Increasing Maintenance Efficiency in Urban Rail Systems: A Proactive Approach Using Time Series Forecasting

Abstract

Electric trains in Thailand are facing increasingly intensive maintenance demands due to their extended service life. This study introduces a proactive approach to optimizing maintenance strategies for urban rail signaling systems, with the aim of enhancing efficiency and reducing maintenance work orders by at least 40%. Engineering management techniques were applied through the Seven Quality Control Tools, and a predictive maintenance model was developed using time series forecasting. The Decomposition Model, selected for its lowest forecasting error, was employed to generate maintenance schedules. As a result, maintenance work orders decreased by 81.82%, the mean time between failures (MTBF) increased by approximately 5,529.98 hours, and the mean time to repair (MTTR) decreased by approximately 2.69 hours within four months. These outcomes demonstrate the effectiveness of integrating predictive and preventive maintenance strategies to enhance system reliability and operational performance.