基于K-Means算法的小区垃圾分类站选址问题研究

Study on the Location of Garbage Sorting Station in Residential Area Based on K-means Algorithm

为促进垃圾分类政策实施,科学、合理地在居民生活小区设置垃圾分类站,建立选址模型和成本模型对垃圾分类站建设运营成本及居民满意度负效应成本进行求解,并对K-means聚类算法与模糊C-means聚类算法进行比较。通过对某小区每栋居民楼到垃圾分类站的平均距离分析得出,K-means聚类算法计算得出的平均距离相比模糊C-means聚类算法缩短了约17%,在成本模型中建设运营成本降低了1万元,居民满意度负效应成本降低了0.68万元,验证了模型的可行性及K-means聚类算法的优越性。在未来的研究中可对算法进行改进,以进一步优化成本,确定全局最优。

In order to promote the implementation of the garbage classification policy,scientifically set up the number of garbage classification stations in the residential quarters,we establish the location model and cost model for the construction and transportation of garbage classification stations.By comparing K-means clustering algorithm with fuzzy C-means clustering algorithm,we find that Kmeans clustering algorithm is 17%lower than that of fuzzy c-means clustering algorithm.In the cost model,the construction and operation cost is reduced by 10000 yuan,and the cost of satisfaction negative effect is reduced by 6800 yuan,which verifies the feasibility of the model and the superiority of the algorithm.It can further improve the algorithm to continue to optimize the cost and determine the global optimization.