Maintenance Decision Support System for Multiple Units Based on High-Quality Knowledge Graphs

As rail transit rapidly develops and its applications expand, the maintenance of multiple units (MUs) plays a pivotal role in ensuring the safety of train operations and enhancing operational efficiency. However, the maintenance process encompasses a vast amount of complex and dispersed professional knowledge, presenting significant challenges in the effective management and utilization of this information. Existing knowledge management methodologies often struggle to integrate maintenance-related knowledge of MUs, leading to low rates of knowledge utilization. To address these issues, this study proposes a decision support system (DSS) that integrates knowledge graphs with large language models (LLMs). We have designed an automated graph construction framework to systematically extract knowledge from professional literature, thereby building high-quality knowledge graphs. By incorporating the BGM-RSRL prompt framework with LLMs, we have developed an efficient DSS. Experimental results demonstrate that the system accurately identifies complex maintenance requirements for MUs and provides comprehensive decision support

Decision support system · Graph · Knowledge base · Knowledge engineering · Knowledge graph · Knowledge-based systems · Maintenance engineering · Advanced Graph Neural Networks · Advanced Research in Systems and Signal Processing · Machine Fault Diagnosis Techniques