An Intelligent Bus Dispatch System Based On Fuzzy Genetic Algorithms

With the rapid advance of urbanization process, a large number of city population growths have caused great pressure to the city transportation. Urban road resources are limited, and develop the public transportation is one of the effective methods to reduce the pressure on urban traffic. In recent years, the intelligent transport system become a hotspot in research of transportation, using GPS and GPRS technology to realize electronic bus stop and vehicles to forecast has been put into use in many cities. In this paper, we study a kind of dynamic transit scheduling model based on real-time traffic data, according to the real-time traffic and traffic situation decision optimal departure intervals,Bus scheduling determine the departure interval of the next period, and the collection of the passenger flow data is the current period of time, so the vehicle dynamic scheduling needs based on the short-term passenger flow forecast. Bus passenger flow data has the characteristics of periodic and mutability, therefore, use of the same period of historical data and real-time data collection of traffic flow for short time prediction. Using MATLAB’s least squares support vector machine toolbox, training the known passenger flow data to establish the forecast model, and then put the passenger flow data into the model to get the predicted value.Mathematical model to calculate the departure interval can be seen as an optimization problem, use the genetic algorithm optimized by fuzzy control to solve the model. The genetic algorithm has defects such as premature convergence, using fuzzy logic control to improve the genetic algorithm. The improved algorithm in the prophase and metaphase evolution stages keep good population diversity and restrain premature convergence; In the late stage of evolution, protect neighborhood close to the optimal solution of the population and speed up the convergence.Finally, considering the passenger waiting time and the full load rate of the bus to establish the objective function, take the predicted passenger flow data and real-time acquisition of vehicle information as input variable of the mathematical model, using the improved fuzzy genetic algorithm to solve the Mathematical model.