Development of Design / Training / Management System for Visual Inspection Work Based on Human Visual Characteristics

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K15237
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 25010:Social systems engineering-related
• Research Institution: The University of Electro-Communications
• Principal Investigator: Nakajima Ryosuke 電気通信大学, 大学院情報理工学研究科, 助教 (70781516)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 目視検査 / 外観検査 / 作業設計 / 作業訓練 / 作業管理 / 作業改善

Outline of Final Research Achievements

In this study, a measurement system was developed to quantitatively evaluate the visibility of defects in visual inspection. In addition, a procedure and systematization of a method for designing an inspection method that achieves both inspection accuracy and inspection efficiency according to the characteristics of defects to be detected were implemented. As a result of these examinations, inspection environment, inspection method, training method, and management method are shown for target products of many visual inspection processes. Then, with the cooperation of the actual manufacturing industries, the inspection environment and inspection method are designed based on the methodologies obtained in the studies, and their usefulness is verified. As a result, it has been confirmed that even defects that were previously difficult to detect at the production site can be detected relatively easily, which has the effect of improving inspection accuracy and inspection efficiency.

Free Research Field

インダストリアル・エンジニアリング

Academic Significance and Societal Importance of the Research Achievements

本研究では、高度な技能が求められる目視検査作業を対象として、検査対象と検査環境、検査者の3つの観点から整理し、高い検査精度と検査効率を実現するための要件を検討した。そして、人間の視覚特性に合わせて検査環境と検査方法を設計・訓練するシステムを開発するとともに、それらの作業を訓練・管理するためのシステムを構築した。これらの研究の結果、実際の目視検査工程において製品や検出すべき欠点の特徴に合わせて検査精度と検査効率を両立する検査環境と検査方法の設計が可能となり、さらに訓練・管理システムを用いることで、特に高度な技能を要しない作業者でも高い検査精度と検査効率で作業が遂行できることが示された。