RELIABILITY MODELLING AND CONFIDENCE LIMIT ESTIMATION OF ELECTRIC MULTIPLE UNIT INCORPORATING MIXED DISTRIBUTED LIKELIHOOD RATIO

The Electric Multiple Unit (EMU) is a sophisticated asset renowned for its high reliability. The implementation of advanced health management technology can enhance the operational reliability and safety of EMUs. Preventive replacement interval planning, a key aspect of EMU lifetime management, requires significant investment in funds, manpower, and resources. However, it has the potential to substantially reduce maintenance management expenses and improve the profitability of EMU operations. This paper introduces a method for estimating reliability confidence limits for EMUs under multi-stage maintenance scheduling to extend replacement maintenance cycles. By employing the mixed distribution model as a two-level foundational framework, the point estimates of the fundamental model are determined using the EM algorithm, and the confidence intervals of the mixed distribution model are assessed through likelihood ratio statistics. A practical case study utilizing actual EMU data demonstrates the feasibility of extending preventive replacement cycles and validates the rationality of the model.