摘要
针对使用蚁群算法对共享单车进行配送得到最优解的问题,通过设定蚂蚁觅食路径上初始信息素浓度和信息素更新方法获得更优的配送距离,对蚁群算法进行改进。将对共享单车使用量有影响的特征因素输入到XGBoost模型中进行预测,根据转移概率通过轮盘赌的方式选择接下来要访问的租赁点,利用提出的基于初始信息素衰减的方法进行更新,以快速得到更优的解。通过对某学校内的14个租赁点进行实验,在初始信息素衰减的基础上,改变初始信息素浓度,可以在较短时间内获得更短的配送距离,比基本蚁群算法的距离缩短了约1%,实验结果验证了该算法的有效性。
To solve the problem of using ant colony algorithm to get the optimal solution for the distribution of shared bicycles,this paper improves the ant colony algorithm by setting the initial pheromone concentration and pheromone update method on the ant feeding path to get a better distribution distance.The characteristic factors that affect the usage of shared bicycle were input into the XGBoost model for prediction.According to the transfer probability,the next rental point to be visited was selected through roulette,and the method based on the initial pheromone attenuation was used to update,so as to get a better solution quickly.Through the verification of 14 rental points in a school,based on the attenuation of the initial pheromone,changing the concentration of the initial pheromone can get a shorter distribution distance in a short time,which is about 1%shorter than the distance of the basic ant colony algorithm.The experimental results show the effectiveness of the algorithm.
作者
吴会丛
王敬
Wu Huicong;Wang Jing(College of Information Science and Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,Hebei,China)
出处
《计算机应用与软件》
北大核心
2020年第9期35-41,55,共8页
Computer Applications and Software
关键词
共享单车
蚁群算法
信息素浓度
更新信息素方法
最短距离
Shared bicycle
Ant colony algorithm
Pheromone concentration
Update pheromone method
Shortest path