论文部分内容阅读
为解决多层分销链中预测订货忽略当前库存状况和决策者参与的问题,运用Multi-agent和模糊理论构建了不确定条件下的协同预测订货模型,采用模糊数对分销链中的不确定性需求和单位成本参数进行了表述并建立了模糊规则库用于分销链的协商谈判。应用遗传算法和解模糊运算求解分销链中整体或局部优化条件下的目标函数,并给出了公司间冲突的模糊协商算法。仿真实验表明,不确定条件下的分销链预测订货量没有出现信息放大,模糊协商谈判获取了符合双方利益的预测订货成本。与确定性条件下的计算相比,运用模糊理论解决分销链中基于Multi-agent的预测订货是合理可行的。
In order to solve the problem of the current inventory status and the participation of decision makers in the multi-layer distribution chain, the multi-agent and fuzzy theory are used to construct the collaborative forecast ordering model under uncertain conditions. The fuzzy numbers are used to determine the uncertainty in the distribution chain Demand and unit cost parameters were expressed and the establishment of a fuzzy rule base for the distribution chain negotiations. The objective function under the condition of overall or partial optimization in distribution chain is solved by genetic algorithm and fuzzy operation, and the fuzzy negotiation algorithm of intercompany conflict is given. The simulation results show that there is no information amplification in the order quantity of the distribution chain under the uncertain conditions, and the fuzzy order negotiation obtains the forecast order cost in line with the interests of both parties. Compared with the deterministic calculation, it is reasonable and feasible to apply fuzzy theory to solve the multi-agent-based forecast order in the distribution chain.