State Prediction and Parts Replacement of a Manipulator Based on Assembly Model

Managing individual parts over their entire life cycle is essential to realize effective reuse of mechanical parts for the development of a sustainable society. Therefore, we propose a scheme in which a part manages itself and supports user maintenance activities. For this purpose, we are developing network agents programmed to follow their real-life counterpart parts throughout their life cycle. We refer to this network agent as a “part agent”; it provides users with appropriate advice on the maintenance of its part and promotes the circulation of reused parts. A part agent uses life cycle simulation (LCS) of the corresponding part to predict the future state of the part by using its life cycle model based on the conditions of its environment and usage of the part. Different from life cycle design, a part agent performs the LCS repeatedly to predict the short-term future mainly in the use/operation phase. LCS uses models representing aspects of the part that a part agent corresponds to, such as the life cycle of the part, its behavior, and its deterioration. One essential point for the models is the representation of the product assembly structure. Assembly plays a crucial role in dealing with the cases that occur in the LCS; an assembly product may fail because of the deterioration of its part, and the repair and replacement of that part can affect the life cycle of the assembly. To manage the reuse of parts in the LCS of assembly, the assembly model should manage the change of properties and behavior of parts because of deterioration, as well as the replacement of parts, including their properties and the effect of change of properties on the assembly. In a previous study, we proposed a method for creating a causal model with stochastic information that represents the progress of deterioration of the part and developed a method for predicting its state based on the model by using the LCS of parts. We have created the assembly model for a manipulator. In this paper, we describe a case study of the prediction of the state of parts and their replacement based on the assembly model.