During a product launch or product quality planning process, it’s often quite common to look at process capability numbers and roll those into relevant corresponding documents.

The intent in this process is genuine; however it’s frequent that a failure to comply with product specifications after the launch process will occur regardless, especially with new designs or processes.

Manufacturers with low process capability numbers in given areas are typically very attuned to those numbers and inherently develop compensating improvements to ensure nonconforming product does not get out the door.

The greater deception actually occurs around the areas where the process shows high levels of capability.  Frequently some failure mode occurs which is adjacent to these highly capable processes.

For example let’s say a grinding process has very high CPK values.  This is a great, however the unanticipated concerns which are often not defined on process documents are typically the failure modes seen by the customer.  It’s usually the items that are difficult to anticipate which often rear their head as the source of a concern. And because they are difficult to anticipate they never end up on the product quality planning documents, nor design FMEA’s, or any other proactive format utilized to eliminate these concerns.  This becomes a vicious circle in organizations that dump excessive resources into the planning process with the hopes of creating the perfect launch.  To make matters worse these companies tend to ignore the potential of downstream problems because they proactively and diligently applied resources at the front of the development cycle.

So next you might ask what is the solution?

Shift the less productive time and resources spent at the front of the process to a higher frequency of inspections and auditing.

Let’s be clear.  I am not endorsing the elimination of a proactive planning and development process.  However as these planning and statistical concepts have matured many have assumed inspections and audits could be eliminated and there lies a problem.

Let me give you an example.  During plant visits to automotive assembly plants I have noticed distinct differences in the levels of inspection and auditing directed where no apparent quality concerns existed. Other companies plants however would take a high tech approach based upon capabilities and statistics to insure their product quality.

The clear winner in terms of product quality was the plant with the added layer of low tech inspections and audits.  Ironically this plant had all the same advanced analysis and statistical technologies but they did not depend on these techniques to insure product quality.  When a failure eventually appeared this company caught the problem in their plant before it got to their customer.