Abstract: The Train Operation Monitoring and Recording Device (LKJ) serves as the core safety protection equipment for railway locomotives, whose operation data supports multiple services including equipment maintenance, driver performance assessment and transportation optimization. Current mainstream LKJ data analysis systems are developed with hard-coded fixed analysis logic, which cannot adapt to evolving business requirements and updated data formats, resulting in poor flexibility, high maintenance costs and high technical barriers for development. Based on the technical stack of XML, FlinkCEP and AviatorScript, this paper develops a general low-code LKJ data analysis tool. A three-layer configurable decoupled architecture is constructed to enable visual drag-and-drop low-code development. A two-level automatic mapping mechanism is designed to convert visual models into executable sequence matching and dynamic computing logic automatically. A standardized library of analysis functions is also established to realize the reuse of business logic. Leveraging a distributed stream computing engine, the tool processes massive time-series LKJ data with built-in libraries of feature matching elements and computing operators. Practical applications show that this tool features favorable scalability and maintainability while ensuring analysis accuracy, and provides universal technical support for efficient mining and utilization of railway LKJ data.