ミズヤマ ハジメ   MIZUYAMA Hajime
  水山 元
   所属   青山学院大学  理工学部 経営システム工学科
   職種   教授
言語種別 日本語
発行・発表の年月 2007
形態種別 学術雑誌
査読 査読有り
標題 製造履歴データの探索的分析のためのニューラルネット援用型多段階品質情報推移モデル
執筆形態 共同
掲載誌名 日本設備管理学会誌
巻・号・頁 166-176頁
著者・共著者 淺田克暢, 山田賢太郎
概要 This paper presents an exploratory data analysis approach called the multi-stage quality information model (MSQIM) based on artificial neural networks (ANNs) for streamlining the process of detecting causal factors of quality defects in a multi-stage production system with the help of a manufacturing database. This approach utilizes an ANN model to extract information on the resultant quality of a product, i.e. quality information, contained in the process data collected until each process step. This results in plural ANN models corresponding to the process steps of the production system. The approach then traces how quality information varies along the process steps, in both quantitatively and qualitatively. How the quality information amount changes makes it possible to identify the process steps that require further focus. Studying the relationships between consecutive ANN models guides hypothesis generation about causal factors of defects and defect-causing mechanisms. An industrial example shows how this approach works.