一种基于冷链低碳物流路径的混合优化算法-《计算机技术与发展》

文章信息/Info

Title:: A Hybrid Optimization Algorithm Based on Low-carbon Cold Chain Logistic Route

Author(s):: SHAO Ke-nan; LYU Cheng-yao; ZHANG Shuai-shuai; GONG Jing; School of Science,Nanjing University of Posts and Telecommunications,Nanjing 210023,China

Keywords:: cold chain logistics; hard time window vehicle scheduling; combinatorial optimization; simulated annealing algorithm; genetic algorithm

摘要:: 冷链物流运输路径优化问题是国内外学者近几年研究的一个热点问题。首先分析了冷链物流运输的特点和需求, 并且提出了考虑综合代价、带硬时间窗和容量约束的单物流中心车辆路径优化模型,其中综合代价包括固定代价、车辆运输代价、货损代价、制冷代价、时间惩罚代价和碳排放代价。车辆路径规划问题是一个 NP 问题, 需要使用智能优化算法对解空间进行搜索。该文对遗传算法和模拟退火算法这两种经典的优化算法进行结合和参数优化,提出了一种模拟退火-遗传混合算法(GA-SA)。该算法在搜索范围和搜索精度上有较好的性能,能够在同数量级运算量的情况下求出更优的解。对上述模型进行参数设置形成仿真实验并使用混合算法和以上两种经典算法对仿真实验进行求解,通过求解结果的对比,验证了混合算法的有效性。

Abstract:: The optimization of cold chain logistics transportation route has been a hot topic in recent years. We firstly analyze the characteristics and requirements of cold chain logistics transportation and propose an optimization model of vehicle path in a single logistics center that takes into account the comprehensive cost,hard time window and capacity constraints. The comprehensive cost includes the fixed cost,vehicle transpor-tation cost,cargo damage cost,refrigeration cost,time penalty cost and carbon emission cost. Vehicle path plan-ning is a NP problem,which requires intelligent optimization algorithm to search the solution space. Two classical optimization algori-thms, genetic algorithm and simulated annealing algorithm,are combined and optimized in parameters,and a simulated anneal-genetic hybrid algorithm (GA-SA) is proposed,which has better perfor-mance in search range and search precision,and can get better solution under the same order of magnitude of computation. The parameters of the above model are set to form a simulation experiment which is solved by the hybrid algorithm and the above two classical algorithms, and the effectiveness of the hybrid algorithm is verified by comparing the solution results.