Instead of personal identification by looking his picture or by hearing the person's voice, we want to do it by computer. We have to find the same type of variable for each person in MT system. Each person must have enough database for his identification. How to read handwritten letter is also an identification problem. As there are many person and many letters, it is needed to find a much simple method than calculating a pattern distance. The method stated in this chapter uses standard distance of unit space member to check whether this belongs to this unit by that standard deviation. It looks like the control chart introduced by W.Shewhart. However the method stated here is quite different from the control chart method. Because there are many members.

Real-estate appraisal so far has expressly considered the relatiqnship between real-estate price and greenery on the real-estate premises or in the surrounding region. The quality of the real estate setting, however, is a recognized pricing factor in the real-estate appraisal standard, and greenery undoubtedly has an effect on the economic valuation of real estate, because it constitutes part or all of the setting. In an effort to understand the influence of greenery on the economic value of real estate, regions in the suburbs of southwestern Tokyo were selected for appraisal and the Mahalanobis-Taguchi system was applied to the forecasting of their real estate prices. The analysis in this study was based on signal patterning, using a single-data unit space instead of applying the Schmidt orthogonal expansion to the unit space as is done when the TS (Taguchi-Schmidt) method is employed. Despite the use of a single-data unit space, highly accurate results were obtained. Because it has fewer unit space constraints, this type of analysis can be carried out on a larger number of signals, and it reduces the economic problems associated with collection of unit space data. Elimination of the major data constraints also clarified the amount of effect on price, which is an essential factor in practical appraisal. The many advantages of this method place it on a par with the TS method.

At present, full sampling inspections of quality characteristics are used in the automatic assembly of switches, but even this does not eliminate all defective articles. Use of the Mahalanobis-Taguchi (MT) system to detect defective articles more reliably was studied and introduced into the production process. The measured characteristic was the operating-force-returning-force curve, which reveals many defective trends. Independently developed support software was used to convert raw data to optimum characteristic quantities (by grouping the data) and set factors. We also used a proprietary approach to signal typing and calculation of Mahalanobis distance from signals in studying conditions and evaluating appropriate scaling. The system we developed is able to respond to a change in switch model by deriving new optimum conditions immediately. After this system was introduced, there was a corresponding reduction in complaints; this was considered to be a result of use of the MT methodology.

In injection molding, a hardened produt is obtained by injecting a heated melt of plastic resin into a metal mold and cooling it in the mold. Differences in shrinkage ratio can occur when the resin hardens in the mold, yielding products that depart greatly from their target dimensions. Whereas past evaluations were carried out by comparing actual products with their blueprint dimensions,in this study we evaluated transferability of the mold dimensions to a test piece. As a result, we were able to reduce point-to-point variations in shrinkage. To minimize measurement error, we also studied the optimization of a measuring method using a non-contact three-dimensional image measuring instrument.

For ISO 14001 compliance, there is a growing need to shift from machining with water-based coolants to dry or semi-dry machining. In the drilling of oil holes in engines, shifting to semi-dry machining is also necessary from the standpoints of energy conservation and zero emissions. Engine oil holes have to be drilled at a low cutting efficiency to avoid drill break due to the great depth and small diameter of the holes, so the cycle time is extremely long. The present study, which applied quality engineering to find optimum conditions for high-speed semi-dry drilling of deep holes, addressed an urgent need to improve the efficiency of a drilling process that had been converted to semi-dry machining. Based on the energy evaluation concept, this study measured electric power consumption and the weight of the material removed. Since the experiment itself was carried out by drilling deep holes with high efficiency, it was anticipated that many drills would break, and in fact they did, so the experimental data were analyzed by the pass/fail method as well as by evaluating weight of material removed and electric power consumption, and the results were compared. The optimum conditions obtained from both types of analysis were practically the same. Semi-dry drilling under the optimum conditions produced various economic benefits, including an energy saving effect and a zero-emission effect.

In semiconductor fabrication processes, it is desirable to utilize on-line quality engineering to establish stabilized fabrication processes through optimized process control. In this study, for the purpose of establishing on-line engineering in semiconductor fabrication processes, a system that many people can utilize was studied and verified through practical use. In work analysis and loss calculations, an attempt was made to eliminate hesitation and improve working efficiency by standardizing data and creating tools that made the data more usable and more generally applicable. Through practice in actual processes and establishment of these methods, an increase in the use of on-line quality engineering was observed, and it was also shown that on-line quality engineering could uncover various problems present in process control.