论文部分内容阅读
针对混合型偏好信息的集结问题,利用直觉模糊实现了5类混合型偏好信息的统一,提出利用冲突系数和Jousselme距离综合表示的证据冲突度计算专家权重的方法,最后利用Demsper组合规则对专家权重修正后的证据进行组合.实例分析表明,所提出的偏好信息统一方法能够保持转化前后方案的优先顺序不变,通过专家权重对专家意见的修正使得证据合成结果能体现多数专家的一致意见,具有较高的实用价值.
In order to solve the problem of aggregate preference information aggregation, five kinds of hybrid preference information are unified by using intuitionistic fuzzy method. The method of calculating expert weight by using conflict coefficient and Jousselme distance synthetic evidence conflict degree is proposed. Finally, Demsper combination rule is used to evaluate expert weight The case analysis shows that the proposed preference information unification method can keep the priorities of programs before and after conversion unchanged, and the amendment of expert opinions through expert weights makes the result of evidence synthesis reflect the consensus of most experts. High practical value.