灰色层次分析法及其应用

本文根据层次分析法并借助灰色理论建立了灰色层次分析法，导出了灰矩阵最大特征值的一些性质，介绍了灰正互反阵的最大特征值与对应特征向量的两种数值解法．     

Fuzzy Application to the Analytic Hierarchy Process for Robot Selection

The problem of Robot Selection is of great relevance in the present times of automation. Traditionally such problems were addressed using conventional techniques of Multi Criteria Decision Making such as The Analytic Hierarchy Process (AHP) and The Multi Attribute Utility Theory (MAUT). This paper proposes a methodology for solving common Robot Selection problems using a modification of the conventional AHP by incorporating ‘Fuzzy Linguistic Variables’ in place of numbers. The methodology encapsulates creation of Fuzzy Interface for conversion of input and output variables into suitable linguistic variables. Further, employing the fuzzification process by assigning the linguistic variables to numerical values of the membership functions and formulating suitable decision rules, the procedure culminates into the defuzzification process for converting fuzzy output into crisp value and obtaining the result in the form of Fuzzy Score. The proposed model is explained using a numerical example. The paper also presents a validation of the proposed methodology over real world problems and provides directions for future research towards the end.Additional support was provided by the National Science Foundation.     

基于层次分析和不确定决策的应急调度路线决策模型

应急物流最显著的特点表现在时间的紧迫性,决策者应在较短的时间内完成调度方案.针对出救点到应急点有多条道路可供选择的情况,由专家根据当前的报导信息和个人的经验对各条道路的通行情况进行两两比较,构造出区间数判断矩阵.根据区间数判断矩阵计算区间数权重,构造可能度矩阵,进一步计算各条道路的权重向量,从而确定出救路线.数值算例表明了算法的有效性和实用性.     

The Analytic Hierarchy Process in an uncertain environment: A simulation approach

Traditionally, decision makers were forced to converge ambiguous judgments to a single point estimate in order to describe a pairwise relationship between alternatives relative to some criterion for use in the Analytic Hierarchy Process (AHP). Since many circumstances exist which make such a convergence difficult, confidence in the outcome of an ensuing AHP synthesis may be reduced. Likewise, when a group of decision makers cannot arrive at a consensus regarding a judgment, some members of the group may simply lose confidence in the overall synthesis if they lack faith in some of the judgments. The AHP utilizes point estimates in order to derive the relative weights of criteria, sub-criteria, and alternatives which govern a decision problem. However, when point estimates are difficult to determine, distributions describing feasible judgments may be more appropriate. Using simulation, we will demonstrate that levels of confidence can be developed, expected weights can be calculated and expected ranks can be determined. It will also be shown that the simulation approach is far more revealing than traditional sensitivity analysis.     

企业综合评价的层次分析法

本文根据层次分析法的基本原理，在企业综合评价中建立各元素的结构模型，确定权重分布。为企业综合评价提供一套科学、合理、有效的指标权重体系。     

Haulier Selection—An Application of the Analytic Hierarchy Process

A project to reduce the despatch costs of a manufacturer is described. This involved the construction of a model that would enable the lowest-cost haulier to be selected and answer the question of which of that haulier's depots should be used to handle the despatches. The existing information system did not generate the necessary data for the regular use of a model based on actual despatches. Furthermore, the pattern of demand was fairly volatile, so that the least-cost haulier might have been expected to change quite often. Therefore, the use of the analytic hierarchy process to derive the necessary data was explored. Attempts were made to compare the results of a complete census of traffic carried out in the initial stage of the project with the data obtained from the analytic hierarchy process.     

基于层次分析的主成分分析法及其应用

在分析了传统的层次分析法和主成分分析法的优缺点的基础上,扬长避短的将两种方法结合在一起,讨论了基于层次分析的主成分分析法.与新兴的决策评估方法相比,例如直觉模糊层次分析法,基于层次分析的主成分分析法更具有简洁性和直观性.最后,将方法应用于登陆作战的目标选择,得到了与实战结果相同的结论,验证了方法的可行性.     

提高大学生综合素质的层次分析

利用层次分析的方法,有力地说明了在提高大学生综合素质诸因素当中,基础知识尤为重要,从而为教学改革的决策提供了理论依据。     

层次分析法中基于SVDD群体信息提取方法

针对层次分析法中群决策问题提出一种基于支持向量域描述(SVDD)的集结方法.首先利用生成树的方法把判断矩阵进行一致性剖分;然后利用支持向量域描述的方法排除干扰信息,找出群体公共信息,引入群体相容性,最优解等概念,提出并证明了关于群体信息球的特性;根据最大特征值法把群体信息球中的向量合成为群决策最优解,即关于方案的排序向量.并通过一个具体示例给出该方法的算法步骤同时显示了该方法的可行性和有效性.     

模糊德尔菲层次分析法及其应用

将传统的层次分析法、模糊评价方法与群体决策的德尔菲方法有机地结合起来,给出了一种系统分析的新方法—模糊德尔菲层次分析法(FDAHP),该方法在群体的德尔菲调查的基础上,将群体对事物的两两判断进行模糊处理,并将群体整体判断的结果作为决策环境参数,进一步根据群体决策的乐观系数最终确定出群体对事物的综合权重.用一个简单例示说明了该方法的具体执行步骤及其可行性.最后对该方法的优缺点作了简要讨论,并提出了今后相关问题的可能研究课题.     

模糊层次分析法在资源外包对象选择中的应用

资源外包是现代企业的一种重要经营战略.本文利用模糊层次分析法对资源外包对象的选择进行了定性和定量相结合的研究,以期为企业有效开展资源外包实践提供相关的理论上的指导.     

Some Comments on the Analytic Hierarchy Process

Some comments are made on the analytic hierarchy process of Thomas L. Saaty, which has gained widespread acceptance as a valuable tool for multicriteria decision-making. Saaty's validation of the method against physical laws is criticized, a multiplicative scale is suggested for making judgements, and the problem of rank reversal is discussed with reference to two published papers.     

用层次分析法对学生进行综合评价

学生的综合评价工作是学校的重要工作内容.如何从思想、学习、生活等各个方面对其进行综合考察,并使之科学化、合理化,是各个院校学生工作一直探索的问题.采用能给定量描述的决策问题带来极大方便的层次分析法(Analytic Hierarchy Process),把学生评价体系归结为5个准则27个指标,通过计算各级判断矩阵,得到了各个指标的权重,并把最终的计算结果应用到学生评优评三好等多项实际工作中.     

层次分析法一致性指标的临界值研究

层次分析法中一致性检验的显著性水平(即判断矩阵偏离一致性的程度)问题在已有研究文献中尚未得到很好的解决。论文在现有一致性指标CI基础上,从随机过程的视角,采用蒙特卡罗模拟法构造一致性指标的随机样本,根据样本确定了显著性水平为0.05和0.01情形下对应的一致性指标的临界值,为评价和调整判断矩阵提供了理论标准。同时,本研究能够确定判断矩阵偏离一致性的程度,并给出决策制定的显著性水平。     

The Analytic Polynomial-Time Hierarchy

Motivated by results on interactive proof systems we investigate an ?-?hierarchy over P using word quantifiers as well as two types of set quantifiers. This hierarchy, which extends the (arithmetic) polynomial-time hierarchy, is called the analytic polynomial-time hierarchy. It is shown that every class of this hierarchy coincides with one of the following Classes: ∑, Π (k?0), PSPACE, ∑ or Π (k?1). This improves previous results by Orponen [6] and allows interesting comparisons with the above mentioned results on inter-active proof systems.     

积因子求权重层次分析法在城市排序中的应用

克服传统层次分析法确定因素权重主观性太强的不足,结合积因子方法给出了积因子赋值求权重层次分析法,并利用该方法对黑龙江省各城市按可持续发展指标进行排序,得到符合实际的结果.     

On Sensitivity Analysis in the Analytic Hierarchy Process

The analytic hierarchy process (AHP) is a popular multiobjectivetool of decision analysis. We describe several ways of enhancingthis decision-making process through the use of sensitivityanalysis, an extension to AHP which is relatively unstudied.Sensitivity analysis can be useful in eliminating alternatives,enhancing a group decision process, or in providing informationas to the robustness of a decision. Concentrating at the firstlevel of the decision hierarchy, we create a weight space whichrepresents all possible combinations of weights for the first-levelobjectives. This weight space is then partitioned into subsets,and spatial information is generated from it. We use a smallexample to demonstrate our ideas.     

基于改进层次分析法的模糊优选模型

给出一种逼近理想解的层次分析法,并将其应用于模糊优选模型,基于该方法给出了中国16个城市的综合实力评价实例,证实了模型的可靠性和实用性.     

Fuzzy logic modifications of the Analytic Hierarchy Process

The Analytic Hierarchy Process (AHP) is a measurement methodology based on pair-wise comparisons that relies on judgment to derive priority scales. During its implementation, one constructs hierarchies, then makes judgments or performs measurements on pairs of elements with respect to a criterion to derive preference scales, which are then synthesized throughout the structure to select the preferred alternative.One of the areas where the AHP finds application is in the subjective phases of risk assessment (RA), where it is used to structure and prioritize diverse risk factors, including the judgments of experts. Since fuzzy logic (FL) has been shown to be an effective tool for accommodating human experts and their communication of linguistic variables, there has been research aimed at modeling the fuzziness in the AHP (FAHP), and recently the focus of some of that modeling has been with respect to RA.The literature discusses more than one FAHP model, which raises the question as to which are the prominent models and what are their characteristics. In response to this question, we examine three of the most influential FAHP models. The article proceeds as follows. It begins with a brief overview of the AHP and its limitations when confronted with a fuzzy environment. This is followed by a discussion of FL modifications of the AHP. A RA-based likelihood score example is used throughout. The article ends with a commentary on the findings.