Transit Service Reliability Based On Stops-Skipping Strategy

Urban traffic jam has developed into a severe problem recent years, hence giving priority to the development of urban public transit is considered as a critical path to solve the problem. Improving the attract ability of urban public transit to residents primarily relying on building efficient transit service system and massive attention has been paid to this topic from researchers. Bus IC card has been implemented for years; therefore data draw from IC card records are considered to be reliable to some point. Hence, research of unreliable phenomenon of transit service and proposing optimizing strategy afterwards base on IC card record has strong potential in improving bus service reliability.At first, this dissertation gives a comprehensive literature review which had demonstrated a clear research structure of transit service reliability; specifies the definition of transit service reliability; proposes6categories of metrics from the purpose of passengers and gives corresponding formulas and benchmarks afterwards; lists several regulation strategies to prevent or recover from disorder situation.Then this dissertation depicts one of the most common issues of transit service-bunching, which is a primary reason that introduces unreliability into transit service system. Considering the typical process and reasons of bunching, a case study which IC card records came from a specific route of Beijing transit service system is given. To conduct the research of bunching, the data of headway between adjacent arriving buses and its coefficient of variation from morning peak IC card records are gathered.Later on, based on the analysis of the necessary of stops-skipping strategy and the disadvantages of traditional stops-skipping strategy together with results from other researchers, this dissertation proposes an improved stops-skipping strategy to relieve transit service unreliability. A transit service stops-skipping model is finally proposed, whose object function is to minimize Total Ride Time (both waiting time and on-board time). The stop-skipping strategy is formulated as a nonlinear integer programming problem. The problem solution relies on an exhaustive search method. The study is conducted to find every potential stops-skipping range and identify the one that minimized object function value. Finally Monte-Carlo simulation base on data from a specific route is used to tentatively test the strategy. The analysis of impact of passenger arrive rate on object function as well as the improved stops-skipping strategy is also included. The result of simulation indicates that, when on bunching, improved stops-skipping strategy has obvious positive effect on reducing Total Ride Time of passengers.