Research Of Optimization For Regional Bus Departure Plan

It is a key issue to be resolved in the field of public transport that improving the operation and management of public transport system to better satisfy the increasing travel demand of people and serve the economic and social development. Therefore, combined with the trends of systematic, network and regional development of public transport, this paper focus the departure plan, a critical step in bus operation and management, and proposed regional departure plan optimization model with corresponding solution. Key technologies in this paper will considerably assist improving coordinate scheduling of regional public transport, as well as the service level.The major research of this paper includes:(1) Establish departure plan optimization model for regional networks of public transport and use appropriate solving method. In this model, the starting point and interval of each line are set as variables, the minimum cost of passengers waiting time (including transfer waiting time and non-transfer waiting time) it built as the target, economic cost and service level are constraints. Then employ Particle Swarm Optimization (PSO) to solve the model, obtaining each line's starting point and non-uniform interval, as the basis for departure timetable.(2) Flow date is the fundamental basis for public transport planning. In the view of the high cost and unreliability, as well as lack of consideration for influencing factors, of existing OD survey, this paper carried out OD matrix estimation method. Based upon the number of passenger boarding-alighting and stop attraction, which were obtained by processing IC and GPS date, together with the analysis of travel characters of passengers, this paper built a structured algorithm for OD matrix estimation with an example followed. This algorithm is simple and easy to implement with all data dependent on computer information, and is more accurate and more economical than existing methods. Thus, it lays a foundation the for bus operation optimizing. (3)Establish link travel time prediction model. With consideration of road conditions, traffic flow and lane share, this paper employ multilayer hierarchical model to estimate time-varying parameters, and proposes bus travel time prediction model based upon multilayer hierarchical approach. The example shows that this model is accurate and easy to use.(4)Combine theory and practice. This paper set Guangzhou to apply three models, and verified the validity of them.This paper focuses on the study of three mathematical models when discussing regional transit departure plan optimization. In this progress, author combines scientific and practical ideas, try to make theories be of practical application value. On the whole, the results of this research could solve the regional public transport departure plan optimization, providing constructive comments for improve transit operation and service.