4M情報を活用した作業分析・改善点抽出方式の開発

抄録

To achieve high-efficiency production, it is essential to develop accurate production plans and engage in continuous on-site improvements. In recent years, with the diversification of customer preferences, there has been an increasing demand for customized products. In the case of customized products, the working time can vary significantly depending on the model and specifications, even for the same process. Therefore, to improve the accuracy of the production plan, it is necessary to identify the attributes that cause variation and set standard working times. Furthermore, to extract improvement points in an environment with few expert workers, it is necessary to quantify the experience and know-how of exerting workers. In this paper, we propose an approach to integrate and analyze 4M information (huMan, Machine, Material, Method) from MES (Manufacturing Execution System), cameras, and sensors. We develop technology to accurately estimate standard working time for each process and a technology to automatically extract improvement points. The experimental results for the manual assembly process of customized products show that the accuracy of standard working time can be improved by 34%. Additionally, we confirmed that this technology can extract 85% of the work improvement points extracted by expert workers.