TABLE OF CONTENTS In Chapter 4, it is claimed that perhaps the majority of quality problems are caused by variation in quality characteristics. The evidence is that typically only a small fraction of units fail to conform to specifications. If characteristic values were consistent, then either 100% of units would conform or 0%. Robust design methods seek to reduce the effects of input variation on a system's outputs to improve quality. Therefore, they are relevant when one is interested in designing a system that gives consistent outputs despite the variation of uncontrollable factors.
Taguchi (1993) created several "Taguchi Methods" (TM) and concepts that strongly influenced design of experiments (DOE) method development related to robust design. He defined " noise factors" as system inputs, z, that are not controllable by decision-makers during normal engineered system operation but which are controllable during experimentation in the prototype system. For example, variation in materials can be controlled during testing by buying expensive materials that are not usually available for production. Let m n be the number of noise factors so that z is an m n dimensional vector. Taguchi further defined " control factors" as system inputs, x c, that are controllable both during system operation and during experimentation. For example, the voltage setting on a welding robot is fully controllable. Let m c be the number of control factors so x c is an m c dimensional vector.
Consider that the r th
