考虑碳排放影响的物流园区布局优化研究

Research on layout optimization of logistics parks considering impact of carbon emissions

合理的物流园区功能区布局有助于提高物流作业效率、降低物流成本.在考虑碳排放影响基础上,提出一种结合拓展环境负荷模型(Stochastic Impacts by Regression on Population,Affluence,and Technology,STIRPAT)与系统布局规划理论(Systematic Layout Planning,SLP)的物流园区布局优化方法,针对物流园区的功能布局需求,采用Kaya扩展恒等式与对数平均迪氏指数模型(Log Arithmic Mean Divisia Index,LMDI)进行因素选择及分解,通过构建STIRPAT碳排放模型,对员工规模、产业结构、能源结构及能源强度等因素进行回归分析,探讨物流园区碳排放的影响机理.然后,以改进SLP法为依据,采用灰色预测模型划分货运需求状态并基于马尔科夫链进行预测结果修正,以货运搬运成本最小、综合相关性最大、碳排放量最低为优化目标,采用多种群遗传算法进行布局优化方案求解.实例分析结果表明:物流园区布局优化时,非物流关系及综合相关性分析中需要考虑与园区运营及运输组织相关的产业结构和能源强度因素,与非支配排序遗传算法(Non-dominated Sorting Genetic Algorithm-Ⅱ,NGGA-Ⅱ)求解的对比分析结果显示,采用STIRPAT-SLP方法所得到的布局方案更符合物流作业的流程关系,在求解收敛速度、优化目标的兼顾性以及实际布局的合理性等方面具有相对优越性,更符合物流园区功能区布局的实际需求.

A well-designed layout of a logistics park can improve logistics efficiency and reduce associated costs.This study proposes an optimization method for logistics park layout,taking into account the impact of carbon emissions.The method combines the Stochastic Impacts by Regression on Population,Affluence,and Technology(STIRPAT) model with the Systematic Layout Planning(SLP)method.To select and decompose factors,the Kaya extended identity and Logarithmic Mean Divisia Index(LMDI) method are used based on the functional layout requirements of the logistics park.The STIRPAT carbon emission model is constructed,and regression analysis is carried out to explore the influencing mechanism of carbon emissions in logistics parks.Factors considered in this analysis include employee size,industrial structure,energy structure,and energy intensity.Then,based on the improved SLP method,the gray forecast model is used to divide the freight demand status,the forecast result is corrected based on the Markov chain.With the optimization objectives of minimizing freight transportation cost,maximizing comprehensive correlation,and reducing carbon emissions,a multi-population genetic algorithm is used to solve the layout optimization scheme.The case study results show that when optimizing the layout of the logistics park,the industrial structure and energy intensity factors related to park operation and transportation organization should be considered in the non-logistics relationship and comprehensive correlation analysis.Moreover,comparison and analysis with the Non-dominated Sorting Genetic Algorithm-Ⅱ(NSGA-Ⅱ) show that the layout scheme obtained by the STIRPAT-SLP method is more in line with the flow relationship of logistics operations.Additionally,it has relative superiority in terms of solution convergence speed,compatibility of optimization objectives,and rationality of actual layout.This method is more suitable for the actual needs of functional area layout of logistics parks.