Study on introduction and effect of the inspection tool in consideration of a human factor in the manufacturing industry

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K13766
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 07080:Business administration-related
• Research Institution: Ibaraki University
• Principal Investigator: HARAGUCHI HARUMI 茨城大学, 理工学研究科(工学野), 講師 (70796325)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 生産管理 / 作業者訓練 / 人的要因 / 機械学習 / 検品作業 / 品質管理

Outline of Final Research Achievements

This research is a three-year project that focuses on the introduction of inspection tools that take into account human factors in the manufacturing industry and the verification of the effectiveness of these tools. An automatic inspection tool based on a discrimination model that reflects the differences in judgement of different inspection workers using machine learning was proposed for inspection targets for which a simple quantitative threshold cannot be set. In the course of the research, it became clear that the improvement of the accuracy of the sample itself, and thus the equalisation of the discrimination accuracy of the inspection workers, was essential to improve the accuracy of the discrimination model, and we proposed a sample labelling tool and a worker training tool. For the operator training tool, continuous training experiments were conducted with student subjects, and the changes in the inspection operator's skills were analysed.

Free Research Field

生産管理

Academic Significance and Societal Importance of the Research Achievements

本研究では，歯科医療器具部品の検品を支援するツールの開発と導入効果の測定を行った．対象となる部品はひとつとして同じ形状のものが無いため単純な閾値設定による自動検品は不可能なうえ，検品作業者によっても判断が異なる場合がある．そこで，作業者による判断の違いを反映した部品の分類を行い，機械学習を用いた検品支援ツールを開発した．また，サンプルのラベル付けツールおよび作業者訓練ツールの提案を行い，作業者の判断基準を平準化することによって，機械学習で使用するサンプルの精度が向上し，最終的に検品支援ツールの精度も向上させることが分かった．