Considering the importance of joining information and making it reusable as organizational knowledge in the concurrent engineering, we study the configuration of knowledge database necessary to design. The fact that design is an activity which analyzes "the inverse problem", i.e., determines various design characteristics necessary to realize the functional requirements derives its real substance of difficulties. In order to improve the quality and efficiency of design processes, they are very important to make an easy access to "knowledge" necessary to realize the functional requirements and to consolidate the environments of supporting thinking processes. To realize them, we present the methodologies of fitting up environments of which fundamentals are qualitative and quantitative representations of relations between functional requirements and design characteristics using FTA and diagrams representing their numerical relationships. The intermediate representation of design characteristics is also important to make the logical relations between them clear. Collecting and accumulating the knowledge and data of design based upon the above principles make an organizational accumulation of design technology information and also make it join and reusable.

View full abstract

The coefficient of determination R^2 and the coefficient of determination adjusted for the degrees of freedom R*^2 are common statistics to evaluate the goodness of fit in a linear regression model. The coefficient of determination doubly adjusted for the degrees of freedom R**^2 is also used in applied research in the field of the quality control. It should be noted that R*^2 and R**^2 are proposed to select an adequate model and that the statistic which has good performance for the model selection cannot always perform well for the point estimation of the goodness of fit. In this paper, we consider the small sample properties of R^2, R*^2, R**^2, max{0, R*^2} and max{0, R**^2} as point estimators of the population coefficient of determination. We have the following results: (1) all types of estimators are unreliable when the sample size is small: (2) if we have to use the estimator, max{0, R*^2} is most desirable: (3)R^2 has the severe upward bias: (4) R**^2 and max{0, R**^2} have the severe downward biases, and their MSE's are larger than others, which imply that R**^2 and max {0, R**^2} are not suitable estimators for the population coefficient of determination.

View full abstract

The method of selecting a quality improvement strategy by using the ratio of improvement rate of a product function to its required cost has been generally used by many manufacturers. However, it is necessary to consider that each customer has preferences and expectations attached to the product and any changes to the product that may occur. Therefore, the quality improvement strategies to be selected by the manufacturers do not always agree well with customer satisfaction levels. It is the purpose of this paper to suggest a method of strategy selection for quality improvement by using the ratio of the degree of contribution towards customer satisfaction to the cost required for the improvement.

View full abstract

This paper presents a method using the wavelet transform that can detect and recognize unnatural pattern on sequential process data. The wavelet transform can characterize the local regularity of signals by decomposing them in time-frequency domain. First, we introduce the basis of the wavelets and discuss the use of effective mother-wavelets for identifying these unnatural process patterns, such as Shift, Peak, Trend and Cycle. Then, we show that the local feature of the process is extracted at various levels of the wavelet decomposition, and the unnatural pattern is identified through the threshold value of wavelet transform values in each level made by a Type I error and by reducing empirical error rates of the recognition. In this procedure, the wavelet decomposition in conducted sequentially, which enables us to monitor the wavelet transform values of all decomposing levels in real-time. For the evaluation of the proposed method we conduct a simulation study in which it is compared with cusum-charts.

View full abstract

Recently many factories are introducing automatic equipment in order to improve productivity and achieve low cost products. Minor-stoppages are one of the problems that hinder automatic productions. Mechanisms of minor-stoppages are not investigated generally, even though there are many case studies with investigations on each of the minor-stoppages. This paper tries to generalize mechanisms of minor-stoppages that are obtained through collections and analyses of many actual cases. Analyses are based on the point of view that there might be common mechanisms for many minor-stoppages occurred in varieties of equipment. As a result, the following conclusions are obtained. 1) Phenomena of minor-stoppages can be generalized 2) based on 6 basic working-functions of the equipment. Also, mechanisms of minor-stoppages can be generalized by dividing phenomena into three steps, that is, deviations from working conditions, failure in working function, and its influence. Causes of minor-stoppages are generalized based on 6 elements of the automatic equipment. These results help analysts to observe phenomena, list up causes, and come up with solution easily.

View full abstract

Design of experiments for screening tests focusing on setting the test condition is discussed. It is assumed that population consists of both good and defective items, also good items don't fail during the test period. Using Weibull regression models for the lifetime distributions, maximum likelihood approach is applied. The optimum treatment for screening defective items and the optimum combination of observation points to estimate population parameters with highest precision are discussed. Applying this model to screening defective LSI, the effectiveness of the model, the result of the analysis and its procedure are presented.

View full abstract