Public Transport Optimal Dispatching Based On Particle Swarm Optimization And Patrheno-genetic Algorithm

With the fast development of science, technology and modernization, todaypeople’s living standard and the demand of trip quality are increasing gradually.However, China is a populous country and per capita road resources are very limited.Urban expansion and private cars enhancement year by year will lead to terribletraffic congestion, bad citizen’s trip environment and so on. Therefore, in order toeffectively alleviate urban traffic pressure and improve its backward appearance, thedevelopment of public transport has become imperative. Due to its importantposition and positive significance in the development of public transport, busoptimization dispatching has not only been paid more attention by relevantdepartments but also become scholars’research hotspot.Although there are a lot of optimization methods used to solve bus schedulingproblem, but in view of its complexity and particularity, this problem has not beensatisfactorily resolved yet. This thesis introduces several common intelligenceoptimization methods especially particle swarm optimization (PSO) andpartheno-genetic algorithm, and designs PSO and partheno-genetic algorithm byutilizing the strong complementarity between these two algorithms. The algorithmuses the solutions of PSO as the initial population of partheno-genetic algorithm, aswell as adding migration, reinsertion and other operations in intermediate sub-group.At the same time, the algorithm stability and convergence rate have been improvedeffectively.In the process of researching bus scheduling problem, firstly this thesisconsiders the influence of bus company earnings, passenger waiting time and buscrowded degree, and then establishes a mathematical model, which can correct theinterest tendency of company or passenger. Secondly the optimal departure timeinterval is obtained through solving this model by PSO and partheno-geneticalgorithm. Thus provide a theoretical basis for the bus optimization dispatching.Finally, simulate and analyze the traffic data of a certain bus route by MATLAB.Simulation results show the superiority of the model and the algorithm.