品質工学
Online ISSN : 2189-9320
Print ISSN : 2189-633X
ISSN-L : 2189-633X
12 巻, 4 号
選択された号の論文の8件中1~8を表示しています
論説
  • ―品質工学の初期1984年まで, 品質から機能へ―
    田口 玄一
    2004 年 12 巻 4 号 p. 5-13
    発行日: 2004/08/01
    公開日: 2016/12/24
    ジャーナル フリー

    The role of engineers is to create products/processes to perform a certain function;new products/processes never exist in nature. They can choose any system and set parameter levels of the system as they wish. These are called control factors. Such activities require cost and functionality predictions. When limited experiments are conducted in R&D laboratories it is impossible to predict exact output responses because laboratory conditions are quite different from downstream conditions for example, from large scale manufacturing condition or customer requirements. There are two strategies related to design of experiment for dealing with the reproducibility of design optimization: (1) Find some good functionality measure so that the design has enough robustness at customers hands. Signal to noise was introduced to measure the robustness. (2) Use OA, an orthogonal array, to inspect the downstream optimality by showing no significant interactions among control factor on the SNR.

    This chapter tells a short history up to 1984 when the following two books were published by Japanese Standards Association. 1: Measurement study group. Parameter Design for New Product Development. 2: Taguchi, G.(ed.). Reliability Design for New Product Development.

開発と研究
  • -コピー画像の評価方法-
    柏谷 修平, 桑田 敦史, 石毛 和典, 竹田 誠, 田村 希志臣, 齊藤 之男, 矢野 宏
    2004 年 12 巻 4 号 p. 31-36
    発行日: 2004/08/01
    公開日: 2016/09/30
    ジャーナル フリー

    Copy image quality is generally evaluated at the time of shipment according to the copier manufacturer's design criteria. In practice, the designers of copiers do not fully know what types of images users want, so there is a tendency to set the shipping evaluation criteria by experience and intuition. This study aimed at enabling optimum design by using evaluation sample images and relating users, subjective evaluations with physical image evaluations by the MT system. In this first report, methods of evaluating physical characteristics of copy images were studied, and the S/N ratios of the evaluated characteristics were determined. Based on the resultant data, a unit space was created from images that satisfied users, and it was verified that lower rated images had greater Mahalanobis distances from the unit space.

  • -単位空間の最適化による判定能力の向上-
    桑田 敦史, 柏谷 修平, 石毛 和典, 竹田 誠, 田村 希志臣, 齊藤 之男, 矢野 宏
    2004 年 12 巻 4 号 p. 37-43
    発行日: 2004/08/01
    公開日: 2016/09/30
    ジャーナル フリー

    This is a second report on the study in which users' subjective evaluations were related with physical image evaluations of copy image quality by the MT system. In this report, the unit space that was created in the first report was reexamined. The selection of measurement items in the first examination improved the S/N ratios but lowered the sensitivity, resulting in a unit space with inferior decision capability. This was due to variations in signal levels, so level precision was improved and the measurement items were reselected, whereby a unit space with superior decision capability was achieved.

  • 上田 収
    2004 年 12 巻 4 号 p. 44-49
    発行日: 2004/08/01
    公開日: 2016/09/30
    ジャーナル フリー

    For optimal design of a system consisting of multiple subsystems, which requires optimization and integration of the subsystems, consideration of the effects of interfaces between the subsystems raises evaluation problems,because even when optimal designs appear obtainable for the individual subsystems,the design of the actual entire system including those subsystems may not necessarily be optimal. This study addressed the optimization of a medium frequency (AM) amplifier, by assuming that the AM amplifier includes two subsystems:an antenna amplifier, and a receiver that acts as the load of the antenna amplifier. An analysis taking the effects of the interface between the subsystems into consideration was carried out to obtain an optimal design of the entire system. This showed that the entire system could be tuned efficiently to the desired characteristics by a two-phase design procedure, together with optimization of the subsystems.

  • 中島 尚登, 矢野 耕也, 小宮 佐和子, 坂本 和彦, 高木 一郎, 橋本 健一, 大畑 充, 戸田 剛太郎
    2004 年 12 巻 4 号 p. 50-57
    発行日: 2004/08/01
    公開日: 2016/09/30
    ジャーナル フリー

    In the medical field, evidence-based medicine (EBM) is attracting attention as a means of avoiding variations in diagnosis and treatment due to doctors' differ ing experience and knowledge. EBM provides a course of action in which the best available scientific evidence is first understood and then medical treat ment is given to individual patients in consideration of their specific clinical sit uations and value systems. The practice of EBM entails integration of a doctor's individual clinical expertise with clinical evidence obtainable from systematic studies, clinical epidemiological studies, and other studies performed at exter nal institutes. EBM provides a method of prior and posterior evaluation to avoid the common tendency of clinical diagnosis and treatment to vary depend ing on the experience and knowledge of individual doctors . A doctor faced with a large number of scientific opinions, clinical test results, and so on must make decisions by a pattern recognition process. The Mahalanobis-Taguchi (MT) method is helpful in EBM because it provides a common standard, not influ enced by doctors' experience and knowledge, that can be applied to this type of pattern recognition. In the present experiment, a unit space was formed from thirty (30) persons in normal health, and D2 was calculated from patient data. Disease progress and effects of treatment were evaluated from trends in both the D2 values and clinical test results. Worsening or improvement of a patient's condition could be determined from an increase or decrease in the value of D2. The single numerical D2 value is all that is needed to arrive at a common diag nosis independent of the doctor's knowledge and experience.

  • 阿部 誠, 渡邉 泰行, 豊福 勝也, 福永 晋
    2004 年 12 巻 4 号 p. 58-65
    発行日: 2004/08/01
    公開日: 2016/09/30
    ジャーナル フリー

    The purpose of this study wasthe optimization of vehicle component Characteristics to increase the passengers' sunvival space in collisions. A one-dimensionalnon-linear spring and mass model was used to simulate a truck undergoing a fronta barriertest. There are many more variations between different truck models than between different passenger automobile models, includding larged differences in platform Structure and payload capacity. Traditional methods select presumed worst-case conditions from all possible combinations of factors and Carry Out a tuning process to increase the survival space. These tuning methods, however, deal with only a single variation, and do not show Whether the tuning result will work for other variations. The authors therefore carried out a parameter design process in which differences in component weights in different variations and changes in strength characteristics due to deterioration were considered. The parameter design process was iterated by checking closely spaced leves of the control factors to find the optimal direction. This method enabled the optimal conditions to be estimated while maintaining the reproducibility of the gain.

  • -シミュレーションによる転写性の検討-
    白川 智久, 神原 憲裕, 杉山 昭, 齊藤 之男, 矢野 宏
    2004 年 12 巻 4 号 p. 66-72
    発行日: 2004/08/01
    公開日: 2016/09/30
    ジャーナル フリー

    This experiment used simulations to explore possible conditions by which molded products similar to the mold shape could be obtained. The experiment succeeded in finding conditions by which similarity between mold shape and molded product shape can be increased and demonstrated the possibility of reducing the number of mold prototypes, together with the further possibility of proto-type-less injection molding. The shrinkage ratio tendencies at different points in the molded products were also clarified.

事例研究
  • 留目 剛
    2004 年 12 巻 4 号 p. 73-82
    発行日: 2004/08/01
    公開日: 2016/09/30
    ジャーナル フリー

    Frames supporting the processing units of ful-color copiers and printers require high rigidity and accuracy. Traditionally, the general practice was to add rigidity to parts of the frames. A product recently put into volume production suffered a significant problem of frame shape deformation. To solve this problem, the frame shape was optimized by quality engineering and CAE using the finite element method. Six test pieces were optimized in this way and compared with each other to determine the degree of quality. The individually optimized test pieces were mated with an actual machine and further optimized for minimum deformation. As a result, a frame shape having 0.7-mm deformation under test conditions alld 0.16-mm deformation in actual machine conditions was obtained. This method also reduced the development cost by 45% and the number of process steps by 37%.

feedback
Top