ANALYSIS OF BUSINESS LOSS AND SYSTEM RISK CAUSED BY NONSTANDARD AND EXCESSIVE QUALITY

The loss function by Professor G.Taguchi is defined by the quality as "losses to be given to society after shipment, but excluding losses due to the function itself". Furthermore, he proposed an evaluation with a loss function that approximates that the loss is proportional to the square of the deviation from the target, and it was economically reasonable that the quality control using the standard deviation and the least squares method are appropriate. The probability distribution of the quality of main manufacturing parts (hybrid integrated chips) used in products is assumed to be a normal distribution. The distribution function gives the probability density function of the quality measurements for the individual products, and we here introduce a major loss function in quality engineering. Generally, the shipping-side standard and the receiving-side standard differ; therefore, the underquality and overquality are analyzed. The loss cost significantly fluctuates because of product quality problems, process lead time, and so forth, thereby affecting the profit risk. These system risks can be mathematically analyzed. We report calculation results for process risk probability based on actual data