Monthly Archives: January, 2013

Entropy and Assignable Causes

Left alone, a process in the Ideal State will migrate toward the State of Chaos over time. Wear and tear of parts in use will lead to breakdowns and failures. Even mothballed machines cannot escape deterioration and decay caused by the environment. Entropy affects everything.

However, the effects of entropy can be repaired. Signatures of deterioration and failure made apparent by process control charts provide clues as to what repairs need to be made when. But, a process isn’t affected just by entropy. Other assignable causes keep it from operating in a predictable fashion. Replacing worn out parts might pull a process out from the State of Chaos back to the Brink of Chaos, but it’s just a matter of time before it returns to it.

Process managers, through the proper use of process control charts, must counteract the effects of entropy and assignable causes to help their processes achieve the ideal state and stay there. This is a never ending cycle.

Note: I learned this material from reading Dr. Wheeler’s writings. My post is intended to reflect my current understanding. None of the ideas herein are original to me. Any errors are my failures alone.

References

Advertisements

The State of Chaos

When a process is out of control and it is producing nonconforming product it is in a state of chaos. The State of Chaos is one of the four states a process can be in as shown in “What State Is Your Process In?“. The manufacturer cannot predict how much nonconforming product his process will produce in any given hour or day. At times the process will produce nothing but conforming product. Then without warning it will produce nothing but nonconforming product. It might seem as if there were ghosts in the machine.

A process in such a state is affected by assignable causes that are easily identified through the use of process control charts. The effects of these assignable causes have to be eliminated one at a time. Patience and perseverance are necessary. It is essential that the process be brought under statistical control and made predictable. Once the process has achieved stability further improvement efforts can be made to reach the ideal state.

ChaosProcess

Note: I learned this material from reading Dr. Wheeler’s writings. My post is intended to reflect my current understanding. None of the ideas herein are original to me. Any errors are my failures alone.

References

The Brink of Chaos

Of the four states a process can be in (see “What State Is Your Process In?“) the most sinister state is the one where it is producing 100 percent conforming product but is operating in an unpredictable way. That is, the process is not under statistical control. Such a process is, in fact, on the brink of chaos. But, hold on. There is no nonconforming product, therefore there is no problem, right? It is easy to get lulled into complacency by this happy circumstance.

BrinkProcess

But because the process is not under statistical control it is impossible to predict what it will do in the next instance. Various assignable causes are affecting the process in an unpredictable fashion. The effect of these causes could very well be the production of nonconforming product without any warning. When that happens the process has moved into a state of chaos.

The only way to address a process on the brink of chaos is to use process control charts to identify assignable causes and eliminate their effects one-by-one and bring the process under statistical control. You can then start other improvement efforts like moving the process mean to the process aim and reducing the process variation by minimizing the influence of common causes affecting the process.

Note: I learned this material from reading Dr. Wheeler’s writings. My post is intended to reflect my current understanding. None of the ideas herein are original to me. Any errors are my failures alone.

References

The Threshold State

A process that is predictable or in a state of statistical control, but producing nonconforming product can be described as being in the Threshold State. This is one of the four possible states that a process can be in as noted in “What State Is Your Process In?“. But, what might such a process look like?

A process in the threshold state might be operating with its mean higher than the process aim,

ProcessMeanHigh

or it might be operating with its mean lower than the process aim,

ProcessMeanLow

or it might be operating with a process dispersion greater than the product specification window,

ProcessDispHigh

or it may be operating with some combination of a shift in its mean and breadth of its dispersion.

Nevertheless, the fact that such a process is in statistical control means that it will continue to produce consistent product so long as it stays in control. This in turn means that the producer can expect to produce a consistent amount of nonconforming product hour after hour day after day until a change is made in the process or a change is made to the specifications.

It is important to say here that exhorting your workers to work harder or to “Do it right the first time” or to show them the examples of nonconforming product from a process in the threshold state will not lead to improvements. They are not the cause for the failures. The causes for the nonconforming product are systemic and must be dealt with at the system level. Focusing on the worker will only serve to demoralize and frustrate them. It may lead to tampering with the process turning a bad situation worse.

You can always share your process data with your customer to demonstrate its stability and ask for a change in the product specifications. However, if specifications cannot be changed your only recourse is to modify your process to shift it from the threshold state into the ideal state. Adjusting the process mean to match the aim is usually relatively simple. In comparison, reducing the process variation requires an understanding of the common causes affecting the process and their respective effects – a much more involved activity.

While you are working on improving your process you are still producing nonconforming product. Until such time as you achieve the ideal state for your process, you must screen every unit or lot before shipping product to your customer – a 100 percent inspection and sort. This should be treated as a temporary stop-gap measure. You must recognize it as an imperfect quality control method and be mindful that defectives will escape.

Note: I learned this material from reading Dr. Wheeler’s writings. My post is intended to reflect my current understanding. None of the ideas herein are original to me. Any errors are my failures alone.

References

The Ideal State

In “What State Is Your Process In?” I noted that a process can be in one of four possible states. Here I write about the Ideal State wherein a process is predictable and is producing 100 percent conforming product.

A process that is predictable is one that is in a state of statistical control. The variability of the product from one unit to the next is randomly distributed about the average and bounded within statistically established limits – its natural limits (red solid lines in the figure below). So long as the process remains “in control”, it will continue to produce units within these limits.

Image

Complete product conformity comes about when the process’s natural limits fall within the product’s specification limits (blue solid lines in the figure below). This depicts the ideal state for a process.

Image

In order for a process to achieve this ideal state

  • The process must be inherently stable over time. This means that in the absence of external disturbances – what Dr. Shewhart referred to as assignable causes – the process’s natural variability does not change over time. (Note: There are processes that are inherently chaotic. An excellent reference to such processes is “Nonlinear Dynamics And Chaos” by Professor Steven Strogatz)
  • The process must be operated in a stable and consistent manner. The operating conditions cannot be selected or changed arbitrarily. Often machine parameters are tweaked in response to natural fluctuations in the process’s output. These actions add to the process’s natural variation disrupting its stability. Dr. Deming demonstrated the effects of such tampering through the “Nelson Funnel Experiment”. (Bill Scherkenbach has an excellent discussion of it in “Deming’s Road to Continual Improvement”.)
  • The process average must be set and maintained at an appropriate level. If you refer to the charts above, you can imagine the consequence of moving the process average up or down from its aim. The result is the production of nonconforming product either on the high or low side.
  • The natural tolerance of the process must be less than the specified tolerance for the product. This is obvious upon a first glance at the second chart above.

A process that satisfies these four conditions will be in the ideal state and the manufacturer can be confident that he is shipping only conforming product. In order to maintain the process in the ideal state he must use process control charts. He must act on the signals they provide to promptly identify assignable causes and eliminate their effects.

Note: I learned this material from reading Dr. Wheeler’s writings. My post is intended to reflect my current understanding. None of the ideas herein are original to me. Any errors are my failures alone.

References

What State Is Your Process In?

You can take one of two approaches to controlling the quality of your product. Once manufactured, you can compare it against its specifications and sort it as either conforming or nonconforming. This, however, will guarantee the greatest degree of variability between units as anything within specifications is considered acceptable. Alternatively, you can work to run the manufacturing process as consistently as possible to produce units that vary as little as possible from one another.

There is no path to improving performance with the first approach where you sort manufactured product as conforming or nonconforming. However, there is a clear path to improving performance through the use of process control techniques introduced by Dr. Walter Shewhart. But, how do you gauge improvement? One measure is achieving a state of statistical control for the process so that its behavior is predictable. Another measure is the manufacture of 100 percent conforming product by the process.

When these two measures are taken together, there are four possible states a process can be in (see figure below): The process is

ProcessState

Note: I learned this material from reading Dr. Wheeler’s writings. My post is intended to reflect my current understanding. None of the ideas herein are original to me. Any errors are my failures alone.

References

Great Generals

As to the influence and genius of great generals — there is a story that Enrico Fermi once asked Gen. Leslie Groves how many generals might be called “great.” Groves said about three out of every 100. Fermi asked how a general qualified for the adjective, and Groves replied that any general who had won five major battles in a row might safely be called great. This was in the middle of World Wat II. Well, then, said Fermi, considering that the opposing forces in most theaters of operation are roughly equal, the odds are one of two that a general will win a battle, one of four that he will win two battles in a row, one of eight for three, one of sixteen for four, one of thirty-two for five. “So you are right, general, about three out of every 100. Mathematical probability, not genius.”

— John Keegan, The Face of Battle (Viking, New York, 1977)