基于自学习遗传算法的共享单车投放点选址研究

Research on location selection of shared bicycle delivery point based on self-learning genetic algorithm

本文将居民作为客户,共享单车投放点作为设施,客户与其最近设施之间的最短距离为目标,基于P-中值问题对实际问题进行建模并对共享单车投放点进行近似最优解的求解。在P-中值模型的基础上,本文针对遗传算法本身由于初始参数固定导致灵活性不高的短板,提出参数自适应的遗传算法,并将其与标准遗传算法在求解精度等多方面进行对比。与遗传算法对比,自适应遗传算法在共享单车选址这类计算规模较大、一般搜索效率较低的问题上,在提高收敛速度的同时也可以避免陷入局部最优解。本文在西班牙马拉加省417个社区的实际案例中,分别使用自适应遗传算法和遗传算法对其共享单车投放点进行选址求解,结果表明,在同等条件下适应遗传算法在提高收敛速度的同时可以求得更优解。

In this paper,residents are regarded as customers,shared bicycle drop points are regarded as facilities,and the shortest distance between customers and their nearest facilities is the goal.Based on the P-median problem,the actual problem is modeled and the approximate optimal solution of shared bicycle drop points is solved.On the basis of the P-median model,this paper proposes a parameter-adaptive genetic algorithm for the shortcomings of the genetic algorithm itself,which is inflexible due to the fixed initial parameters,and compares it with the standard genetic algorithm in terms of solution accuracy and other aspects.Compared with the genetic algorithm,the adaptive genetic algorithm can improve the convergence speed and avoid falling into the local optimal solution in the problem of large-scale calculation and low search efficiency such as location selection of shared bicycles.In this paper,under the actual case of 417 communities in Malaga Province,the adaptive genetic algorithm and the genetic algorithm are used to solve the location of the shared bicycle placement points respectively.The results show that the adaptive genetic algorithm can improve the convergence speed while improving the convergence speed under the same conditions.A better solution can be obtained.