群决策中两类三端点区间数判断矩阵的集结方法

研究群决策过程中三端点区间数互反判断矩阵和三端点区间数互补判断矩阵的集结. 采用OWA(Ordered weighted averaging) 方法将决策者的偏好信息集结为两个三端点区间数判断矩阵. 基于三端点区间数判断矩阵的完全一致性概念, 建立三端点区间数判断矩阵的权重求解模型. 根据群决策背景下专家群最大一致的目标, 建立求解专家群体偏好权重的模型. 在第二阶段建立群偏好权重分布范围估计模型, 最后通过可能度方法以排定各方案的最终优劣顺序.     

基于证据理论/层次分析法的贝叶斯网络建模方法

针对依据专家知识推断贝叶斯网络中条件概率表(CPT)时存在的个体推断信息缺乏完备性和精确性以及整体集成结果缺乏科学性的问题,提出了基于证据理论/层次分析法(DS/AHP)的能够从专家推断信息中提取最优条件概率的方法.首先,通过引入DS/AHP方法中的知识矩阵提出了有利于实现判断对象更直观、判断方式更完善的推断信息提取机制;其次,在此基础上遵循由前至后的推断顺序提出了贝叶斯网络的构建过程;最后,应用传统方法与提出方法对同一贝叶斯网络中的缺失条件概率表进行了推断.数值对比分析表明,所提方法能够在提高计算效率的同时将累计总偏差降低41%,验证了所提方法的科学有效性和应用可行性.     

混合AHP法在ERP系统供应商评价模型中的应用

提出了一种基于混合AHP的供应商评价模型。利用标准AHP确定评价体系中各指标权重,使用基于区间AHP方法对各方案进行评定,求得评价区间值,定义了适用的区间排序法则,对评价区间值进行排序来选择供应商。该方案解决了标准AHP方法利用数值构造判断矩阵时存在的判断不确定性问题,在实际的ERP系统中得到了应用和实现。     

诱导连续区间有序加权平均算子及其在区间数群决策中的应用

基于诱导有序加权平均(IOWA)算子和连续区间有序加权平均(C-OWA)算子,提出一种诱导连续区间有序加权平均(IC-OWA)算子,并讨论了该算子的优良性质.针对区间数互补判断矩阵提出了连续偏好矩阵的概念,定义了基于专家评判水平偏差的诱导连续区间有序加权平均(DIC-OWA)算子,并给出一种基于该算子的区间数群决策方法.最后通过算例说明了该方法的可行性.     

基于一致性指标的两类不确定偏好信息集结

研究了两类区间数判断矩阵偏好信息的集结问题.首先,基于Saaty提出的数字互反判断矩阵一致性检验指标(CR),给出了区间数互反判断矩阵的满意一致性条件;然后利用互反与互补判断矩阵之间的关系求解出数字互补判断矩阵的一致性指标(CGCI),并在此基础上给出了区间数互补判断矩阵的满意一致性条件;最后建立了一致性指标最大的两类区间数判断矩阵偏好信息的集结模型,并用此模型解决了供应链中伙伴企业的选择问题.     

基于概率可信度的区间数排序方法

针对多属性决策理论的研究需要,提出一种基于概率可信度的区间数排序方法,给出了区间数排序问题的形式化描述.为明确排序规则,定义了区间数的二元序关系.采用概率密度函数表征区间数的特征信息,并构建出区间数序关系的概率可信度模型.通过建立互补判断矩阵,将区间数排序问题转换为矩阵运算过程,实现了不同类型区间数的比较.实验结果表明,所提出方法能够有效克服传统排序方法在适用范围方面的局限性,具有排序速度快、求解质量高等优点.     

基于互补判断矩阵的DS/AHP方法

为克服传统DS/AHP方法中因度量标度选择问题和知识矩阵求解问题而带来的推断信息精确性差和集成结果有效性有待商榷的缺陷,结合传统方法的建模思路分析指出其中存在的问题,通过引入互补判断矩阵提取决策者对方案组与识别框架之间的相对偏好推断,在此基础上提出由残缺互补知识矩阵向BPA函数转换的定理,并结合证据折扣思想和证据组合规则构建了开展推理判断的改进方法步骤,应用一个算例说明了提出方法的具体操作过程。     

基于证据距离的改进DS/AHP 多属性群决策方法

证据推理/层次分析(DS/AHP)方法采用了AHP法的层次结构模型和证据理论的分析过程,为不确定多属性决策问题的解决提供了新思路,但其在构造知识矩阵中用0代表残缺信息是不合理的.鉴于此,对DS/AHP方法进行了改进,并将改进后的方法拓展到群决策中,研究了专家群体权向量的确定方法.具体地,引入证据距离的概念,通过计算专家证据的综合距离来对专家赋权,体现了群决策中的多数人规则.     

基于区间判断矩阵的层次化风险评估方法

传统的AHP方法由于判断值受人为影响因素太大,导致在衡量多因素权重时常常出现结果不一致,影响了结论的准确性和评估结果的可信任性。建立了基于“网络－主机－服务－评估因子”的层次风险评估模型,利用层次化的计算模型评估网络的风险等级。从给出的区间判断矩阵入手,将区间判断矩阵一致逼近到一般的数字判断矩阵,提出一种自动修正判断矩阵的层次分析法,得到各层元素的近似权重。通过实例验证,表明该方法能精确地、自动化量化实时风险势态状况。     

基于近似随进占优的概率区间数信息多属性决策方法

针对传统区间数包含信息量太少,导致决策结果具有很大不确定性的问题,提出了基于VIKOR的概率区间数多属性决策方法。该方法首先将概率信息引入区间数,定义概率区间数的运算规则和集结算子。接着基于近似随机占优理念,定义了一种新的主客观结合的占优度,在此基础上确立了概率区间数的比较规则。最后将专家给出的概率区间数决策矩阵转化为占优度矩阵,并利用 VIKOR 方法对其求解。算例分析表明,该方法对概率分布的缓急变化具有很好的稳定性,而对于概率分布的偏态趋势变化具有敏感性。得到的决策结果在与传统方法一致的基础上更为准确。     

A Model for Decision Making with Missing,Imprecise, and Uncertain Evaluations of Multiple Criteria

In real‐life multicriteria decision making (MCDM) problems, the evaluations against some criteria are often missing, inaccurate, and even uncertain, but the existing theories and models cannot handle such evaluations well. To address the issue, this paper extends the Dempster–Shafer (DS)/analytic hierarchy process (DS/AHP) approach of MCDM to handle three types of ambiguous evaluations: missing, interval‐valued, and ambiguous lottery evaluations. In our extension, the aggregation of criteria's evaluation takes the following six steps: (i) calculate the expected evaluation interval and the ambiguity degree of each group of decision alternatives regarding each criterion, (ii) from them to obtain the preference degree of each group of decision alternatives, (iii) apply the DS/AHP method to obtain the mass function distribution of each group of decision alternatives, (iv) use the Dempster's rule of combination to get the overall mass function of each group of decision alternatives with respect to all criteria, (v) according to the overall mass function to count the belief function and the plausibility function of each decision alternative, and (vi) set the overall preference ordering of decision alternatives by our regret‐avoid ambiguous principle and then find the optimal solution. Finally, we give an example of real estate investment to illustrate how our approach is employed to deal with real‐life MCDM problems.     

An expert system for multi-criteria decision making using Dempster Shafer theory

This paper outlines a new software system we have developed that utilises the newly developed method (DS/AHP) which combines aspects of the Analytic Hierarchy Process (AHP) with Dempster–Shafer Theory for the purpose of multi-criteria decision making (MCDM). The method allows a decision maker considerably greater level of control (compared with conventional AHP methods) on the judgements made in identifying levels of favouritism towards groups of decision alternatives. More specifically, the DS/AHP analysis allows for additional analysis, including levels of uncertainty and conflict in the decisions made, for example. In this paper an expert system is introduced which enables the application of DS/AHP to MCDM. The expert system illustrates further the usability of DS/AHP, also including new aspects of analysis and representation offered through using this method. The principal application used to illustrate this expert system is that of identifying those residential properties to visit (view), from those advertised for ales through a real estate brokerage firm.     

三值命题逻辑系统中理论的随机发散度的分布

在三值命题逻辑系统的随机逻辑度量空间[(F(S),ρp)]中,指出理论的[p]-随机发散度是和随机三值分布序列[p=(p1,p2,…)]的具体取值密切相关的,证明了全体原子公式之集[S]的[p]-随机发散度随着三值分布序列[p]的不同取值可以充满整个[(0,1]]实数区间。     

中小型制造企业供应伙伴选择的UVAHP群决策法

针对中小型制造企业零部件供应伙伴的选择问题,在分析AHP(层次分析法)优、缺点的基础上,提出了一种为克服AHP的缺点而采用不确定投票表决的方法来构造判断矩阵,且统一处理伙伴选择的各评价指标下的区间数值型属性值与不确定语言型属性值、增益指标属性值与损益指标属性值的UVAHP协作伙伴选择方法.给出了该方法的基本原理和基本步骤,建立了其相应的数学模型,并通过实例验证了其正确性.该方法对于中小型制造企业零部件供应商选择具有较高的参考价值.     

A method of aggregation in DS/AHP for group decision-making with the non-equivalent importance of individuals in the group

Group decision-making (GDM) is an active area of research within multi-criteria decision-making (MCDM). This paper develops the nascent DS/AHP method of MCDM as an effective tool in GDM. Attention here is given to the aggregation of evidence from individual members of a decision-making group who are considered non-equivalent in their importance within the group. A discount rate value is defined for each member of the group depending on their perceived individual levels of importance. This discount rate attenuates the evidence from an individual by re-assigning more value to their concomitant level of ignorance. The adjusted evidence from each group member is then combined to derive the group's collective decision. A small real world GDM problem concerning the choice of university course textbooks is utilised throughout the analysis presented.     

Interval Evaluations in the Analytic Hierarchy Process By Possibility Analysis

Since a pairwise comparison matrix in the Analytic Hierarchy Process (AHP) is based on human intuition, the given matrix will always include inconsistent elements violating the transitivity property. We propose the Interval AHP by which interval weights can be obtained. The widths of the estimated interval weights represent inconsistency in judging data. Since interval weights can be obtained from inconsistent data, the proposed Interval AHP is more appropriate to human judgment. Assuming crisp values in a pairwise comparison matrix, the interval comparisons including the given crisp comparisons can be obtained by applying the Linear Programming (LP) approach. Using an interval preference relation, the Interval AHP for crisp data can be extended to an approach for interval data allowing to express the uncertainty of human judgment in pairwise comparisons.