Support of part reuse based on Bayesian estimation by part agents

2018 Fiscal Year Final Research Report

• Project/Area Number: 15K05772
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Design engineering/Machine functional elements/Tribology
• Research Institution: Chuo University
• Principal Investigator: Hiraoka Hiroyuki 中央大学, 理工学部, 教授 (20165161)
• Project Period (FY): 2015-04-01 – 2019-03-31
• Keywords: ライフサイクル工学 / 部品リユース / ベイズ推定 / ネットワークエージェント / 循環型社会

Outline of Final Research Achievements

For the promotion of reuse which is effective for the realization of the circular society, a system was developed that gives the advice on the reuse to the user by the part agent which manages the life cycle of an individual part by combining RFID and network agent. Bayesian estimation method was applied to deal with the deterioration and failure that are difficult to predict. A method was developed that generates a causal relation model for the failure estimation based on the deterioration process, and, on the basis of it, generates Bayesian network by giving initial conditional probability by the deterioration simulation. Furthermore, an application method of a product model for simulation was developed, as well as a method to present disassembly instructions for removing parts from assemblies over images of the assemblies using AR technologies.

Free Research Field

計算機援用設計製造

Academic Significance and Societal Importance of the Research Achievements

手法の有効性は今後さらに確認する必要があるが，劣化・故障のプロセスにベイズ推定を適用する手法は新規性があると考えている．リユースの促進は循環型社会の実現と発展に大きく貢献することが期待でき，社会的意義が大きい.