供应商选择问题的随机相关机会规划模型

Stochastic Dependent Opportunity Programming Model for Supplier Selection Problems

针对经典的供应商选择问题,在供应商报价、延迟交货率以及废品率不确定的情况下,同时结合随机相关机会的特点,以总成本、延迟交货量、货物不合格量小于其对应目标值的可能性最大为优化目标,建立随机相关机会规划模型,并且设计一种嵌入神经网络和随机模拟的混合自适应遗传算法。并用Microsoft Visual C++编程得以验证,详细说明了算法的关键内容与步骤,结合具体事例验证模型和算法的有效性,并与简单的遗传算法进行对比以验证文章中模型和算法的可行性。

For the classic vendor selection problem,under the uncertainty of supplier quotation,delayed delivery and rejection rate,combined with the characteristics of the stochastic dependent chances,the probability that the total cost,delayed delivery amount and unqualified quantity of goods are less than their corresponding target values is the greatest as the optimization goal,a stochastic dependent chance programming model is established,and a hybrid adaptive genetic algorithm embedded with neural network and random simulation is designed.In Microsoft Visual C++programming,it can be verified that the algorithm is more advantageous than the traditional genetic algorithm.