模糊需求的联合补充问题研究

研究了需求不确定的联合补充问题.用梯形模糊数表示不确定需求,建立了问题的模糊规划数学模型,目标函数为最小化总成本,包括订货成本和库存持有成本;采用遗传算法对模型求解,给出了编码方案,并讨论了选择、交叉、变异等遗传算子.用数值实例验证了所提出的模糊数学模型及求解算法,并对确定需求模型结果进行了对比分析.     

基于模糊层次分析法的集成软件质量评估模型

通过建立软件质量点模型和软件质量度量模型,并运用模型集成的方法,构建了软件质量的集成评估模型。通过模型转换方法将其转换为层次分析模型,并针对模型中权重指标难以确定的问题引入了模糊分析的方法。最后通过一个实例验证了该方法的可行性。     

基于梯形模糊AHP的信息安全风险综合评估

为提高风险评估的可信度,针对传统的层次分析法在确定因素权重方面存在的问题,引入梯形模糊数来确定权重,提出了基于梯形模糊层次分析法的综合评价方法,解决了风险评估中定性指标定量评估的难题。通过实例分析,验证了该方法的合理性和可行性。     

层次分析法和Theil指数在软件需求风险评估中的应用研究

软件的需求风险分析对项目的成败起着重要的决定性作用.为了对需求风险进行定性评价、定量评估,首先对软件项目的需求风险影响因素进行识别,建立风险因素的模糊变换矩阵,然后采用层次分析法确定各因素的权重,再使用Theil指数模型综合考察各影响因素权重对最终风险的影响,建立模糊指数组合预测模型,最后依据需求风险数据集进行模型验证...     

基于模糊理论的Ad Hoc网络可信性研究

在分析Adhoc网络的通信能力的基础上,确定网络可信度影响因子,并采用模糊层次分析法计算可信度影响因子的权重值.提出了基于模糊理论的AdHoc网络可信性评判模型.该模型对Adhoc网络进行了单因素评判和综合评判,并通过实例运算证明了模型的科学性与合理性.     

基于模糊层次分析法的计算机网络安全评价

根据计算机网络的特点及网络安全涉及的因素,建立计算机网络安全评价的层次结构模型,运用基于三角模糊数的模糊层次分析法实现对网络安全的综合量化评价。实例的计算结果表明模糊层次分析法适用于定性判断的量化综合,它为计算机网络安全评价提供了一种探索性方法。     

基于三角模糊数层次分析法与网络仿真的网络整体性能评价研究

结合模糊理论提出了改进的模糊AHP算法——基于三角模糊数层次分析法(Triangular Fuzzy Number Analytic Hierarchy Process,TFAHP),使得结论更加客观、合理,同时简化了AHP算法的步骤。在提出基于THAHP网络整体性能评价模型的基础上,阐述了应用网络仿真技术进行网络整体性能评价的方法。包括性能指标选取、仿真试验设计、仿真性能参数计算和仿真结果处理与分析的方法,并给出了对一个具体的多媒体网络进行网络整体性能评价的应用实例。     

模糊层次分析法在方案优选方面的应用

给出了一种实用的层次分析方法(AHP),即模糊层次分析法(FAHP),讨论了它的原理和步骤。特别是模糊层次分析法,它是基于模糊集合理论的。它采用模糊一致关系去实现模糊推导,使用模糊一致矩阵去实现动态自适应。分析了该方法的适用范围,并提供了具有一般特点的方案优选的实例。提出了模糊层次分析方法的不足之处。     

基于模糊层次分析法的网络业务性能评估模型

针对现有网络性能评估方案不能兼顾用户的业务偏好和网络实际情况以及指标权重计算方法存在片面性的问题,提出一种面向网络业务性能的综合评价模型。该模型考虑到用户偏好,以网络应用业务为导向,首先建立一种评价层次结构,分别确立层次结构中的准则权重和方案权重;然后对实际测量数据进行规格化处理;最后利用模糊层次分析法(AHP)进行计算得到目标链路中各业务的性能评估值。实验结果表明,所提模型不仅可以评估目标链路的整体性能情况,还可以结合用户业务偏好,对单个的网络业务的性能进行有效评估,有利于更好地实现区分服务的网络。     

模糊层次分析法和人工神经网络模型在电梯风险评估中的应用

针对特种设备风险评估模型中的人工神经网络(ANN)技术一般采用定量加权平均法. 由于其权值是以专家经验为依据, 导致评估所需时间太长. 针对这个缺点, 建议将模糊层次分析法(F-AHP)和ANN相结合, 用于对电梯的风险评估中. 实验证明用此方法进行评估所用的时间比ANN方法要少, 且评估的准确性没有降低.     

基于粒子群算法的模糊层次分析模型

针对模糊层次分析法中存在的模糊判断矩阵一致性检验和修正困难、元素权重计算繁琐的问题,从模糊判断矩阵的定义角度出发,构建了基于粒子群算法的模糊层次分析模型(PSO-FAHP),提出了包含模糊判断矩阵一致性修正及各元素排序过程的非线性带约束优化问题,引入粒子群算法实现了问题的求解,并分析了模型的合理性。最后通过数值算例对比了模型的计算结果,验证了模型的正确性。对模糊层次分析法的实践应用具有一定的指导意义。     

基于模糊层次分析法的软件易用性过程综合评价方法

对软件易用性过程进行综合评价是开发可信软件产品的必要因素。通过构建软件易用性生命周期模型和软件易用性度量模型搭建软件易用性过程综合评价模型, 通过此模型对软件开发中的易用性问题进行有效的指导, 规范易用性开发过程, 从而保证能够产出高易用性的产品; 利用模糊层次分析法对此模型进行度量评价, 量化评估易用性过程优劣; 通过一个实例分析说明该方法可以有效地应用于软件开发过程中, 能够尽早地确保软件产品的易用性, 为开发高易用度的软件产品指明了方向。     

层次分析法和神经网络相融合的教学质量评价

为提高教学质量评价准确性,提出一种基于层次分析法和神经网络相融合的教学质量评价方法（AHP-BPNN）。采用层次分析法构建评价指标体系,筛选出对评价结果有重要影响的指标作为BP神经网络输入,采用神经网络建立教学质量评价模型。仿真结果表明,AHP-BPNN不仅简化神经网络的结构,而且提高了教学质量的评价精度和评价效率,是一种可行、有效的教学质量评价方法。     

An approach to budget allocation for an aerospace company—Fuzzy analytic hierarchy process and artificial neural network

Budgetary allocations of resources are made in all businesses, but their volume and composition vary; and efficient budget allocation is fundamental to flow in businesses. The objective of the allocation problem is to determine the required budget for each department (or section) of a company so as to maximize the sum of the company's benefits. The purpose of this paper is to find a suitable degree of fuzziness for preference rankings and to demonstrate an example of budget allocation using artificial intelligence programming, and fuzzy analytic hierarchy process (FAHP). An efficient budget allocation method using FAHP will be provided for businesses. This method is suitable for use in evaluating proposed policies (including tangible and intangible information). A comparison between FAHP and artificial neural network (ANN) will be also made in this paper.An aerospace company's budget allocation problem is investigated as a case study in this research, which will illustrate how to solve this problem. The case study utilizes a two-stage interview (semi-structured interview and in-depth interview) to select their budget allocations given a number of tangible and intangible criteria. The results from the case study are pertinent to other real-world allocation problems that share many of the characteristics of problems, such as decision makers’ subjective opinions.     

Modification of the fuzzy analytic hierarchy process via different ranking methods

The use of fuzzy set theory in the analytic hierarchy process (AHP) has gained popularity in recent years as part of the multiple criteria decision-making (MCDM) process to more realistically reflect human judgment. However, due to the nature of fuzzy calculations, this situation imposes more computational load. The aim of this study is to propose methods for obtaining accurate weights from fuzzy pairwise comparison matrices with the least amount of computational load possible. In this context, two different fuzzy AHP (FAHP) methods based on fuzzy numbers ranking methods have been proposed and these proposed methods are compared with commonly accepted FAHP methods. Magnitude-based fuzzy AHP (MFAHP), which is one of the proposed methods, has outperformed all other methods according to accurate weight and computational load. Although the other proposed method, called the total difference-based fuzzy AHP (TDFAHP), gave better results than the frequently used Chang's fuzzy extent analysis method, it could not produce more accurate weight results than many other methods in general. But performance analysis shows that it is as good as the MFAHP in terms of computational load.     

Selection of non-traditional machining processes using analytic network process

Non-traditional machining (NTM) processes are now being widely used to generate intricate and accurate shapes in materials, like titanium, stainless steel, high strength temperature resistant (HSTR) alloys, fiber-reinforced composites, ceramics, refractories and other difficult-to-machine alloys having higher strength, hardness, toughness and other diverse material properties. Generation of complex shapes in such materials by the traditional machining processes is experienced to be difficult. For effective utilization of the capabilities of different NTM processes, careful selection of the most suitable process for a given machining application is often required. Selection of the best suited NTM process for a work material and shape feature combination requires the consideration of several criteria. In this paper, an analytic network process (ANP)-based approach is proposed to select the most appropriate NTM process for a given machining application taking into account the interdependency and feedback relationships among various criteria affecting the NTM process selection decision. To avoid the difficult and time consuming mathematical calculations of the ANP, a computer program is also developed in Visual Basic 6.0 with graphical user interface to automate the entire NTM selection decision process. It simply acts as an ANP solver. The observed results from the ANP solver are quite satisfactory and match well with those obtained by the past researchers.     

直觉模糊层次分析法处理医院用药风险评价

基于直觉模糊集的模糊逼近理论,给出了将直觉模糊互补判断矩阵转换为模糊逼近矩阵的方法,提出了直觉模糊环境下的AHP方法,简称为直觉模糊层次分析法,将其应用于对医疗机构的用药风险的评价问题,给出了各子风险类别的权重。经检验,是一种实用性较强的医疗机构用药风险评价方法。