摘要
针对由单个制造商、单一产品和多个客户构成的供应链系统,建立了分散控制下系统利润最大化模型,提出了新的客户选择可变方案,分别设计了遗传算法和分枝定界法对问题进行了求解。通过实例仿真与前人提出的客户选择不可变方案进行了比较分析,结果证明,分枝定界法更适合求解规模较小的问题,而遗传算法可以通过调整种群规模和遗传算子来解决规模较大的问题;与客户选择不可变相比,当客户选择可变时,系统能获取较大的期望利润。
The batch plan problem with random demand is considered for a supply chain system consisting of one manufacturer and multi-retailers. The manufacturer produce single product. A model is built based on decentralized control, so that the profit of system is maximized. New eliant choice variable plan is proposed.A genetic algorithm and a branch-and-bound algorithm are used to solve the problem,Numerical examples show that the branch-and-bound method can be only used for solving problem of smaller size, while genetic algorithm can be used for solving problems of larger size. Comparing with client choice invariable plan, the system can obtain bigger expected profit by client choice variable plan.
出处
《控制工程》
CSCD
2007年第4期434-437,共4页
Control Engineering of China
基金
国家自然基金资助项目(10671001)
关键词
分散控制
遗传算法
分枝定界法
distributed control
genetic algorithm
branch-and-bound method
