Research On Scheduling Optimization Of Peak Shared Bike Based On Improved Genetic Algorithm

In recent years,with the rapid development of economy,urban traffic congestion and environmental pollution are becoming more and more serious.To solve the traffic congestion,people begin to advocate low carbon travel mode,sharing a bike as a new type of green transportation way arises at the historic moment,park with its ease of use and flexibility,make up for the accessibility of public transportation and the shortcoming of flexibility is not high,not only help to ease the traffic congestion in cities,and to a certain extent,solve the residents last mile Travel problems.But with Shared cycling competition within the enterprise,the market share of bicycle type and quantity is increasing,due to Shared cycling space can transfer a high flexibility,tend to cause a certain period of time the distribution of the regional Shared cycling serious imbalance,leading to a variety of traffic problems and severely disrupted the normal traffic order.This phenomenon is caused by the lack of scientific and reasonable scheduling of shared bikes in the face of unbalanced spatial distribution of shared bikes.Therefore,it is necessary to conduct a systematic study on the scheduling problem of shared bikes,build a scheduling optimization model,determine a scientific and reasonable scheduling scheme,and put forward relevant scheduling suggestions for shared bikes enterprises.In this regard,the research work of this paper includes the following aspects:(1)literature review.Firstly,this paper summarized the relevant research results of scholars on the scheduling optimization problem of shared bikes by sorting out and summarizing the literature;Secondly,the origin and definition of shared bikes are summarized,and the system structure and function positioning of shared bikes are analyzed and introduced.;Finally,the demand characteristics and scheduling problems of shared bikes are analyzed.(2)Data analysis.Visualized analysis was made on the real cycling data of Mobike in Shanghai in August 2016 to summarize the laws of shared bikes in both space and time.In terms of the distribution of time rides,there are morning and evening peaks of shared bikes on weekdays,and the peak duration is about 2 hours.;In terms of spatial location distribution,shared bikes are mostly clustered around public places such as traffic stations,highlighting the role of shared bikes as traffic connecting points.(3)Model construction and calculation example analysis.Firstly,based on k-means clustering analysis,the dispatch region of Shanghai's cycling data was divided,and the time series analysis model was used to predict the incoming and outbound volume of shared bikes in each dispatch region.Secondly,a scheduling optimization model with fixed time window was built to minimize the total scheduling cost as the objective function,and an improved genetic algorithm was designed to solve the problem.Finally,taking Shanghai Mobike as an example,MATLAB software is used for modeling and solving.By comparing the solving results of the algorithm before and after improvement,the effectiveness of the algorithm is verified.By analyzing the relevant parameters of the model,relevant scheduling management suggestions are put forward for bike-sharing enterprises.