Study On On-line Scheduling Methods Of Urban Public Transport Based On Quick Response To Passengers' Needs

Vigorously developing urban public transportation is the most effective way to solve transport problems, such as urban traffic jams, frequent accidents and serious pollution of the environment. It is of great practical significance to study how to organize various resources reasonably for public transportation operation, because it is critical to maximize the use of public transports advantages and enhance the overall efficiency and service levels of public transportation system.How to schedule buses flexibly in accordance with the pattern of passenger flows based on the ITS technologies environment was studied in this paper. That is significant to maximize operating efficiency of public transportation system. Passenger flows were divided into two parts, which were daily passenger flow that is relatively stable part and special passenger flow with certain OD distribution which is unusual.To meet the needs of the daily passenger flow, a new bus scheduling method, named short-circle bus scheduling method based on quick response to passenger needs, was proposed. This method is different from the traditional one as follows: a) shortening the scheduling cycle, b) use of forecasted high-precision passenger flow data, c) headways are variable with passengers'needs. And a multi-objective model was presented under the single-type vehicles scheduling mechanism, in which the social and economic benefits were considered. The multi-type vehicle scheduling method also was studied so as to further improve scheduling flexibility and operating efficiency by optimizing the structure of transportation capacity. Also genetic algorithm and cooperative co-evolutionary algorithm were designed to re-solve those models online.To solve the problem in which OD distribution is too concentrated, stop selected bus scheduling method based on path choice was created. The method allow buses stopping at selected stops not every one and moving along the optimal paths instead of the original bus lines, meanwhile some stops with less needs are passed.