Study On Large-scale Activity Evacuation Plan Based On Scheduling Optimization Algorithms

Recently,more and more large-scale activities are held in cities.And large-scale activities have the characteristics of large number of people,high population density,and emergency evacuation time.It is necessary to evacuate the audience quickly to avoid traffic congestion and even stampede.Therefore,aiming at the evacuation problem of large-scale activities,study how to develop an evacuation plan based on scheduling algorithm to speed up the evacuation of the audience has practical application value.Although many scholars at home and abroad have done extensive research on the evacuation of large-scale activities,the existing evacuation plan mainly studies the timetable and route of public transportation around the large-scale activity venue(source station),ignoring the problem that the audience cannot be quickly evacuated in the “last mile” between the source station and the public transportation station.Therefore,in order to solve the “last mile” in the large-scale activity evacuation problem,this paper proposes a bike scheduling optimization algorithm based on K-Means clustering optimization algorithm and improved genetic algorithm.The main research work of this paper is as follows:(1)This paper proposes an evacuation plan based on the scheduling optimization algorithm,employing the method to deploy and schedule the shared bikes.And this paper studies the vehicle routing problems and solving algorithms.Simultaneously,this paper proposes a method of crowd flow forecast.Firstly,the public transportation stations around the source station are screened to determine the destination station of the audience.Secondly,the traffic flow of each station is forecasted based on the capacity of the station and the distance from the source station.(2)Combining K-Means clustering optimization algorithm and improved genetic algorithm,this paper proposes a two-stage bike scheduling algorithm.Firstly,it dynamically adjusts the number of clusters to meet the balance between the load capacity of the schedulingvehicle and the number of bikes in the cluster.Meanwhile,it optimizes the selection of the cluster center to ensure that the cluster centers are as far as possible.Then,aiming at the limitation that the genetic algorithm is easy to fall into the local optimum,from the perspective of population diversity and convergence,this paper proposes the elite retention strategy and eugenic strategy to improve the genetic operator and applies it to figure out the scheduling path in each cluster.(3)Based on the above studied contents,this paper designs a large-scale activity audience evacuation system,including: a)building a client to provide a visual interface;b)generating a corresponding evacuation plan according to a specific large-scale activity;c)simulating the evacuation process of the audience by a simulation evacuation algorithm.The experimental results show that the two-stage bike scheduling algorithm proposed in this paper has faster convergence speed and better global optimization ability than the traditional genetic algorithm and the existing improved algorithms.On the evacuation problem in the “last mile” of large-scale activities,the evacuation plan based on the scheduling optimization algorithm proposed in this paper can effectively improve the evacuation efficiency and reduce the evacuation time.