Research Of Demand Response Based Real-time Custom Bus System

Along with the high speed of urbanization in China, the scale of cities is growing all the time, at the same time, the growing number of personal used vehicles demonstrated an accent of mobilization in mega-cities. Traffic jam, air pollution and traffic accident has become the so called’city illness’. Priority for public transport is now the fundamental policy in our city traffic, like many other contraries. Nevertheless, because of the wrong city planning and constructing time sequence, there are many mega residential areas or new city areas exist. Traffic demand is always exceeding the supply in this area at commuting time, other than these time, traffic supply exceeds demands a lot, empty buses can be seen in street a lot, energy and human resources are wasted. The common problem in these areas is the low passenger density, and the disperse of passengers, when buses run according to fixed route and time table, transport capacity can not be utilized sufficiently.This essay analyzes the demand response bus(DRB) abroad, after comparing three different kinds of DRB, a demand response based real-time custom bus system is built to fit low density traffic demand. It belongs to flexible route bus, except for fixed stops, temporary stops can be ignored, namely the bus can choses which demand to response, and change its’ route accordingly. This essay analyzes the custom bus system, demonstrates different kinds of customized models, introduces key factors that consist this system, and design the customized process in the end;Two models are built to describe the system, namely the ’one car model’ and the ’two car model’, each model is separated into two phases also, according to the depart of bus, the two phases are pre-departure phase and the post-departure phase, Open Vehicle Route with Time Window(OVRPTW) and Multi-Objective Optimization model are built to describe the two phases.After analyzing the characteristics of these two models, this essay bring up the solving methods, Genetic Algorithm(GA) is used to solve the OVRPTW, encoding method, cross operator, mutation operator and selecting operator are designed according to the model. Sequential decision based NSGA-II algorithm is designed to solve real time problem and multi-objective optimization model, taking the latest Pareto front as the fix route of next phase.This essay mainly design a demand response based real time custom bus system, according to the characteristics of low passenger density areas, it can be popularized in certain areas, the transportation capacity can be sufficiently used and provide a high level public transportation service at the same time, High technology transportation and Green Transportation is thus realized.