基于动态自适应布谷鸟搜索算法的多目标闭环供应链网络优化

Multi-objective Closed-loop Supply Chain Network Optimization Based on Dynamic Adaptive Cuckoo Search Algorithm

针对不确定环境下的闭环供应链网络设计问题,构建以最小网络成本、碳排放量和顾客满意度损失为目标的闭环供应链网络规划模型。采用多面体不确定集描述不确定参数,建立基于多面体不确定集的多目标鲁棒优化模型,同时提出一种基于动态步长和动态发现概率的自适应布谷鸟搜索算法,并引入群搜索策略以增加种群的进化效率,结合案例企业的运营数据,分别采用动态自适应布谷鸟搜索算法和非支配排序遗传算法求解模型,验证改进型布谷鸟搜索算法的优越性。最后为验证模型的鲁棒性,将多面体鲁棒优化模型与确定模型、盒式鲁棒优化模型以及区间多面体鲁棒优化模型进行对比,验证所提模型对不确定扰动的有效抑制作用。

Aiming at the problem of closed-loop supply chain network design in uncertain environment,a closed-loop supply chain network planning model is constructed with the goal of minimum network cost,carbon emission and loss of customer satisfaction.The polyhedral uncertainty set is used to describe the uncertain parameters,and a multi-objective robust optimization model based on polyhedral uncertainty set is established.An adaptive cuckoo search algorithm based on dynamic step size and dynamic discovery probability is proposed,and the group search strategy is introduced to increase the population advancement efficiency.Combined with the operation data of the case enterprise, the dynamic adaptive cuckoo search algorithm is used respectively.The superiority of the improved cuckoo search algorithm is verified by the dynamic adaptive cuckoo search algorithm and the non-dominated sorting genetic algorithm.Finally,in order to verify the robustness of the model,the polyhedron robust optimization model is compared with the deterministic model,the box robust optimization model and the interval polyhedron robust optimization model to verify the effective inhibition of the proposed model to the uncertain disturbance.