产品设计方案择优的语言群决策方法研究

针对产品设计方案择优过程中评价信息具有的主观性和不确定性问题,以及现有产品设计方案择优存在的不足,提出了一种基于二元语义信息公理的语言型多属性产品设计方案择优方法。该方法将各专家语言评价信息转化为二元语义形式并使用二元语义信息集结算子分别得到了产品设计方案群组二元语义决策矩阵和群组二元语义指标权重向量,确定了产品设计方案指标主、客观组合权重向量;依据信息公理思想分别计算了各产品设计方案的二元语义信息量并确定最优产品设计方案。研究结果表明,该方法能够避免语言型评价信息转换与集结造成的信息损失,对产品设计方案指标进行组合赋权能够全面反映主观和客观程度,产品设计方案决策过程计算简单、准确有效,具有潜在实用价值。     

基于多粒度语言的动态联盟合作伙伴群决策

客观事物的复杂性和人类认识的局限性使得评价指标的数据很难用一个精确数表示,具有不同程度的不确定性。在比较分析现有各种表达不确定性方法的基础上,提出了基于多粒度语言的虚拟企业合作伙伴选择方法。首先将多粒度语言一致化为标准语言评价集中的语言,然后基于正态分布的方法求位置权重向量,采用扩展的二元语义组合有序加权几何平均(ET-COWGA)算子集结各专家评价值得到群体综合的各属性评价值。根据局部优化和全局优化结果得到精确的属性权重,利用加权几何平均(ET-WGA)算子集结各属性值,得到方案的综合评价值。最后通过一个实例说明了合作伙伴选择的整个决策过程,并论证了该方法在最后优化决策过程中的有效性。     

考虑专家偏好关联与心理行为的地铁车门系统失效风险评估

为了在传统失效模式及影响分析(FMEA)中考虑专家偏好信息关联与心理行为,提出基于Choquet积分与交互式多准则决策(TODIM)的FMEA风险评估方法。该方法采用区间直觉模糊数表征风险评估信息的模糊性与不确定性,利用能够刻画专家偏好信息关联的2-可加模糊测度和Choquet积分确定专家和风险因子权重,运用区间直觉模糊加权算术集结算子集结专家评价信息,应用反映专家风险评估中心理行为的TODIM方法确定失效模式占优度,以此获取失效模式风险排序。通过地铁车门系统失效风险评估案例验证了所提方法的有效性与可行性。     

考虑客户需求偏好的产品创新设计方案多属性决策评价

为了使产品创新设计方案的评价和决策过程充分考虑客户的需求偏好特征,研究了考虑客户需求偏好的产品创新设计方案多属性决策评价方法。基于对客户需求偏好的分析,提出集成Kano模型和粗糙集理论的评价指标权重确定方法,提升了指标权重确定的合理性并使客户需求偏好信息有效地反映到评价过程中。考虑到方案评价中决策信息的模糊性和不确定性,构建了基于粗数决策信息处理的产品创新设计方案多属性决策灰色关联分析模型,通过融入上述指标权重并计算各方案的粗数区间差异系数来实现产品创新设计方案的优选。以电冰箱产品创新设计方案评价决策为例,验证了所提方法的可行性和有效性。     

基于不完全互惠偏好关系的产品工业设计方案决策

为解决产品工业设计方案决策中的偏好缺失问题,引入不完全互惠偏好关系,提出了产品工业设计方案决策流程。研究了不完全互惠偏好关系的基本理论、运算法则与缺失偏好处理算法,并据此确定产品方案偏好矩阵的一致性权重,借助层次分析法获取决策者信任权重,建立共识度模型评判决策群体偏好一致性,根据不完全互惠偏好关系一致性权重和决策群体信任权重确定非共识决策的反馈机制,利用粒子群优化算法和最小成本法对产品工业设计方案决策的不完全互惠偏好矩阵进行优化,以促使决策共识达成,输出产品方案整体支配度,确定方案优劣。以汽车充电桩的产品设计方案决策为例,开发了软件原型系统,验证了所提方法能够有效处理产品工业设计方案决策中的偏好缺失问题,提升设计决策的质量和效率。     

基于语言直觉模糊集的应急物流预案决策

为解决应急物流决策问题,提出一种考虑预案实施人员诉求和定性表征预案偏好的语言区间直觉模糊群决策方法.为克服已有算子出现的集结失效现象,有效集结语言区间直觉模糊数组,提出?ukasiewicz运算法则和集结算子;以群共识最大化为目标,基于距离测度构建专家权重优化模型,进而获取可靠性强的综合评价矩阵;利用字典序关系提出语言区间直觉模糊逼近理想解法,以获取各个属性的正、负理想值,进而提出应急物流预案群决策方法.通过实例表明所提方法的合理性和科学性.     

基于决策者方案偏好的新产品开发方案选择方法

为充分体现新产品开发方案选择决策过程中决策者的主观偏好对决策结果的影响,提出基于决策者方案偏好的新产品开发方案选择方法。鉴于方案选择过程中信息的模糊性和不确定性,评价指标属性值以区间数形式给出,以减少决策信息的丢失。利用向量规范化方法规范化评估团体决策矩阵;通过企业内决策者给定其对备选方案的主观偏好值,并引入修正系数进行修正;通过区间数相离度获取使评估团队与决策者偏好总偏差最小的属性最优权重向量;集结属性权重向量和规范决策矩阵,得出各备选方案的综合属性值;通过两两比较区间数形式的综合属性值,建立可能度矩阵并计算其排序向量进行方案的排序择优。应用实例及案例对比分析表明,该方法对新产品开发方案选择是有效且可行的。     

产品规划中基于多种偏好信息的顾客需求分析方法

为了有效地评估产品规划中市场环境的模糊性,提出了基于多种偏好信息的顾客需求分析方法。首先,决策者采用偏好的信息形式评估各项顾客需求的重要度;其次,利用转换函数将多种形式的偏好信息一致化为模糊互补判断矩阵,应用行和归一化方法计算排序向量,进而集结决策成员的权重获得顾客需求的初始重要度;最后,合成初始重要度与各项需求相应的修正系数,并通过规范化处理得到顾客需求的重要度。PE颚式破碎机新产品的开发实例表明,本文中提出的方法是有效可行的。     

基于模糊物理规划与互反馈的产品工艺方案绿色性分析

针对绿色工艺决策中合理且有实际物理意义的目标权重难以确定的问题,提出基于决策者主观偏好和决策信息互反馈的产品工艺方案绿色性分析技术。通过对用户层、满意偏好捕获层、方案决策层、方案评估层与候选方案层的分解细化,实现了工艺方案绿色性分析的多层架构。运用模糊集理论对决策模糊知识进行处理,依据捕获的决策目标偏好建立反映决策者主观信息的偏好函数,搭建模糊语境下绿色满意工艺方案的物理规划决策模型。通过决策者与工艺方案间的满意信息互反馈,实现基于模糊物理规划与互反馈的工艺方案绿色满意度决策。结合某汽车零部件工艺方案的绿色性分析对所述方法进行了说明和验证。结果表明,该方法可以有效避免决策目标权重确定的局限性,提高决策满意度。     

基于多种偏好信息的工程特性优先度确定方法

为有效评估质量功能配置中信息的模糊性,提出基于多种形式偏好信息的工程特性优先度确定方法.顾客需求与工程特性关联关系、工程特性自相关关系的评估信息容易获取且精确度较高,使用效用值进行判断;通过加权算术平均算子将判断信息与评估成员权重集结,进而将群体关联度矩阵、群体自相关矩阵和顾客需求最终重要度合成,得到各个工程特性的关联重要度;市场竞争性分析有较大的不确定性,引入多种形式的偏好信息表征各项工程特性的竞争潜力;利用转换函数将多种形式的信息一致化为模糊互补判断矩阵,进而使用行和归一化方法得到工程特性的竞争潜力因子;合成关联重要度和竞争潜力因子,确定工程特性的综合优先度.通过PE颚式破碎机新产品的开发实例,验证了所提方法的有效性和可行性.     

A fuzzy hybrid group decision support system approach for the supplier evaluation process

The quality and delivery performance of a typical company depends largely on the capability of its suppliers. It is therefore important for a company to evaluate the performance of its existing suppliers. However, the problem of supplier evaluation is rather complex when a group of decision makers responsible for these suppliers are geographically dispersed. Evaluating decisions requiring the elicitation of vague and imprecise linguistic information expressed in fuzzy terms. For this purpose, a hybrid fuzzy group decision-making approach for supplier evaluation is proposed, integrating the fuzzy analytic hierarchy process (F-AHP) and the fuzzy preference-ranking-organization-method-for-enrichment-of-evaluation (PROMETHEE) group decision support system (F-PROMETHEE GDSS) method. F-AHP is used to determine the weight of multiple criteria while F-PROMETHEE GDSS is applied for the final ranking of suppliers from a group decision viewpoint. The approach advances the supplier evaluation context in two aspects. First, it treats each decision maker uniquely in terms of the preference function and threshold levels. Second, it provides rapid communication between teams through video conferencing. The hybrid approach is used to evaluate the suppliers of a dairy company.     

基于Vague集的机械设计方案多目标模糊决策方法

在现有多目标模糊决策方法的基础上，提出了一种改进的基于Vague集的多目标模糊决策算法及其在机械设计方案决策中的应用方法。通过建立一个函数来评价每个方案满足决策者需要的适合度，其中方案的特征由Vague集表示，改进的评价函数算法包含了肯定隶属度、否定隶属度和不确定度（踌躇度）等因素，并且算法中所定义的函数含义更明确。实例阐明了本文基于Vague集的多目标模糊决策方法的合理性，机械设计方案多目标模糊决策取得了较为满意的结果。     

A multicriteria intuitionistic fuzzy group decision making for plant location selection with ELECTRE method

Plant location selection, the process of finding a suitable location for plant of a project has a strong influence on the success of industrial venture with the goal to minimize cost and maximize the use of resources. Thus, plant location selection problem is a multicriteria decision-making problem involving several conflicting criteria on which decision makers' knowledge is vague and imprecise. Therefore, in this study, the elimination and choice translating reality (ELECTRE) method is proposed with intuitionistic fuzzy sets for selection of appropriate plant location in group decision-making environment to tackle uncertainty of the information provided by decision makers and a plant location selection problem is considered to illustrate the proposed intuitionistic fuzzy-ELECTRE method. The ratings of alternatives with respect to each criterion and the weights of each criterion are taken as linguistic terms further characterized by triangular intuitionistic fuzzy sets.     

Linguistic harmonic mean operators and their applications to group decision making

Harmonic mean is the reciprocal of arithmetic mean of the reciprocal, which is a conservative average to be used to provide for aggregation lying between max and min operators. In this paper, we develop some new aggregation operators such as linguistic harmonic mean (LHM) operator, linguistic weighted harmonic mean (LWHM) operator, linguistic ordered weighted harmonic mean operator, and linguistic hybrid harmonic mean (LHHM) operator, which can be utilized to aggregate preference information taking the form of linguistic variables and then study some desirable properties of the operators. Based on the LHM and the LHHM operators, we propose a practical method for group decision making with linguistic preference relations and also give an illustrative example. Furthermore, based on the LWHM and LHHM operators, we develop a multiple attribute decision making applied in the group decision making to improving advanced manufacturing technology selection process.     

A fuzzy bi-objective mixed-integer programming method for solving supply chain network design problems under ambiguous and vague conditions

Supply chain (SC) network design problems are complex problems with multi-layer levels and dynamic relationships which involve a considerable amount of uncertainty concerning customer demand, facility capacity, or lead times, among others. A large number of optimization methods (i.e., fuzzy mathematical programming, stochastic programming, and interval mathematical programming) have been proposed to cope with the uncertainties in SC network design problems. We propose a fuzzy bi-objective mixed-integer linear programming (MILP) model to enhance the material flow in dual-channel, multi-item, and multi-objective SCs with multiple echelons under both ambiguous and vague conditions, concurrently. We use a computationally efficient ranking method to resolve the ambiguity of the parameters and propose two methods for resolving the vagueness of the objective functions in the proposed fuzzy MILP model. The preferences of the decision makers (DMs) on the priority of the fuzzy goals are represented with crisp importance weights in the first method and fuzzy preference relations in the second method. The fuzzy preference relations in the second method present a unique practical application of type-II fuzzy sets. The performance of the two methods is compared using comprehensive statistical analysis. The results show the perspicuous dominance of the method which uses fuzzy preference relations (i.e., type-II fuzzy sets). We present a case study in the food industry to demonstrate the applicability of the proposed model and exhibit the efficacy of the procedures and algorithms. To the best of our knowledge, a concurrent interpretation of both ambiguous and vague uncertainties, which is applicable to many real-life problems, is novel and has not been reported in the literature.     

基于模糊多属性群决策及OWA算子的民机备件供应商评估

针对民机备件供应商的评估,从质量、资信、竞争力、响应以及服务等方面出发,构建了包含准则层与操作层的评估层次结构,其中操作层中定性属性信息以群决策、语意形式给出;准则层指标在权重信息与模糊语意量化词匹配的基础上,采用OWA算子的优化算法计算对应操作层的权重值;最后综合操作层的属性信息与权重信息,确定供应商的评估值,进而通过实例分析说明了本方法的实用性。     

Comparison of fuzzy AHP and fuzzy TOPSIS methods for facility location selection

Facility location selection is a multi-criteria decision problem and has a strategic importance for many companies. The conventional methods for facility location selection are inadequate for dealing with the imprecise or vague nature of linguistic assessment. To overcome this difficulty, fuzzy multi-criteria decision-making methods are proposed. The aim of this study is to use fuzzy analytic hierarchy process (AHP) and the fuzzy technique for order preference by similarity to ideal solution (TOPSIS) methods for the selection of facility location. The proposed methods have been applied to a facility location selection problem of a textile company in Turkey. After determining the criteria that affect the facility location decisions, fuzzy AHP and fuzzy TOPSIS methods are applied to the problem and results are presented. The similarities and differences of two methods are also discussed.     

A new fuzzy group multi-criteria decision making method with an application to the critical path selection

In this paper, we propose a new fuzzy group multi-criteria decision making method and apply it to determine the critical path in a project network. The criteria used here are time (expected duration), cost, risk, and quality of the project activities that are considered critical in project management. As each criterion has its independent level of importance in the critical path selection, the weights of the project criteria are also considered in the analysis. Considering that the information in terms of various project activities and criteria weights are often incomplete and/or uncertain in real-world situations, the essential information in terms of the criteria and project activities are obtained using triangular fuzzy numbers and/or linguistic variables that are mapped to triangular fuzzy numbers, wherever appropriate. The proposed method involves fuzzy evaluation based on the fuzzy information of the possible project paths on each criterion leading to the strength and weakness index scores of the project paths. We define a measure of criticality termed as the total performance score of each project path obtained using its strength and weakness index scores. The path that has the highest measure of criticality is selected as the critical path. A numerical illustration is provided to demonstrate working of the proposed methodology. Further, a case study from manufacturing engineering industry is also presented to better justify the applicability and potentials of the proposed methodology. A comparison with closely related fuzzy multi-criteria decision methods for the critical path selection is done to analyze the performance of the proposed methodology.     

Extension of the ELECTRE method for decision-making problems with interval weights and data

Recent research is recognizing that multicriteria decision making should take into account the concepts of uncertainty, risk, and confidence. In some cases, precise determination of the exact value of the attributes is difficult. Consequently, to surmount this potential problem, their values are considered as fuzzy and intervals. The elimination and choice translating reality (ELECTRE) is one of the most widely used methods to rank a set of alternatives versus a set of criteria to reflect the decision maker’s preference. What is customary in almost all recent papers regarding ELECTRE is the utilization of fuzzy concepts in decision-making process. In this paper, we propose a novel algorithmic ELECTRE method which assimilates the concepts of interval weights and data for the facilitation of evaluation of a set of alternatives against a set of criteria in this situation. Finally, the executive procedure of our proposed ELECTRE method is illustrated by applying it to the problem of supplier selection.