品質
Online ISSN : 2432-1044
Print ISSN : 0386-8230
44 巻, 4 号
選択された号の論文の7件中1~7を表示しています
特集
  • 吉野 弘規
    原稿種別: 特集 『ものづくりにおける解析手法の最前線』
    2014 年 44 巻 4 号 p. 362-367
    発行日: 2014/10/15
    公開日: 2017/06/05
    ジャーナル 認証あり
    A low computational complexity and high-accuracy method for clarification of oil temperature in a hydraulic electronically controlled rear differential was developed for implementation to a vehicle control unit. Because the target system uses a motor-driven electronic hydraulic pump to increase hydraulic pressure, a correlation was considered to exist between oil temperature and motor current. Theoretical analysis of the structure of the system demonstrated that a linear model was sufficient for clarification of the state of oil temperature from motor current, and a clarification model for the state of oil temperature using logistic regression was formulated. In addition, in order to balance accuracy and frequency of clarification, a method of calculation of posterior probability when continuous multiple clarifications were executed using only the output of the logistic regression model was developed. The results of application of the method to data obtained from the actual target system confirmed that the proposed method was able to clarify the state of oil temperature with a high degree of accuracy.
  • 近藤 拓
    原稿種別: 特集 『ものづくりにおける解析手法の最前線』
    2014 年 44 巻 4 号 p. 368-372
    発行日: 2014/10/15
    公開日: 2017/06/05
    ジャーナル 認証あり
    Many reports have described the application of statistical quality control (SQC) techniques to efficiently resolve issues in manufacturing. However, factor analysis can be difficult to implement for phenomena with large numbers of factors that have complex causal structures. In conventional factor analysis, factors are organized into an Ishikawa diagram considering the causal structures of the products or processes. However, these structures are frequently ignored at the analysis stage. Consequently, factor analysis becomes difficult if the causal structure is complex. Structural equation modeling is capable of considering multiple causal structures and is therefore regarded as a possible solution for this issue. However, the details of structural equation modeling are unclear since no research has been published about its application to similar issues in the general industrial field. Therefore, the details and effectiveness of structural equation modeling were identified by applying this method to factor analysis of the manufacturing processes of transmission parts.
  • 室崎 隆
    原稿種別: 特集 『ものづくりにおける解析手法の最前線』
    2014 年 44 巻 4 号 p. 373-378
    発行日: 2014/10/15
    公開日: 2017/06/05
    ジャーナル 認証あり
    We studied a welding inspection for terminal welding of Fuel Pump with image processing. Color Extraction Method with blob analysis had gray zone of 13%. The gray zone is a vague area which is mixture area of good pieces and defective ones. There are two methods to determine a decision boundary. One is the Mahalanobis distance method that calculates the decision boundary given the distribution of learning data statistically. The other is SVM method that calculates the boundary that maximizes the margin of the border to identify all of the training data provided. We decided to adopt the SVM method in order to ensure the margin which does not flow out absolutely defective. First, we made a SVM tool for selecting the three features and could display the SVM border surface in 3D. But, the SVM method possibilities of misjudgment still exist for the defective work that cannot be covered by the supervised learning. Then, we applied SVM with the process capability index (CP). As a result, we could decrease excessive judgments as defective goods greatly.
  • 仲矢 実, 李 新春
    原稿種別: 特集 『ものづくりにおける解析手法の最前線』
    2014 年 44 巻 4 号 p. 379-384
    発行日: 2014/10/15
    公開日: 2017/06/05
    ジャーナル 認証あり
    MIRROR PLANT, the on-line plant simulator can perfectly simulate dynamic plant behavior. One of applications of MIRROR PLANT visualizes inside the plant where sensors cannot be actually installed. The plant operator can see the key performance index related with product quality in real time. And the operator can also see the prediction of the index. MIRROR PLANT uses physical plant modeling based on physical and chemical laws. However, only the physical model cannot express some phenomena in the plant, especially those which are not physically analyzed yet. We proposed a hybrid model of a physical model and a statistical model based on historical plant operation data. Combining the physical model and statistical model improves the estimation accuracy of the statistical model. In addition, by applying the just-in-time (JIT) modeling to the on-line update of the statistical model, it can deliver exact estimation even during rapid changes in the plant behavior.
連載
論文誌編
報文
  • 鈴木 理史, 田中 研太郎, 宮川 雅巳
    原稿種別: 報文
    2014 年 44 巻 4 号 p. 419-428
    発行日: 2014/10/15
    公開日: 2017/06/05
    ジャーナル 認証あり
    Xiao et al.(2012)は,カンファレンス行列を用いた3水準スクリーニング実験計画を与えた.この計画は取り上げたm個の因子に対して2m+1回の実験を行うもので,すべての線形主効果が2因子交互作用および2次主効果と交絡することなしに不偏推定できる.しかし,2因子交互作用と2次主効果は部分的に交絡し,不偏推定量を得ることはできない.そこで本論文では,この計画にm+1回の追加試行を加えることで,特定の因子と他のすべての因子とのm-1個の2因子交互作用と,その因子の2次主効果に対して不偏推定量が構成できるようになることを示す.また,この不偏推定量の存在によって最小二乗推定量の不偏性が示される.数値実験により最小二乗推定量の推定精度を精査する.
  • 黒木 学, 林 崇弘
    原稿種別: 報文
    2014 年 44 巻 4 号 p. 429-440
    発行日: 2014/10/15
    公開日: 2017/06/05
    ジャーナル 認証あり
    本論文では,観測データが線形構造方程式モデルにしたがって生成されているとき,中間特性を用いて処理変量から品質特性への総合効果を推測する問題を考える.処理変量から品質特性への直接効果がなく,かつ中間特性が一つの場合,Cox(1960)は逐次的回帰モデルを用いて総合効果を推定したほうが,単一の回帰モデルを用いた場合よりも漸近的な意味で推定精度が高いことを示している.しかし,実際の品質管理の現場では,処理変量が単一の中間特性を経由してのみ品質特性に影響を与える状況は少ない.そこで,本論文では,複数の中間特性を含んだ補助特性の集合を用いた場合の総合効果の推定精度について考察し,単一の回帰モデルを用いた場合よりも逐次的回帰モデルを用いて総合効果を推定したほうが推定精度が高いことを示す.また,総合効果の推定精度の観点から補助特性の選択指針を与える.さらに,数値実験とIC製造工程の分析事例をとおして,本論文で与えた結果の有用性について検討する.
feedback
Top