Abstract: Because the arrival time prediction of public transit is influenced by many factors and all kinds factors on the prediction accuracy can’t be measured,it is difficult to use the traditional mathematical model to solve this problem.In this paper,a fuzzy neural network model based on genetic algorithm was used to predict the arrival time of the bus.In this model,genetic algorithm and fuzzy inference system are integrated into the multi-layer feed forward neural network.And the initial value of each parameter of the network is initialized and updated by the membership degree of fuzzy rules.At the same time,multi-population adaptive genetic algorithm for macro searches is used to improve the network optimization ability.The paper used a bus line in Chengdu city running time prediction as an example to make the simulation.The simulation results show that,the fuzzy neural network based on genetic algorithm of bus arrival time prediction model has higher accuracy and reliability.

Key words: Bus arrival time prediction,Multilayer feed forward neural network,Fuzzy logic system,Multi population adaptive genetic algorithm