The Application Of Fuzzy Multi-objective Programming In Public Transit Optimal Allocation

Fuzzy multi-objective programming (FMOP) is an important optimization way to solve many complicated but practical problems. This paper presents the modeling and the solution of FMOP, and its application in public transit optimal allocation.Public transportation is an important component of urban transportation. Urban modernization, especially the improvement of urban functions, is inseparable from urban traffic optimization and improvement. A reasonable public transportation can effectively ease the contradiction between the capacity and traffic, keep the interest balance between passengers and bus companies, reduce passengers' travel costs and improve the environment as well. Furthermore, it can increase the efficiency of vehicles, and improve the economic benefits and social effects of the bus company.Given the multiple objectives and uncertainty when dealing with the problems in bus allocation, the author adopts FMOP to work these problems out in order to make the best use of buses. This paper shows the characteristics of transport system in China, followed by a model multi-objective programming model that can simulate the bus driving with the help of FMOP technique, and gives solution. Finally, the author select some cases to testify the validity of this model. The main contents of this paper are as follows.1. Accurate and logical data are prerequisite and basis for dispatching buses. This article describes the details and ways of investigation to these data, analyzes data based on the distributive characteristics of passenger flow in time and space, which provides a reasonable basis for dispatching buses.2. The public transit fuzzy multi-objective linear programming (PTFMOLP)model is established. The multiple objectives include the passagers' interests which is made of the cost of waiting a bus and the cost paid in bus and the profits of the bus company. Then with this model, the author introduces some approaches and solutions of PTFMOLP.3. Based on the PTFMOLP model, this paper applies two approaches to solve the problem: the fuzzy programming approach and the"fuzzy optimal solution"approach based on genetic algorithm. To use MATLAB to finish related calculations and above-mentioned models to analyze some specific bus.4. This paper shows the fuzzy comprehensive evaluation method and makes a remark on optimal program. Finally, the readers can get the orders of excellence of these programms.