Puplic Transportation Timetable Optimization Based On GPS Data Analysis

Bus timetables is the most important bridge between public transport companies and passengers seeking a reliable transit service. Inappropriate or inaccurate timetable will not only get passengers into trouble, but also led to the overall impression of the deterioration of public transport. At present, bus timetable execution strategy using by the companies can only control the first and last site. It ignores the phenomenon that the buses operating efficiency among sites are low, and the stability of stations is different. If not controlled, the public transport efficiency and service level will decline, and public transport passengers will be lost.With development of intelligent transportation technology, especially these GPS devices widely used in-car, bus GPS data can record the exact time of each vehicle reaching each site, which provides technical and data supporting for studying the operating characteristic of buses in the line. But the traditional research of buses is primarily focused on the running time factors and predicting arrival time. However this way cannot solve the problems of public transport. Therefore, this paper analyzes the implementation timetable from the operating characteristics of buses based on a large number of bus GPS data. According to the identified problems, this paper proposes to set key sites. It can improve the stability of the bus running between stations by controlling the operating time between key sites, and also optimize schedules and improve the efficiency of public transport. Specifically, this paper mainly studies the following aspects:(1) Based on GPS data, it can get a true reflection of the spatial and temporal distribution about running buses, analyze the composition of buses running time, and compare the effects of time and vehicle. At last it also analyzes the relationship between the running time and dwelling time.(2) In order to optimize schedules, and increase the stability of buses running between stations, this paper defines key sites and designs efficient methods of finding the key sites. So that it provides a guideline for optimizing schedule.(3) For ensuring a true reflection of the distribution of the running time, this paper selects the normal distribution and other 4 probability distributions to describe the law of running time between stations. It obtained the conclusion that the normal distribution is the best fitting form.(4) This paper studies these two technology in timetable optimization. The first one is grid spacing algorithm which is based on key sites controlling. The other one is the distribution that describes running time between stations. Then it can get the more stable and reliable timetable in the stations.