Practical health monitoring diagnosis method for aged bridges and proposed for bridge inspection work efficiency

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K04223
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 22010:Civil engineering material, execution and construction management-related
• Research Institution: Tottori University (2022-2023); Fukushima National College of Technology (2021)
• Principal Investigator: Emoto Hisao 鳥取大学, 工学研究科, 准教授 (90556698)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 橋梁健全度評価 / 点検技術者養成 / Artificial Intelligence / MR-HMD / Augmented Reality

Outline of Final Research Achievements

The results of this project showed that the use of random forests as a bridge health condition method is effective. We also proposed a data expansion method for the problem of low training data (i.e., damaged data and its evaluation), which is typical of the civil engineering field. Using Random Forests, we were also able to calculate importance levels that can derive important factors for health condition. To improve the efficiency of inspection work, a tool for training seminars using MR-HMD was developed for the training of inspection technicians. In addition, by dividing the content of its seminars into levels, the company was able to create a system that is effective for learning. In addition, we have attempted to improve efficiency by using AR to display inspection results.

Free Research Field

維持管理工学

Academic Significance and Societal Importance of the Research Achievements

わが国の橋梁の70％は、市町村が管理する橋梁である。市町村や地方の技術者のレベルや技術者不足を補うためには、橋梁点検技術の向上や新規産業からの参入とその若手技術者の教育が重要となる。さらに、その結果を評価する実用的な健全度診断が重要となる。橋梁の健全度診断では、技術力による差異が発生する。これは、暗黙知を含む項目が存在することもあり、その重要な項目の抽出も重要となる。
以上から、点検業務の効率化のためにMR-HMDを利用した点検講習会用のツール開発と橋梁の健全度診断における暗黙知の抽出や健全度判定の点において学術的・社会的な意義がある。