基于遗传算法的潜射反舰导弹综合作战效能分析

提出了直接根据单指标相对隶属度的模糊评价矩阵，构造层次分析法中的判断矩阵用以确定各评价指标权重的方法，给出了用遗传算法检验和修正判断矩阵的一致性和计算判断矩阵各要素的权重的模糊综合评价模型，对潜射反舰导弹作战效能进行了综合评估和排序．实例表明，该方法简便通用，计算结果较为客观和稳定．     

残缺区间数互补判断矩阵可接受的判定及其排序方法

本文定义了残缺区间数互补判断矩阵、残缺区间数互补判断矩阵的有向图、加性一致性残缺区间数互补判断矩阵.对残缺区间数互补判断矩阵进行了研究,得到了残缺区间数互补判断矩阵是可接受的一个充要条件,用Matlab编程实现了判定残缺区间数互补判断矩阵是否可接受的算法.对于可接受残缺区间数互补判断矩阵,提出了对一致性和不一致性残缺区间数互补判断矩阵都适用的排序方法.最后通过实例验证了该方法的可行性和有效性.     

具有模糊偏好认知矩阵的超对策稳定性分析

针对一类具有模糊偏好认知矩阵的超对策问题，给出了具体的稳定性分析方法。基于相对模糊距离的概念和模糊有序加权平均（F0WA）算子提出了三角模糊互补判断矩阵的一种排序方法。由此方法得到局中人的结局偏好认知向量，进而给出了一阶超对策的稳定性分析过程。最后，以实际冲突为例说明这一方法。     

基于AHP和模糊综合评价方法的儿童早教机设计

目的针对产品进行创新设计及方案选定,提高产品设计过程中的合理性和科学性。方法运用AHP(层次分析法)与模糊综合评价法,形成儿童早教机的综合性评价。首先,通过层次分析法进行层次划分,构建指标评价体系;其次,建立判断矩阵计算各要素权重值,对结果进行一致性检验并排序;最后,结合模糊综合评价法对三种设计方案进行决策,解决多方案优化排序问题。结论通过运用层次分析法与模糊综合评价法,有效地为设计方案提供客观的决策依据,并对其他同类别产品设计研究提供一定的参考价值。     

基于模糊AHP和TOPSIS的企业知识吸收能力评价

建立了企业知识吸收能力的评价指标体系,并提出了基于模糊AHP和TOPSIS的评价方法。借助模糊AHP方法确定评价指标的三角模糊权重,并用重心法对模糊权重进行反模糊化;运用模糊TOPSIS方法构建模糊加权评价矩阵,计算各企业方案与理想解之间的相对贴近度,从而比较出各企业吸收能力强弱。通过实例分析验证了该方法的实用性和有效性。     

基于互补判断矩阵排序的供应商绩效评价方法

提出了一种基于互补判断矩阵排序的供应商绩效评价方法.由多位专家对供应商的绩效进行两两比较并进行评分;评分结果以互补判断矩阵的形式构造出相应的评价矩阵;利用权的最小平方法计算互补判断矩阵的排序向量;将排序向量中的分量按照大小排列,得到供应商绩效的比较结果.算例的计算过程和结果表明,该方法具有很好的可操作性.     

机械设计方案的模糊评估系统的研究

针对机械设计方案的多样性及存在的诸多模糊因素，提出了一种模糊评估方法。根据机械设计过程中所出现的模糊因素进行分类，建立相应的评价指标分级矩阵，通过相关计算权衡各评价指标的差异程度，然后再对分级矩阵进行加权，最后利用综合信息评估对所选设计方案进行总体评价。     

模糊群决策方法在物流服务商选择中的应用

针对物流服务商选择中评价指标的不确定性和模糊性,提出用三角形模糊数来描述评价指标;构造了基于三角形模糊数和群决策理论的基本算法;并利用模糊数排序公式对决策方案进行排序和择优.最后进行了实例分析.数值结果表明该方法可行且有效.     

基于模糊层次分析的跑道侵入风险评估

目的在模糊层次分析理论的基础上,本文从"人-环境-管理"四维度对造成跑道侵入的主要风险因素进行分析,运用基本排序的方法进行单排序,最终得到所有风险因素的危险度排序,从而为跑道安全运行提出相应措施。方法首先构造了1个准则层和4个指标层的三角模糊互补判断矩阵;其次运用模糊层次分析法对造成跑道侵入的主要因素进行了初步分析,克服了层次分析法中判断矩阵的一致性检验过程相当繁琐而且不易操作的缺点;最后利用所述的三角模糊数排序法对跑道侵入风险20个因素进行危险度总排序。结果本文得出7个影响跑道运行安全水平风险因素的重要程度排序,与实际运行结果状况基本相符。结论本文提出的方法能够较好地评估跑道侵入风险因素的危险程度,根据评估结果可有针对性地提出减少和预防跑道侵入风险控制措施,具有一定的实用性。     

基于偏好信息的实数型不确定交通安全测度研究

由于交通安全问题的复杂性和交通安全测度者认识的不确定性,在实数型不确定交通安全测度过程中可能会给出不同形式的偏好信息。针对互反判断矩阵、模糊互补判断矩阵、残缺互补判断矩阵和混合判断矩阵4种测度对象偏好信息,分别采用特征向量法、最小方差法、残缺互补判断矩阵的线性方程组法和混合判断矩阵的线性目标规划法计算出相应的排序向量,然后采用WAA算子合成出每个交通安全测度对象的综合测度值,并排序。算例研究证明该方法的有效性和易操作性。     

基于三角模糊数的测试性分配方法

在考虑多因素影响的测试性分配中,需要专家通过两两比较来确定诸因素的相对重要性．然而由于人们的判断具有模糊性和不确定性,故在构造比较判断矩阵时所给出的判断往往不是确定的数值．针对这一问题,本文提出基于三角模糊数的测试性分配方法．该方法用三角模糊数表示专家评判数值范围,考虑人判断的模糊性,将无弹性的硬指标转化成模糊的软指标,建立基于三角模糊数的结构模型,同时结合层次分析（APH）方法进行指标权重的确定,由得到的权重进行测试性指标的分配．最后结合实例对本方法进行应用研究,应用结果验证了此方法的可行性．     

Ranking alternatives via maximizing set and minimizing set based fuzzy MCDM approach

Abstract

This work suggests a maximizing set and minimizing set based fuzzy multiple criteria decision‐making (MCDM) model, where criteria are classified into cost and benefit criteria. The final fuzzy evaluation value of each alternative is developed based on the concept of subtracting the summation of weighted normalized benefit ratings from that of weighted normalized cost ratings. Using interval arithmetic of fuzzy numbers can develop the membership functions for the final fuzzy evaluation values. Chen's maximizing set and minimizing set is then applied to defuzzify all the final fuzzy numbers for ranking alternatives. Formulas for the membership functions and ranking procedure of the final fuzzy numbers are clearly presented. The suggested method provides an extension to the fuzzy MCDM techniques available. A numerical example demonstrates the computational process of the proposed method.     

排序决策的三角模糊数方法

指出<卓越绩效评价准则>实施中排序决策的重要意义.提出用语言变量表示作为排序基础的两两对象的比较评价,且用模糊三角数来表示相对属性判断值.提出一组对象比较时的相对属性测度、模糊测度矩阵、模糊相对属性权重向量等概念.并在对模糊相对属性权重向量清晰化的基础上最终完成对一组对象的排序.举出某公司供应商选优方案排序的示例,由3名评判者对4个供应商从供应品质量、供应品价格、供应时间、诚信程度4个方面进行排序,在两两比较基础上,应用三角模糊数方法实现选优,最终结果为u1>u2>u4>u3.     

基于区间二型模糊EDAS的多属性决策方法

针对多属性决策存在的复杂性和不确定性问题,提出基于区间二型模糊平均解距离法 (evaluation based on distance from average solution,EDAS) 的多属性决策方法。采用区间二型模糊集合 (interval type-2 fuzzy sets,IT2FS) 表达评价信息解决专家的偏好信息存在个体化差异问题,并纳入EDAS对备选方案进行排序。以区间二型模糊数表达评价信息构建决策矩阵,以计算得到的综合评价值的去模糊化结果作为最终的方案排序依据。针对EDAS中属性权重需要从外部获取的问题,采用区间二型模糊集合改进的最优最劣法 (best-worst method,BWM) 确定属性权重。最后,以某汽车制造企业选购新能源汽车云服务方案为例,验证所提方法的有效性。     

Group-based product scheme-screening decision method based on fuzzy AHP and evidential reasoning theory

Selecting the favourable product scheme is the first step to successful new product development (NPD). There are usually large numbers of uncertainties in product scheme evaluation and screening process of NPD due to lack of or incomplete reliable information. Considering fully the uncertainties and then conducting correct reasoning could guarantee reliability and rationality of scheme-screening results. As an extension of analytic hierarchy process (AHP), fuzzy AHP inherits multi-merits of the AHP approach and is capable of dealing with fuzzy information effectively, but it still has two weaknesses. One is the well-known ranking reversal problem. Although several researchers have analysed the reasons, we think the root cause for ranking reversal problem is due to the fact that AHP treats weights of attribute criteria and performance scores of alternatives in the same way. Therefore, we intend to deal with attribute weights and performance scores of alternatives separately and introduce evidential reasoning (ER) theory, which is good at uncertain reasoning, into fuzzy AHP to calculate the performance scores of alternatives. On the other hand, in view of the difficulty in resolution for fuzzy weights from fuzzy comparison matrix, a linear goal-programming model is proposed to calculate fuzzy weights, whose objective is to minimise the inconsistency degree of comparison matrix. By combining fuzzy AHP with ER, a group-based hybrid decision model FAHP-ER is developed. The hybrid model not only gets a great improvement in the capability of dealing with uncertainty, but also reflects the most real decision scenario and thinking process of the decision maker. Finally, a case study for schemes screening of the rotor and bearing system in the turbine generator is presented to demonstrate the application of the hybrid decision method.     

Improved fuzzy ranking procedure for decision making in product design

In this paper, we present a method for ranking any number of normal fuzzy numbers using trapezoidal fuzzy numbers as a general form, where rectangular and triangular fuzzy numbers are particular cases of such a form. This general form is supported by 29 cases, which is enough to consider all the possible situations between two normal fuzzy numbers, such as trapezoidal, triangular, or rectangular. The ranking procedure is performed using four ordering criteria into a pseudo-order preference model considering the type of the fuzzy preference relation. Two examples are given to illustrate and validate the applicability and practicality of this fuzzy ranking method. A comparison and an analysis of the proposed method is presented to demonstrate its usefulness and its contribution to the improvement of the decision making processes as a result of its management of vague or imprecise information, and whether or not that information should be allowed to be entered into such processes.     

网页的广告容忍度模糊综合评价

长期以来难以定量比较及监测网页上广告的用户容忍度,为改变该现状,提出一种结合层次分析和模糊评价方法的量化综合评价模型。在文献和实证研究厘清网页上广告元素的基础上构建评价指标体系。该体系包括8个一级指标和64个二级指标;利用层次分析法,在问卷调查和访谈得到指标相对重要性信息的基础上,通过专家评估获得评价指标的权重;在转化指标值的基础上进行多级模糊评价,通过权向量和模糊关系矩阵的合成运算求得网页的广告容忍度。为验证评价模型的实用性,应用于四大门户网站首页为实例,结果表明该模型可以克服广告元素众多、可变的属性进行量化度量,对于广告主选择投放平台及运营商监控网页的广告容忍度具有实际价值。     

Employment Quality Evaluation Model Based on Hybrid Intelligent Algorithm

In order to solve the defect of large error in current employment quality evaluation, an employment quality evaluation model based on grey correlation degree method and fuzzy C-means (FCM) is proposed. Firstly, it analyzes the related research work of employment quality evaluation, establishes the employment quality evaluation index system, collects the index data, and normalizes the index data; Then, the weight value of employment quality evaluation index is determined by Grey relational analysis method, and some unimportant indexes are removed; Finally, the employment quality evaluation model is established by using fuzzy cluster analysis algorithm, and compared with other employment quality evaluation models. The test results show that the employment quality evaluation accuracy of the design model exceeds 93%, the employment quality evaluation error can meet the requirements of practical application, and the employment quality evaluation effect is much better than the comparison model. The comparison test verifies the superiority of the model.