Part Agent that Proposes Replacement of a Part Considering its Life Cycle Using a Bayesian Network

Abstract Managing information on individual parts throughout their life cycle is a requisite to promoting their reuse. To do so, we have been developing a part agent system that uses network agents and RFID technology. A part agent proposes appropriate maintenance actions for a part based on information about its life cycle. This paper describes a method for a part agent to predict possible states of the corresponding part and to select an appropriate action by using a Bayesian network based on information collected on the state of the part, consumer preferences with regard to maintenance, and its usage.