生鲜农产品分拣车与冷藏车的协同作业优化研究

Cooperative Operation Optimization of Sorting and Refrigerated Trucks forFresh Agricultural Products

聚焦生鲜农产品“最先一公里”冷链运输环节中的分拣车和冷藏车协同作业优化难题,本文采用时空网络建模方法和拉格朗日松弛算法等,以多维复杂组合优化问题的降维分解为主要切入点,重点研究基于时空状态网络构造移动分拣车和冷藏车协同作业优化模型;设计基于拉格朗日松弛算法和贪心规则的求解方法,实现大规模高维网络的降维分解,提升算法效率;通过数值实验与CPLEX商用求解器进行对比,验证本文算法在求解较大规模问题时的性能,并通过敏感性分析研究了单位行驶成本对系统运行状态的影响。本文采用时空网络建模方法,为降低多阶段协同作业优化建模的复杂度提供新思路,为高效求解大规模多维网络优化问题提供新方法。

The paper focuses on the cooperative operation optimization of sorting and refrigerated trucks in the“first-mile”cold chain transportation for fresh agricultural products.It employs methods such as space-time-state network modeling and lagrangian relaxation algorithm to address the challenges.The primary focus is on decomposing multidimensional complex combinatorial optimization problems,particularly emphasizing the construction of an optimization model for the cooperative operation optimization of moving sorting and refrigerated trucks based on space-time-state network.Additionally,it designs a solution method based on the lagrangian relaxation algorithm and greedy rules to achieve dimension reduction decomposition of large-scale high-dimensional networks,thereby enhancing algorithm efficiency.Comparative analyses with the CPLEX commercial solver through numerical experiments validate the algorithm’s performance in solving large-scale problems.Sensitivity analysis is conducted to study the impact of unit travel costs on system operation states.By employing space-time-state network modeling,the study provides new insights into reducing the complexity of multi-stage cooperative operation optimization modeling and offers novel methods for efficiently solving large-scale multidimensional network optimization problems.