In this study, first a parameter design of the plasticizing unit in an injection molding machine was carried out, and then the relationship between the characteristic values obtained in the experiment and resin melt temperature data was investigated. If the characteristic values correlate to various quality data,variation can be reduced by parameter design first, and then the injection molding production process can be controlled by general injection molding measurements. For rational evaluation in the stages from process design to process control as described above, it is desirable to perform both a parameter design and an analysis and evaluation by the MT (Mahalanobis-Taguchi) system. A rational design method for the MT system, similar to parameter design, was discovered through this study, and an effective method of designing production processes for injection molded products was found.

There is currently a market demand for systems that can identify faces objectively. Research on facial identification using coordinates of features such as the eyes and nose has made personal identification possible. However, the feature points at which the coordinates are taken have been selected subjectively,and only a small part of the image information has been used. In this study,subjective selection of feature points was avoided to the maximum extent possible by using the facial image pixels themselves as feature points. Since the number of such feature points is very large, the multilevel Mahalanobis-Taguchi (MT) system was used to define a unit space. The results confirmed the feasibility of personal identification based on pixel information. It is anticipated that pixel-based image recognition techniques will be applicable not only to facial image recognition, but also to more general-purpose image recognition.

The injection molding, produces the resin products of prescribed shape by injecting the melted resin into the metal die. In this technology, the filling becomes not uniform in each part, and the difference in the density occurs when the melted resin is filled to the die. As a result, the shape of molded product differs from the shape of die because the difference in shrinkage for each part during cooling. The deformation of shape by this heat shrinkage is forecasted beforehand on the shop. The die is made in the shape which is different from a real product, and the product depends on the shape as hope is obtained finally. However, the experience and intuition of workman are needed in this adjusting process, that is, cutting down of die, and it costs an enormous cost for time and labor. As a result, the cost of injection molding product increases. The shape of hopper that supplies the pellet of resin has not been so considered up to now in the injection molding machine. In this study, the shape of hopper was optimized by the parameter design to aim to improve bridge phenomenon where the pellets are blocked in the hopper. It makes the amount of pellets supplying to the screw uniformly.

Exterior flat house paint must provide a uniform finish without variations in thickness or gloss, even when the paint is applied to a large exterior surface such as a condominium wall. Differing from the interior paint used in factories where temperature and other environmental factors are kept constant, house paint is used under widely varying conditions, and is applied by painters with widely differing skills. In this study, an experiment was carried out to find the construction environmental factors that cause uneven gloss, and a flat paint was developed that is resistant to the effects of these factors and produces an even finish. This product was developed by defining uneven gloss in terms of the amounts of reflected light entering the eye at various angles and carrying out a standard signal-to-noise ratio analysis. This led to the development of a flat paint in just a few months, as compared with at least a year for flat paints developed by conventional methods.

Papers on quality engineering were examined by tracing a history of the comments appended to and prizes awarded to papers that appeared in the Journal of Quality Engineering Society from the first issue through volume 16 No.6. This study produced a picture of the history of the development of quality engineering and also showed that our current task is to reexamine the idea of generic functions. That is, besides producing observations on the ways in which papers on quality engineering were evaluated and tracing the history of quality engineering research, this study led to a consideration of future directions for quality engineering research. Specifically, it became clear that generic functions must arise from consideration of how a system actually works. The possibility of finding new ways of comprehending technology through use of submitted papers will be explored as a future research topic.

An automated guided vehicle (AGV) cannot reach its desired position unless its position is detected and its movement is controlled. This study mounts two sound sources on a moving object modeled on the AGV, one on the left side and the other on the right side of the object. The positions of the sound sources are detected as they emit sounds at different frequencies. In addition, the numbers of revolutions of two wheels that are driven independently from the sound sources are controlled to guide the moving object to the desired position. This paper presents a basic test in which the hardware and software parameters of the proposed guidance system are optimized by the parameter design in quality engineering. Model-based testing as it is, it demonstrates that parameter optimization of the guidance system leads to swift and accurate guiding of the moving object to the desired position. This test result confirms the feasibility of the sound guidance system for AGVs.