Selection of discrete multiple criteria decision making methods in the presence of risk and uncertainty
Selection of discrete multiple criteria decision making methods in the presence of risk and uncertainty
复制标题
DOI：
10.1016/j.orp.2018.10.003
复制
发表时间：
2018-01-01
影响因子：
2.5
通讯作者：
Sanders, David
中科院分区：
文献类型：
--
作者：
Haddad, Malik;Sanders, David
本文提出了一种新的方法来推荐最合适的多准则决策（MCDM）方法的候选方法的子集时，风险和不确定性的预期。在分析多目标决策问题和方法特点的基础上，提出了一种结构化的方法。这种分析的结果为决策者提供了一组建议的候选方法。敏感性分析被应用于建议的候选方法组，以分析预期风险和不确定性时输出的稳健性。自动选择一种MCDM方法，以提供最稳健的结果。MCDM方法处理离散集的替代品。数值例子中，一些MCDM方法进行了比较，并建议通过计算改变任何两个替代品的排名所需的标准权重和性能指标的最小百分比变化。将根据改变替代品排序所需投入的最小百分比变化的最佳折衷，推荐多指标决策方法。不同的情况下，被认为是一些新的命题的基础上潜在的广义场景的MCDM问题。
This paper presents a new methodology to recommend the most suitable Multi-Criteria Decision Making (MCDM) method from a subset of candidate methods when risk and uncertainty are anticipated. A structured approach has been created based on an analysis of MCDM problems and methods characteristics. Outcomes of this analysis provide decision makers with a suggested group of candidate methods for their problem. Sensitivity analysis is applied to the suggested group of candidate methods to analyze the robustness of outputs when risk and uncertainty are anticipated. A MCDM method is automatically selected that delivers the most robust outcome. MCDM methods dealing with discrete sets of alternatives are considered. Numerical examples are presented where some MCDM methods are compared and recommended by calculating the minimum percentage change in criteria weights and performance measures required to alter the ranking of any two alternatives. A MCDM method will be recommended based on a best compromise in minimum percentage change required in inputs to alter the ranking of alternatives. Different cases are considered and some new propositions are presented based on potential generalized scenarios of MCDM problems.