged in cost cutting
during the 1980s that triggered the vicious cycle of increasing breakdowns,
higher maintenance costs, and less planned maintenance, pushing it into the
capability trap. By the early 1990s, Lima lagged well behind other U.S. refiner-
ies. BP began to think about selling or closing the facility.
In 1994, the Lima facility introduced the maintenance learning lab and
other system dynamics tools. It was not a top management intervention: The
original champions were an equipment specialist, a maintenance training super-
visor, and an engineer. Successful pilot projects led to favorable word of mouth;
eventually 80% of all employees participated in the program. Soon dozens of
improvement teams were in place. During the first six months, maintenance
costs ballooned by 30%. Having experienced it in the game, management was
prepared for the worse-before-better dynamic, and focused on the
improvements generated by the action teams.
In January 1996, BP announced that it intended to sell the Lima refinery
and stepped up its cost cutting and downsizing. A few months later BP stunned
the employees by announcing that it could not find a buyer at a satisfactory
price and would therefore close the refinery. The announcement was a deep
blow to the workers and the community. One of the most important businesses
in the community, the refinery employed 450 people and pumped more than
$60 million per year into Lima’s depressed economy. Some employees became
discouraged and questioned the value of the learning lab and improvement pro-
gram. A few transferred to other BP facilities or left altogether. Winston Ledet
described what happened next:

“For those who decided to stay with the ship, a new spirit emerged. They realized
that they needed a future in Lima and should take responsibility for creating that
future. The first step was to ensure that the exit of many experienced people did
not throw them back in the reactive mode. . . . It actually created a clearer focus
for the people who remained. They were all there because they had chosen to be
there.”
Soon the impact of the new maintenance policies and attitudes was clearly
visible (Table 1).
These dramatic improvements did not go unnoticed. On July 2, 1998,
the banner headline of the Lima News announced “Oil Refinery Rescued.” Clark
USA, a privately held Fortune 500 company with refining and distribution inter-
ests, agreed to buy the Lima refinery for $215 million and keep it operating.
The Du Pont and BP cases illustrate the power of a shift in mental models.
The model, game, and workshop do not teach anyone how to maintain equip-
ment. For example, a BP team reduced butane flare-off to zero, saving $1.5 mil-
lion/year and reducing pollution. The effort took two weeks and cost $5000, a
return on investment of 30,000%/year. Members of the team had known about
the problem and how to solve it for eight years. They already had all the engi-
neering know-how they needed, and most of the equipment and materials were
already on site. What had stopped them from solving the problem long ago? The
only barrier was the mental model that there were no resources or time for
improvement, that these problems were outside their control, and that they
could never make a difference.
The modeling process and the resulting game were effective because they
eliminated many of the impediments to learning in the real system. Dynamics
TABLE 1. Improvement at the Lima Refinery
• Pump MTBF up from 12 to 58 months (failures down from more than 640 in 1991 to 131 in 1998).
Direct savings: $1.8 million/year.
• Hydrocarbon flare-off down from 1.5% to 0.35%, saving $0.27/barrel and improving environmental
quality.
• On-line analyzer uptime improved from 75% and not trusted to 97% and trusted, permitting real-time
optimization of product flow. Savings: $0.10-0.12/barrel.
• Safety incidents and lost hours cut by a factor of 4.
• Thirty-four production records set.
• Cash margin improved by $0.77 per barrel of oil processed.
• Total new value created: $43 million/year.Total cost: $320,000/year. Ratio: 143:1.
• Learning initiative under way for other BP facilities around the world.
Source: Paul Monus,“Proactive Manufacturing at BP’s Lima Oil Refinery,” presented at National Petroleum Refiners Association
Maintenance Conference, May 20-23, 1997, New Orleans; J. Griffith, D. Kuenzli, and P. Monus,“Proactive Manufacturing:Accelerating
Step Change Breakthroughs in Performance,” NPRA Maintenance Conference, MC-98-92, 1998; Paul Monus, personal
communication.

such as the progressive slide into the capability trap that normally play out over
years or even decades could be experienced in just a few hours. Unlike the real
world, people could take different roles: A mechanic playing the role of plant
manager might find himself with low uptime and then cut preventive mainte-
nance to avoid equipment takedowns and cut costs. Seeing people from differ-
ent functions and backgrounds enacting the same behaviors helped break the
vicious cycle of self-confirming attribution errors and blame. The systems think-
ing process enabled people to experience for themselves the long-term, organi-
zation-wide consequences of their actions. They discovered how to use initial
successes to create resources for further improvement and how to survive the
short-run drop in performance. They saw how small actions could snowball into
major gains. Most importantly, they learned that they could, after all, make a
difference.
Notes
1. See G. Easton and S. Jarrell, “The Effects of Total Quality Management on Corpo-
rate Performance: An Empirical Investigation,” Journal of Business, 2 (1998): 253-
307; K. Hendricks and V.R. Singhal, “Quality Awards and the Market Value of the
Firm: An Empirical Investigation,” Management Science, 43/3 (1996): 415-436.
2. See Easton and Jarrell, op. cit.; Darrell Rigby, “Management Tools and Techniques:
A Survey,” California Management Review, 43/2 (Winter 2001): 139-159.
3. See J. Pfeffer and R. Sutton, The Knowing-Doing Gap (Boston, MA: Harvard Busi-
ness School Press, 2000); K. Klein and J. Sorra, “The Challenge of Innovation
Implementation,” Academy of Management Review, 4 (1996): 1055-1080.
4. Published summaries can be found in: J. Sterman, N. Repenning, and F. Kofman
“Unanticipated Side Effects of Successful Quality Programs: Exploring a Paradox
of Organizational Improvement,” Management Science, 43/4 (1997): 503-521; N.
Repenning and J. Sterman, “Getting Quality the Old Fashion: Self-Confirming
Attributions in the Dynamics of Process Improvement,” in R.B. Cole and R. Scott,
eds., Improving Theory and Research on Quality Enhancement in Organizations (Thou-
sand Oaks, CA: Sage, 2000), pp. 201- 235; E. Keating and R. Oliva, “A Dynamic
Theory of Sustaining Process Improvement Teams in Product Development,” in
M. Beyerlein and D. Johnson, eds., Advances in Interdisciplinary Studies of Teams
(Greenwich, CT: JAI Press, 2000); R. Oliva, S. Rockart, and J. Sterman, “Managing
Multiple Improvement Efforts: Lessons from a Semiconductor Manufacturing
Site,” in D. Fedor and S. Ghosh, eds., Advances in the Management of Organizational
Quality (Greenwich, CT: JAI Press, 1998), pp. 1-55; J. Carroll, J. Sterman, and A.
Markus, “Playing the Maintenance Game: How Mental Models Drive Organiza-
tion Decisions,” in R. Stern and J. Halpern, eds., Debating Rationality: Nonrational
Elements of Organizational Decision Making (Ithaca, NY: ILR Press, 1997).
5. For formal models of implementation see: Sterman, Repenning, and Kofman op.
cit.; N. Repenning, “Drive Out Fear (Unless You Can Drive It In): The Role of
Agency and Job Security in Process Improvement Efforts,” Management Science,
46/11 (2000): 1385-1396; N. Repenning, “A Dynamic Model of Resource Alloca-
tion in Multi-Project Research and Development Systems,” System Dynamics
Review, 16/3 (2000): 173-212; N. Repenning, “A Simulation Based-Approach to
Understanding the Dynamics of Innovation Implementation,” Organization Science
(forthcoming).
6. Repenning and Sterman (2000), op. cit.

7. See A. Schneiderman, “Setting Quality Goals,” Quality Progress (April 1988),
pp. 55-57; J. Sterman, N. Repenning, and F. Kofman, op. cit.
8. Schneiderman, op. cit.
9. For an example, see Sterman, Repenning, and Kofman, op. cit.
10. The dynamics of project management and software development have been
treated extensively in system dynamics. See, for example, T. Abdel-Hamid, “The
Economics of Software Quality Assurance: A Simulation-Based Case Study,” MIS
Quarterly, 3 (1997): 395-411; D. Ford and J. D. Sterman “Dynamic Modeling of
Product Development Processes,” System Dynamics Review, 14 (1998): 31-68.
11. For an excellent summary of attribution theory, see S. Plous, The Psychology of
Judgment and Decision Making (New York, NY: McGraw-Hill, 1993).
12. See also George Roth and Art Kleiner, Car Launch (New York, NY: Oxford Univer-
sity Press, 2000).
13. See N. Repenning et al., “Past the Tipping Point: The Persistence of Fire-Fighting
in Product Development,” published in this issue [California Management Review,
43/4 (Summer 2001)].
14. These cases are discussed in depth in J. Sterman, Business Dynamics: Systems Think-
ing and Modeling for a Complex World (New York, NY: Irwin/McGraw-Hill, 2000),
chapter 2.
15. See Winston Ledet, “Engaging the Entire Organization: Key to Improving Reliabil-
ity,” Oil and Gas Journal, 97/21 (1999): 54-57.
16. R. Tewksbury and R. Steward, “Improved Production Capability Program at Du
Pont’s Washington Works,” Proceedings of the 1997 Society for Maintenance and
Reliability annual conference.

No creditforsolvingproblemsneverhappened

More Related Content

What's hot (19)

Viewers also liked (20)

Similar to No creditforsolvingproblemsneverhappened (20)

Recently uploaded (20)

No creditforsolvingproblemsneverhappened