- THE MAGAZINE
- WEB EXCLUSIVES
I remember one manufacturer shipping parts for an automotive Tier I supplier at 2 parts per million (PPM). The customer wanted 0 PPM. As a result, the customer wanted to sort all parts, recheck all parts, design a new inspection machine, etc. All in all, the cost of weeding out remaining 2 PPM became exorbitantly high and cut right through the corporate profits. The Appraisal Cost as a part of total cost of quality increased. Therein lies the irony-improving a process performance results in the higher appraisal cost. This is not manufacturing excellence, instead it is perpetuating manufacturing nonconformity.
Continuing manufacturing marginality is born out of on-going verification against a set of margins, or specification limits. Many manufacturing operations today are managed to the limits. As a result, we preserve performance within the specified limits. It just happens that such a system leads to limited attribute type pass or fail information. Failed items are verified and re-verified due to lack of intelligent information, and root cause analysis is difficult to pursue. Due to lack of variable-type detailed information about the process behavior, processes causing manufacturing problems are rarely fixed; instead we continue to treat the symptoms, so appraisal cost does not go down.
In order to reverse a culture of process control, the design must set clear targets, the manufacturing engineering must design the process to produce target performance and implement controls to keep the process around the target, and then use limited inspection or testing to monitor how wide and far the process distribution is from the target. Such an approach will continue to shrink the variance, and continually bring the process output closer to the target. This will continually necessitate less inspection, thus reducing appraisal cost.
I remember my young days while responsible for testing of computer memory devices, my goal was to reduce test time from 300 milliseconds. In the current domain of process improvement, there is practically no goal focusing on reducing cost of inspection, testing and verification as an expected outcome. Thus, we must increase sensitivity to the inspection, testing and verification activity that is known to be a non-value-added activity. Excessive inspection, testing or verification gives false confidence in the capability of marginal manufacturing operations.
Building to target and verifying the product performance against the target leads to real confidence in the manufacturing capability, and eventually results in reduced inspection, testing and verification. Any reduction in the appraisal cost increases corporate profitability and improved product quality.
We can see that the correct way to achieve manufacturing excellence pays for itself while attempting to achieve excellence through inspection costs. The correct way to achieve excellence is through good product and process designs with clearly defined targets, and with the goal of continually aiming to keep manufacturing capability on target while reducing variability around it. Without defining effective targets, the only choice is to add more appraisals. By defining targets, the choice is reduced appraisal and reduced appraisal cost.
Reducing appraisal means fewer gages, less appraisal time, faster appraisals and a robust measurement system. Let’s look into our manufacturing systems, assess the cost of appraisal as a percent of total cost of manufacturing and in absolute dollars. Set a goal to aggressively reduce it-it is not adding much value and going nowhere. Worse yet, it hinders continual improvement and the drive for excellence.