Everyone has an idea about what constitutes process quality. It's often defined as the absence of negatives. However, if the machine crashes, the cause may be related to man (operator), machine, material, method or some combination. So who or what is at fault (i.e., lacking in some key quality)? Leaving the operator out for the moment, the problem is almost always machine and material.

This system approach does not fit well into management's black-or-white thinking. Yet, killer solutions to a single problem can be envisioned in either the machine or material arenas. As an exaggerated example, web breaks could be cured by entraining carbon fiber into the web or by carrying the web through the machine with a forming belt. The problem is not that many solutions are not possible; the problem is that constraints are applied at the outset which often precludes a solution before we even get started.

The “fix-it-but-don't-change-anything (of substance)” constraints are what cause more people to get stuck than any laws of physics. Next time management constrains the solution at the outset, remind them that they are mixing the option-generating and decision-making steps. This is not a minor problem-solving faux pas. It's similar to criticizing ideas in a brainstorming session.

Defining product quality is also problematic. Certainly customer complaints would seem to indicate key qualities are lacking. However, is this the whole story? Could the customer be complaining where the fault is not largely or solely with our web? It reminds me of a recent mission where the supplier sent me to their customer to investigate curl complaints. One objectionable outcome was that the curl caused a photo-eye to misbehave. Yes, indeed the web was quite curled, and we could look into curing curl on the supplier's manufacturing process. However, the far easier solution was to simply move the photo-eye closer to a roller where it would be flat instead of midspan where the curl can act without constraint. Change an entire (supplier) process and product or move an eye. Solutions are available in both locations if one could look beyond the biases and self-interest of the company or department one works in. Once all solutions are listed without constraint, by which I mean interference by management, it will usually be apparent what makes the most sense.

Very often, quality is defined in terms of conformance to targets as measured by the test lab. Here, there is the highest level of disconnect between numbers and reality. Who says that a particular measurable property is important? Who says that there aren't other currently unmeasured properties that might be even more important? Who says that the chosen target is best for the product, process or system? Who says that exceeding a certain variation including process capability will be problematic?

We've presented more questions than answers. This is good. It stimulates thinking. The single-minded person seeks the answers they're looking for. The smart person seeks the answers however they fall. The wise person seeks the questions.

920/312-8466, drroisum@aol.com, www.roisum.com