Diffusion approximation for exploring the correlation between failure rate and bus-stop operation

This study proposed a model incorporating a diffusion approximation method to explore the correlation between failure rate and bus-stop operation, which included bus arrival rate, service time, variation coefficient of service time, variation coefficient of bus arrival headway, and bus stop capacity. According to the results, the linear relationship had more accurate and reliable fitted values (with 6.98% of MAPE and 0.935 of R-square) for single-berth stops and the exponential relationship had better performance (with 0.799 of R-square and 14.86% of MAPE) for multi-berth stops to explore the expression of failure rate. The results of sensitivity analyses showed that failure rate increased with the increasing ratio of arrival rate to service rate, and added berths produced diminishing returns in failure rate and capacity. In addition, the returns in capacity were influenced by failure rate, and variation coefficients of bus arrival headway and service time.