Too much diligence? Who'd have thought....?
Recently, I have been involved in a very interesting debate with my co-workers. Every week, the core manufacturing group meets to discuss what we shall call (for simplicity) customer complaints. We then agree (more often than not) on a root cause and corrective action for each case. Now, in certain types of part failures, root cause analysis is not so simple, and an engineer outside the group is then tasked to do some detailed research on that part. It could take months for the engineer to arrive at a conclusion for the group to use in determining the appropriate corrective action. Typically, there is no other activity on that particular case during that time, so the case would sit idle for a while. However, our new quality engineer, who was tasked with facilitating this meeting every week, felt that a culture change was in order. He felt that instead of simply sitting on a case while the analysis was being conducted, the group should continue to be proactive in identifying possible weaknesses in our manufacturing processes that could have led to that failure. Half the group agreed that the time we had until the results of the engineering analysis could be put to good use this way. The other half, however, disagreed. They felt that without knowing the exact cause of failure, we would be firing shots in the dark and wasting our efforts.
I think both sides have valid points. I strongly believe that as a supplier, we should take quality assurance seriously and be proactive in correcting any weaknesses in our processes. We should also develop the habit of asking ourselves the right questions about our processes and how they impact the final quality of the product. Most importantly, time is too precious to waste when there is a disgruntled customer who wants prompt action on our part.
On the other hand, time could be wasted either way if we force ourselves to make uninformed conclusions. Without solid information, we would basically be looking at all possible reasons for part failure and attempting to put checks in place to circumvent these reasons in the future, whether or not they were actually responsible for the failure. If we are not careful, we could end up torquing every screw, testing every sensor and every switch, and over-designing every mechanical part. This would not be a prudent thing to do at a company that makes equipment as large and complex as ours.
I do not sit on the fence on this one. I think waiting for the formal analysis is the right thing to do, even if it costs us some months. My colleagues know what a feisty proponent I usually am for proactive measures. So they would be surprised that I am against this measure that seems to be “proactive”. There is an illusion of proactiveness here. True, the investigation is indeed proactive. However, the resolution that comes out of it will still be reactive. Let’s take an interlock, for example. We know that the probability that it will fail is high. If we were truly proactive, we would act on that weakness even before we see the interlock fail at the customer site. After it does fail, the probability does not change. So if we do change our process to address that weakness just because we have an open case against that failure, it would still be a reactive measure. If we are taking reactive measures anyway, let take reactive measures that are justified by hard facts.
