一种组合赋权的两阶段优化算法

摘要

To solve the problem of combination weighting with more alternatives , the paper supposes an algorithm of two phase optimization .In the first phase , the paper selects some alternatives through sided t-test based on the order of every alternative in each attribute , which will optimize samples of combination weighting .In the second phase , a combination weighting model is given which combines two goals: minimizing the deviation between combination weight and subjective weight , and between combination weight and objective weight;maximizing the deviation between alternatives .Finally, an example shows the process of the method .%针对含有较多决策方案的组合赋权问题，提出了一种两阶段优化算法。第一阶段根据各个属性下的方案排序，采用单边t检验筛选方案，优化组合赋权的样本；第二阶段构建与主客观权重的偏差最小以及各方案之间的离差最大的组合赋权模型；最后运用算例说明方法的有效性。