Bayesian Estimation for the Reuse of Mechanical Parts Using Part Agents

Fukunaga, Yoshinori; Fukumashi, Yuuki; Nagasawa, Atsushi; Hiraoka, Hiroyuki

doi:10.1007/978-981-13-1181-9_3

Cited by 1 publication

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have still issues to apply this scheme to practical application. One of those issues is creation of the failure model and, as a first step, we are considering a method to improve the failure model based on a library of causal relations [7].…”

Section: Qualitative Prediction Of Failuresmentioning

confidence: 99%

Replacement of Parts by Part Agents to Promote Reuse of Mechanical Parts

Hiraoka

Nagasawa

Fukumashi

et al. 2017

Product Lifecycle Management and the Industry of the Future

Self Cite

View full text Add to dashboard Cite

Replacement of parts is a crucial task for the reuse of parts. In this paper, functions of part agent we are developing that support the reuse are described. A part agent manages life cycle of a part using a network agent and an RFID tag attached on the part and generates advices for the user on the maintenance of the part in order to promote the reuse. Part agent we are developing to support the replacement has the following functions. First function is to detect deterioration of the part from its sensory data. As an example, simple 3dof manipulator is developed with modularized link structure. Part agent is assigned for each module and generates advice on its replacement. Second function is to calculate the possibility of failures of part based on causal relation among user operations and detected events. Third function is to support users to disassemble used parts. Part agent generates disassembly procedure of the part and shows it to the user using augmented reality technique. In this paper, the planned scheme for part agent to support reuse of part with these functions is described and preliminary result of the development is shown.

show abstract

Section: Qualitative Prediction Of Failuresmentioning

confidence: 99%