The quality detective: a case study

Abstract

The idea of modern quality improvement is to improve processes and remove causes for problems rather than passively inspect. People engaged in quality improvement projects work as detectives. Scientific method plays a major role, and statistical methods and design of experiments are the tools for fact finding. Data and graphs are sparks that light up the imagination of the investigator. The ideas coming from the data are catalysts that bring out the best of engineering. An industrial case study is presented that illustrates the quality detective idea. I show how engineering and statistics played dual roles in an iterative learning process. Design of experiments, statistical methods of analysis and engineering were used to remove causes for excessive variation. Practical constraints, time pressure, psychology and relations with management and other team members had an important impact on the experimental strategy. This example illustrates the tremendous economic gains industry can make by using designed experiments.