Engaging the entire organization key to improving reliability

May 24, 1999
Manufacturing Game allows employees to work together to eliminate defects and learn how to translate their ideas into action. Plants can accelerate their maintenance improvements with lower risk by reducing the current workload through the elimination of the defects. Reducing the workload both increases the time available to do further improvements and minimizes the defects that come from poor practices that are inherently part of being in a reactive mode.
Winston J. Ledet
Manufacturing Game Inc.
Manufacturing Game allows employees to work together to eliminate defects and learn how to translate their ideas into action.
Plants can accelerate their maintenance improvements with lower risk by reducing the current workload through the elimination of the defects.

Reducing the workload both increases the time available to do further improvements and minimizes the defects that come from poor practices that are inherently part of being in a reactive mode.

Several large manufacturing and process companies, including Eastman Chemical Co. in Kingsport, Tenn., and British Petroleum plc (BP; now Clark Refining & Marketing Inc.), Lima, Ohio, have used Manufacturing Game Inc.'s interactive workshops and teams to make substantial improvements to their operating and financial performance.

The company uses a "game" which focuses on eliminating defects by engaging front-line employees in a process that communicates the value of improvements, builds passion for the change, aligns them around the right ideas, and launches the action to start making improvements.

The traditional approach to reliability improvement involves big investments in information systems, data gathering, planning systems, mechanic skills, scheduling systems, and reorganizations. While this approach has produced substantial results at some companies, many traditional efforts have sputtered out over time.

Origin of the game

Worldwide benchmarking at DuPont showed that the best reliability maintenance is paradoxically the least expensive. Just as importantly, these same benchmarks show that many of the technical elements of the traditional reliability approach are not present in many of the best-of-the-best performers.

DuPont created a model of plant reliability in the early 1990s. By applying that model to over 25 companies, it was discovered that the key to achieving better reliability at a lower cost is to engage the entire organization in eliminating the defects.

DuPont created a workshop now used with other companies to engage all levels and functions of the organization. The centerpiece of this workshop, the game, is an interactive simulation of a plant that allows participants to self discover the value of changing. It creates a common vision of the right ideas, generates grassroots passion to improve things, and launches improvement actions.

With this process, improvements in operations can begin within 90 days, and bottom line improvements are possible within the first year. The traditional approach typically requires 3-5 years.

The traditional approach

Once U.S. companies or sites decide to pursue increased reliability, they generally take the following traditional approach:
  1. Reduce reactive work by building a preventive (interval based) maintenance program, which requires a computer maintenance management system.
  2. Add a predictive (condition based) maintenance program, which requires acquiring appropriate predictive tools.
  3. Build a proactive (root-cause based) maintenance program, which requires root cause or reliability-centered maintenance training for front-line personnel.
The traditional approach outlined above requires a consistent and coordinated effort over 3-5 years. It also requires a large investment, and performance often worsens before it improves.

Finally, most successful implementations of this approach are "champion" driven, which means they are dependent on one person whose job function may change.

This functional approach has risen out of two valid underlying assumptions:

  1. Maintenance work done in a predictable, planned mode is more efficient and effective than work done reactively.1
  2. All of the systems, information, and people must be in place to make the work more possible to plan and predict.
The traditional approach, however, requires new work to get done (for example, planning, predictive work, machine-history analysis, inventory management, re-engineering). Most of this work is poorly understood so it is inherently slow as people work up the learning curve.

In the meantime, the plant is still in a reactive mode and all of the old work remains. While management might see the new work of planning and predicting as important, it will see the old work, which consists of failures and emergencies, as more urgent.

Also, there is a large monetary investment in skills training, computer systems, and diagnostic equipment that can be an easy target for future cost cutting.

Finally, there is the risk that projects, like installing a new computerized maintenance management system, can take on a life of their own. Companies may lose their focus on improving equipment reliability.

The game approach focuses on eliminating the work that the traditional approach tries to optimize.

Reliability dynamics

Dupont's model suggested and its experience proved that elimination of defects accelerated reliability improvement efforts.

Fig. 1 [78,341 bytes] is a dynamic illustration of this principal.

It begins by eliminating the defects in equipment. Examples of defect elimination include convincing mechanics to laser align all rotating equipment every time, helping operations understand the impact of cavitating a pump, and throwing away parts with known defects.

As defects are eliminated, over time there will be a reduction in the number of breakdowns and emergency jobs. This has two important effects: sources of defects are reduced, and time is created for mechanics to spend on future defect elimination, training, planning, and scheduling.

The reduction of defect sources is shown in the left-hand loop of Fig. 1. This cycle accelerates the improvement once the ball is rolling.

Defects from the failure events will be avoided because they were never introduced.

Defects from workmanship will decline as fewer jobs have to be done, especially rush jobs.

Defects from parts are reduced as fewer new parts are installed because fewer failures occur.

Time created for mechanics is shown in the right-hand loop of Fig. 1. By eliminating defects, some of the old work is eliminated to make room for the new work.

We have seen companies reduce defects dramatically over a very short period of time by working on defect elimination. In less than a year, the BP Lima refinery doubled its mean time between failure (MTBF) rate for pumps from 12 months to over 24 months and has since doubled its performance again to 48 months (Fig. 2 [88,834 bytes]).2

Creating the right culture

It is not enough to get the managers interested and motivated. The challenge is creating an organizational culture of defect elimination.

To create any real change in a large organization, the company must:

  • Communicate the value of the change both for the business and the individual.
  • Build the desire in the organization to get the change implemented and overcome the inevitable hurdles.
  • Focus the majority of people on the right ideas.
  • Be sensitive to the important parts of the organization and culture that should not be changed or be lost.
  • Launch the action required to turn strategies into reality.
The game has been used at DuPont for the past 6 years and outside of DuPont for the past 4 years.

The game is a simplified version of the System Dynamic model created at DuPont. Systems Dynamics is a discipline that came from MIT in the 1960s and was created by Jay Forrester.3

The game is played by a team of six on a 6-ft. multicolored game board. Usually, four to six teams play simultaneously.

The objective of the game is to move as much raw material (yellow poker chips) as possible through the plant (across the board) to maximize profit. Red poker chips represent defects. Players reduce defects by performing reactive-maintenance work, carefully doing planned work, or eliminating defect generators.

The game allows people to experience in a day what would normally take over 1 year to experience in the real world. It gives them the opportunity to take risks that they would not normally take.

There are three roles in the game: operations, maintenance, and business services, which includes stores and logistics.

Participants see the whole system and the impact of decisions made in one function on the others and the performance of the whole plant.

They are also required to play a different role from their normal roles so they see their world from another perspective and gain insight into the pressures faced by their counterparts. The plant in the game starts out with typical reactive performance and the team attempts to make the journey to proactive and profitable operations.

The game is facilitated in a way that participants can self-discover the value of the changes. Having participants invent the answer themselves is critical for creating the passion needed to make a change this big.

Alignment and enthusiasm around the right ideas, however, is still not enough. For real improvement, action must occur.

Action is the second key element of the Manufacturing Game workshops. The second day of the workshop is primarily focused on translating the ideas and excitement developed in the game to the real world through action teams.4

An action team comes together to work on a specific issue with a very clear goal for a short fixed period of time. In our case, the issue is defect elimination. The best teams are small (5-7 people) and cross-functional. They have operators, mechanics, engineers, and procurement people from a given area.

The team identifies defects in their equipment and processes and creates an action plan to eliminate one of them. They have a goal of eliminating that defect within 90 days.

A common objection to action teams is: "We just want to train the managers. Our operators and mechanics don't have time for this."

Only the front-line personnel know, however, where the majority of the defects are. Management can typically point to a few defects that are production bottlenecks, but it cannot identify the thousands of little things that eat up time, process efficiency, and quality.

Action teams must consist of more than managers. To make a big change quickly in reliability, one cannot have a few highly managed and facilitated teams. A large portion, 80-90%, of a site must be eliminating defects.

If every improvement and change has to come through one or a few people, the process will move very slowly, and there will be little enthusiasm from the front-line.

Culture change at Eastman Chemical

Eastman Chemical used the game workshops to accelerate improvement efforts at its Kingsport site. Details can be found in a paper presented at the NPRA Maintenance Conference in May 1998. 5

The teams from the power and services division went back to their areas and identified over 80 defects that could be eliminated. They then focused on the top 10 that they knew were important and possible to impact in a short period of time.

A good example of this approach is the work that they did with metering pumps.

The power and services division had several positive displacement metering pumps that had about two failures/week. Eastman spent around 20 hr/week fixing these pumps.

A team took this on as a defect to eliminate. It researched the problem and contacted the vendor. The vendor inspected the pumps and found that they had been running backwards and as a result, were not getting proper lubrication.

It was not possible to determine the direction that the pumps were running by looking at them. The pump manufacturer's technician also showed the team several tricks to setting up the pumps.

Since this action team completed its work, the pumps have not had a single failure. The entire effort took less than 90 days and required basically zero investment. Eastman will save over $26,000 in direct costs but more importantly, the action freed up one-half of a man year of maintenance time to do more value-added work.

While action teams are helpful in starting the process of eliminating defects, they also play an important role in building a culture where defects are not tolerated. Mickey Logan of Eastman Chemical's power and services division describes this change in culture as "...it starts out as a series of projects, but it has become the way we do business now."

This new approach was also used to solve a chronic problem that Eastman Chemical was having with its generators.

Soon after the first Manufacturing Game workshop, one of the critical turbogenerators came down because of vibration problems.

The mechanics determined that the bearing had been wiped and the surface was no longer smooth, which caused the vibration. They could not, however, determine the root cause of the bearing getting wiped.

The wiped bearing was replaced, and the generator restarted. On start-up, the vibration spiked up and then dropped off to acceptable but a higher-than-normal level. The generator ran over the weekend, but once the operating team saw the data on Monday morning, it did something that was very counter cultural. The team shut it back down.

Normally, a critical piece of equipment that is running in the acceptable range would not be taken down, but the team knew it still had a defect.

The bearing was slightly wiped again when they examined it and again no root cause could be determined. The team replaced the bearing and started it up again. When the vibration spiked on start-up again, the team immediately took the turbogenerator down.

Instead of giving up and letting the generator run with a known defect, the team re-examined the generator. It found what looked like a photographic image of the bearing on the shaft.

After consulting several experts, it determined that the shaft was not properly grounded and an electrolytic effect was causing the damage to the bearing. The brushes that were supposed to ground the shaft were not making good contact anymore. This was causing the bearing to form a pitted surface that caused it to get wiped very quickly.

The fix was easy. Eastman Chemical cleaned the contact brushes, replaced the springs that held the brushes against the shaft, and implemented a preventive maintenance program to make sure that these were maintained.

The cost of this solution was negligible. The team also determined that this same problem had been the cause of several previous generator failures over the past few years, and it came up with a new design to eliminate this problem in the future.

The savings from this change in approach is well in excess of $500,000/year and over 128 manhours/generator/year.

Improvement at BP

The results at BP's Lima refinery, have also been impressive. Details of BP's reliability improvements have been published. 2 6

Within the first year, it significantly improved the reliability of the plant, reduced maintenance costs, and dramatically reduced operating costs.

The operating cost improvements came as a surprise. By focusing on a broad definition of defect elimination and reliability, it cut a lot of waste in its operations. The best example of this approach was the butane-action team.

The butane-action team worked on a defect in the butane sphere that it felt was a safety hazard.

In the summer months, these spheres heated up and the pressure increased. Operators would then open a valve to the flare to release the pressure and avoid safety problems. For years, the operators and mechanics complained, but it was nobody's job to fix it.

Once the team started working on this defect, it quickly determined that the cooler on the compressor was undersized; it was very hot to the touch. The team went to the plant's "bone yard" where scrapped equipment was kept, found a larger cooler, had it refurbished, and passed it on to engineering. Engineering ensured that the fix was approved through the management of change process for safety.

After installation of the new cooler, the pressure on the vessel immediately decreased and the valve to the flare could be closed.

The team eliminated the safety issue and cut out over $1.5 million worth of butane going to the flare annually. As a result of action teams like the butane-action team, the Lima site reduced its hydrocarbon weight loss by more than 1%, which resulted in a $10 million reduction in annual operating costs (Fig. 3 [104,778 bytes]).

The refinery now attributes almost $32 million in annual margin improvement to its reliability improvement effort, or about 56¢/bbl.

The Lima refinery has put 80% of its site staff through the Manufacturing Game workshop, and it has completed over 30 action-team projects.

The culture of this plant has transformed from a mentality of "if it ain't broke, don't fix it" to one of "don't just fix it, improve it." Capacity is significantly higher, and maintenance and operating costs are down.


  1. Hudachek, R., and Dodd, V., "Progress and Payout of a Machinery Surveilance and Diagnostic Program," ASME Proceedings, 1985.
  2. Monus, Paul, "Proactive Manufacturing at BP Oil Lima Refinery," NPRA Maintenance Conference, May 1997.
  3. Forrester, Jay W., Industrial Dynamics, Productivity Press, 1961.
  4. Schaffer, Robert H., The Breakthrough Strategy, Ballinger Publishing Co., 1988.
  5. Smith, Steve, and Ledet, Winston J., "Using Organizational Learning to Accelerate Reliability Improvement," NPRA Maintenance Conference, May 19-22, 1998, San Antonio.
  6. Monus, Paul, and Griffith, Jim, "Proactive Manufacturing at BP Oil Lima Refinery." Society of Maintenance and Reliability Professionals Proceedings, October, 1996.

The Author

Winston J. Ledet is a principal in the Manufacturing Game organization and is one of the original contributors to the creation of the game. Over the past 5 years, he has worked with over 40 companies on reliability improvement efforts in a variety of industries including petrochemicals, paper, automotive, mining, and consumer goods manufacturing.

Ledet previously was an operations consultant with McKinsey & Co. He is an adjunct professor of business dynamics at Emory's Goizueta Business School. He holds a masters in management science from MIT where he focused on operations management.

Copyright 1999 Oil & Gas Journal. All Rights Reserved.