Every product is designed to perform some intended function. Some measurable characteristic, generally referred to as the quality characteristic, is used to express how well a product performs the function. No matter how the quality of the product is measured — by a single criterion or by a combination of multiple criteria, the measure will possess one of the following three characteristics that indicates the direction of desirability of results: bigger is better, smaller is better, or nominal is best. Early in his research, Dr. Taguchi observed that unexpected variation was common to all manufacturing processes and that it was the primary cause for rejection of parts. Parts were rejected upon inspection when they did not conform to a predefined specification. Rejection increases the cost of production. Thus, he recommended that effort should be directed toward minimizing variation, with less emphasis placed on production within fixed tolerance limits. The Taguchi method is most effective when applied to experiments with multiple factors. But the concept of selecting the proper levels of design factors, and reducing the variation of performance around the optimum/target value.