Research on Complex Product Maintainability Configuration Design Method Based on Knowledge Graph

A favorable maintainability design can reduce massive cost and resource consumption. Aiming at the problems that the semantic knowledge of the operation and maintenance stage is not integrated into the configuration design process and it is difficult to obtain the implicit configuration knowledge, this research proposes a maintainability configuration design method for complex products based on a knowledge graph. Firstly, an ontology is constructed to represent the design domain knowledge, maintenance service knowledge, and the semantic relationships between them. Combined with the enterprise database, the maintainability configuration design knowledge graph is constructed as the configuration model. Knowledge graph and FP-Growth are used to mine objective configuration rules and update the rules into the model. Then, the pellet reasoner is used to solve the physical architecture that meets the order requirements of the customer. An optimization model is established, which aims at the maximum maintenance efficiency and the minimum life cycle cost of the product, and the NSGA-II algorithm is used to solve the optimization model to obtain the configuration scheme. Finally, the feasibility of the proposed method is verified by taking the drive device and wheelset composition of a high-speed train bogie as a case study.