机会可控的补偿随机规划模型

On stochastic programming model with recourse and chance control

分析了已有的随机规划处理方法,针对原有方法中存在的不能兼顾补偿和约束机会控制的问题,提出了机会可控的补偿随机规划模型,在补偿随机规划模型的基础上,对允许违背的约束或者相应的补偿增加机会控制.以明确的反应决策者的态度,分析了模型的结构,设计了以遗传算法和单纯形法为基础的分解算法对模型进行求解.实际算例证明了模型和分解算法的有效性.

Uncertain problems can be transformed to deterministic ones by a variety of stochastic programming models from different points, and these models can also reflect the attitude of the decision makers to some degree. A new stochastic programming model is proposed, in which some chance control constraints are used to the constraints that could be violated or the corresponding recourse terms to control the chance and the recourse at the same time. A new decomposition algorithm based on GA and SA is also proposed to resolve the model. Finally it is demonstrated by an example that the proposed model can reflect the decision attitude more clearly, and shows the validity of the decomposition algorithm.