Scheduling Optimization Research On Public Bicycle System

In the context of vigorous promotion of low-carbon transport and encouragement of green travel by the government,the "public bicycle system" has develop-ed rapidly in the major cities in the country.But,the lack of public bicycles to be rented and the lack of places to return public bicycles have appeared in some of the rental points,which greatly reduce people’s satisfaction with the public bicycle system.Aiming at scheduling of public bicycle system in this paper,the models of public bicycle system are established,and the mixture of genetic algorithm and simulated annealing algorithm is designed to solve the models.Through reference to the study of public bicycle scheduling at home and abroad,this paper pays more attention to the scheduling optimization of single-scheduling center of public bicycle system.Firstly,the public bicycle rental system is introduced,including its advantages and facilities;secondly,the impact of scheduling costs and customer satisfaction of rental points on the scheduling path has been studied.Using the mobile phone APP "Wuhan huantou public bicycle" to obtain the scheduling demand information of 15 rental points of Hongshan District in Wuhan City,the mixed genetic simulated annealing algorithm is designed to solve the models,so the satisfactory scheduling programs are obtained.The main results are as follows:(1)The first model takes a minimum scheduling cost as the objective function,and uses the mixed genetic simulated annealing algorithm to get a specific scheduling scheme:only one scheduling vehicle is needed,which the specific scheduling path is 0→9→3→12→5→7→1→2→4→15→10→6→14→8→11→13→0 and the scheduling cost is 73.09 yuan.(2)The second model mainly takes account into the customers’ satisfaction with the rental point.Based on this problem,a satisfaction function with soft time window is introduced,and the weighting coefficient σi is introduced.And the value of σi is designed to be proportional to the scheduling demand quantity mi of the rental point i.The concrete formula of σi,is:The objective function expression is:By solving the model,the maximum weighted average satisfaction of the 15 rental points is 0.95,which requires four scheduling vehicles with the scheduling cost of 155.12 yuan that is higher than the minimum scheduling cost of 73.09 yuan in the first model by 112%.So,the scheduling cost is greatly increased if we blindly pursue the high satisfaction of the rental points.(3)The third model is established with the minimum scheduling cost as the objective under the premise of a certain weighted average satisfaction value.According to the results,it is concluded that the scheduling cost and the satisfaction with the rental points should be considered when designing the scheduling scheme,and it is necessary to balance their proportion according to the specific needs.