Research Of Bus Scheduling Optimization Based On Chemokine Guide BFO Algorithm

As a major construction project, public transport is related to people’s lives. With the rapid development of transportation technology, more and more people pay attention to intelligent transportation scheduling techniques. The scientific, intelligent and paperless of bus scheduling will increase efficiency of bus scheduling, reduce cost so that passengers can get better service. The departure interval of bus scheduling is the most important of the research, because the departure interval is related to the profit of both companies and passengers, which is an import consideration of bus scheduling. Based on the actual situation, this paper establishes a mathematical model of bus scheduling problem. And solved this problem by improved CG-BFO, obtained a satisfactory solution, better than the BFO.The article took into account the factors of the departure plan. We use the bus company’s departure times and passenger’s waiting time as the optimization object, establish a objective function of the two minimum conditions. The departure interval is the decision variable, the maximum and minimum for each period departure interval, the bus company profitability condition and the passengers of bus are the constraints.There are many algorithms to solve this problem. But it has never been a good solution because of the complexity and particularity. Because bacterial foraging optimization algorithm has the characteristics of easy to escape from local optimization, this paper adopts this algorithm to solve the bus scheduling optimization. According analysis find the algorithm has three shortcomings. The first shortcoming is the step size fixity in chemokine caused reduced accuracy. The second shortcoming is the direction randomness in the chemokine caused the convergence speed reduce. The third shortcoming is the disperse probability randomness in the dispersal caused convergence speed reduce. And on this basis, presents improvements of three aspects. Introduced PSO as the mutation operator, provide guiding in chemotactic. Introduce a sensitivity value to make the step adjusting adaptively. In order to keep elite individuals, made the fitness has a large dispersal probability by changed dispersal probability. In the end of the paper, solve the bus scheduling problem by the improved CG-BFO. Through simulation results verify the validity of the model and algorithm and make a comparative analysis with the BFO.