Focusing Improvement Efforts for Total Quality Management

We can also use this production system to examine the effects of a Total Quality Management oriented program to reduce defects produced in the system. The defect rates for each resource are shown in Figure 5. This is again the same production system shown in Figure 1. The resources shown in Figure 5 with a square in the middle have a defect rate of 5 parts per hundred, while those with a circle in the middle have a defect rate of 4 per hundred parts produced. The resources shown as cross-hatched have a defect rate of 3 per hundred, and those which are unmarked have a defect rate of 2 per hundred. The resources with a black dot in the middle have a defect rate of one per hundred parts. If we could reduce the defect rate to half its current level on any one resource, which one should we target first?

Without taking the effects of constraints into account, a conventional answer would be to select one of the resources marked with a square, to reduce its defect rate from 5 parts to 2.5 parts per hundred. However, this does not consider the difference between a constraint and a non-constraint resource, as shown in Figure 2. While all of the non-constraint resources can make more than enough parts to make up for their defect rates, the constraint cannot. If the constraint resource makes a bad part, it is a part that is lost to the system's output and cannot be made up. If the people operating that resource had taken a rest instead of making the defective part, it would have the same result for the system's output as if the defective part had never been made in the first place. This brings up an issue regarding defective parts made at resources in the system other than the constraint.

If a one-hundred-percent screening process could be put in place to reject defective parts that have already been produced, where is the most important place in the system to put it? If this screening process was placed just in front of the constraint resource, then the constraint would not work on any parts that were already defective. Having the constraint work on a part that is already defective is just as wasteful as not making it at all. It would be just as productive to give the workers a break, instead of making a part that cannot be used.
There is no question that defect rates throughout the system should be targeted for reduction. However, rather than try to improve everything at the same time, with a "tidal wave" of improvement washing through the system, it makes sense to target those defect rates first which have the greatest impact on the system. Once the defect reduc­tion is accomplished at the constraint, the processes which follow it in the production sequence should be targeted next. This is because once the parts have been worked on by the constraint, they become in a sense, "gold plated." It would be wasteful to have them become defective after that, since it would reduce the system's output just as much as if the constraint had not made the parts at all.

Also, effective screening processes should be placed in front of the assembly resource K which follows the constraint (marked with a black dot). This would prevent any already defective parts from being assembled with parts that the constraint has worked on, resulting in an overall defect. There is no question that all of the defect rates at all of the resources in the system should be reduced. It should be apparent that focusing the efforts can lead to greater impact on the overall system by multiplying the effective­ness of a Total Quality Management oriented program to reduce the defects produced. The outcome of this focused program would be that tangible results would be produced much more quickly, with much less disruption to the system's operation, and at much less expense.

Conclusion

Constraint Management can be used to multiply the effec­tiveness of Total Quality Management and Just-In-Time improvement programs, without contradicting their phi­losophy or their methods. Through a focus on Throughput, Constraints, and Protective Capacity, Constraint Manage­ment can provide a means to achieve very rapid improve­ment, with much less investment and much less disruption than TQM or JIT programs without this focus. In addition to targeting these improvement programs, Constraint Management calls for a recognition of the importance of the performance measurement and reward systems in either stifling or accelerating the improvements. This stems from a realization that the behavior of individuals is heavily influenced by the ways they are measured and rewarded.

While the illustration of manufacturing improvement pro­vides a useful context to discuss Constraint Management, this should NOT be interpreted to mean that Constraint Management is only useful in manufacturing. Constraint Management has been used very effectively to improve the management of manufacturers, service organizations, edu­cational institutions, and government agencies, among others. A primary value of the Constraint Management approach is that it provides a focus for improvement efforts, including other programs, rather than replacing or contradicting them.