Abstract: To solve the problem of the difficulty selecting many weighting methods in multiple attribute decision, the correlation between subjective and objective weighted vector conflict factor and the assessment result accuracy is analyzed, and then, an adaptive selection algorithm based on conflict threshold is proposed. As verified by the simulation, the results show that the algorithm eliminates the effect on subjective and objective weighted vector conflict to the evaluation result, improves the calculation accuracy, and realizes the adaptive selection based on conflict threshold. Moreover, when the conflict factor is less than the threshold, adaptive selection algorithm will choose time complexity been lower than integrated weighting method of the multiplicative synthesis principle. On the contrary, when the conflict factor is greater than the threshold, it will choose the combination weighted method of high calculation result accuracy based on game theory.

Key words: conflict factor, weighted vector, adaptive, threshold

摘要： 为解决多属性决策中众多赋权方法难以选择的问题，对主客观权值向量冲突因子与评估结果精度的相关性进行了分析，进而提出了一种基于冲突阈值的自适应选择算法。通过实验仿真，结果表明：该算法消除了主、客观权值向量冲突对评估结果的影响，提高了计算精度，实现了基于冲突阈值的自适应选择。而且，当冲突因子小于阈值时，自适应选择算法将选择算法时间复杂度低于乘法合成原理的综合赋权法。当冲突因子大于阈值时，选择计算结果准确度高的基于博弈论的综合赋权法。

关键词: 冲突因子, 权值向量, 自适应, 阈值