摘要
如何将农产品新鲜、快速、低成本地有效配送,一直是配送中心选址需要关注的核心问题。结合易腐农产品特点,将总配送成本分解为运输成本及产品腐败成本两部分,运用G1法和熵值法,构建了包括各需求点的人口、社会、经济及建造成本4个因素在内的以总配送成本最小为目标的选址模型。借鉴经典NP问题中的Steiner点问题,将之前构建的以配送成本最小为目标的选址模型转化成配送中心到各个需求点总距离之和最小的问题。由于此问题属于典型的NP问题,所以文中选用模拟植物生长算法对此问题进行求解,为了使初始生长点更加合理的分布,引入了谢尔宾斯基地毯的原理对算法进行改进,有效地提高了算法的效率。最后以江苏宜兴市为例进行了实证分析,为相关部门决策提供借鉴。
How to deliver products quickly and economically is crucial for exploring a site for distribution center. Combined with perishable agricultural product characteristics, this paper establishes a location model by minimizing the total cost. The total distribution costs include transportation costs and product cost of corruption. This paper constructs an index system including four factors：population,social factors,economy and construction costs. Inspired by the classical NP problem in the Steiner point problem,this paper transformes minimizing total cost into minimizing the distance between the distribution center to the various demand points. Then this paper uses an improved plant growth simulation（PGS） algorithm to settle this model. This paper uses the Sierpifiski carpet algorithm to obtain the initial plan, which can not only find the optimum of the location but also reduce the number of iterations in the algorithm, used the PGS algorithm to find the final plan based on the initial one, andfind out the global optimal solution, taking Yixing city as an example.
出处
《江南大学学报（自然科学版）》
CAS
2013年第6期732-738,共7页
Joural of Jiangnan University (Natural Science Edition)
基金
浙江省高校人文社科重点研究基地支撑子项目(RWSKZD04-2012ZB2)
关键词
易腐农产品配送中心
选址
G1法
熵值法
模拟植物生长算法
perishable agricultural products ~ distribution center,location,G1 method,entropy method,plant growthsimulation algorithm
