基于模糊优选法的灰靶决策模型

针对多指标灰靶决策模型,提出了同时考虑了方案与正靶心和负靶心的距离,还能进一步考虑两个距离之间的折中、权衡问题的模糊优选法,克服了传统的灰靶决策模型只考虑正靶心距对方案进行排序不考虑负靶心距对决策的影响。文末,通过实例说明了本文所提出的方法的实用性和有效性。     

基于灰熵的不确定型决策方法及其应用

基于灰熵理论,提出了一种不确定型决策的量化计算方法.灰熵是离散序列均衡程度的测度,灰色关联度是参考序列和比较序列之间接近程度的测度.该方法将灰色关联度和均衡度合成为均衡接近度,并以此作为决策准则,从而避免了低层次多因素权重确定的主观性,使决策模型更加合理.通过在太原市热力公司决策中进行应用,结果表明,该方法较其它方法更易于操作,结果更符合实际.     

区间灰数表征与算法改进及其GM (1, 1)模型应用研究

针对目前区间灰数的表征和运算过程中存在着较为严重的问题，即：运算结果对灰数的灰度进行不必要的放大，造成信息的严重失真，笔者定义了标准区间灰数与第一和第二标准区间灰数的概念，分析了第一和第二标准区间灰数之间的关系，进一步设计了普通区间灰数与标准区间灰数之间的转换规则，提供了标准区间灰数之间的比较与运算法则，较好地解决了区间灰数之间的大小比较与运算问题。最后，将这一研究成果应用于基于区间数的GM (1, 1)模型预测问题，取得了良好的效果。     

模糊序列的GM(1,1)建模

本文在不改变GM(1,1)模型建模机理的前提条件下,运用模糊回归理论对GM(1,1)模型进行优化,通过定义三角模糊数的左、中、右距离,将模糊序列的GM(1,1)模型的求解转化为线性规划模型的求解,并通过实例加以论证。研究表明模糊序列的GM(1,1)模型保留了GM(1,1)模型所需建模数据少和具有预测功能的特点,且能为决策者提供一个决策区间。     

企业研发项目中止决策的灰靶分析模型

中止决策是研发项目过程管理中不可或缺的环节,是企业提高研发效率、改善资源配置的有效手段。根据该问题模糊多属性分类决策的本质,以案例分析为起点,结合模糊集与灰靶决策理论构建求解属性权重、临界点等分类参数的优化模型,利用所得参数计算待判项目的综合靶心矩,最后按照综合靶心矩所属区间对待判项目进行分类决策。将所提方法应用于一个具体的决策分析过程,其可行性和合理性得到验证。     

基于四维空间相似度的军械紧急调运方案优选方法

分析了军械紧急调运方案优选决策的混合多属性特点。基于TOPSIS方法的基本原理，首先通过方案属性值的四维属性值转化，将不同类型的属性值统一成相同的形式，然后通过定义四维空间相似度，将各类型属性值的相似度计算规范到统一的度量空间，解决了含有实数、区间数、语言值和不确定语言值等多种不同决策属性值的的军械紧急调运方案优选这一混合型多属性决策问题。计算示例表明了该方法的可操性和有效性。     

Fuzzy值测度的生成

继Sugeno[1]关于Fuzzy测度与Fuzzy积分的研究以后,这方面的工作已有较深入的发展。但这些研究的共同特点是测度的值仍为普通实数。文[2]给出了Fuzzy数的一个新描述,它不但使普通的实数和区间数成为它的特例,而且使它们关于各自的四则运算与它的某一子集关于[2]中定义的运算同构,[3]建立了这种Fuzzy数的收敛与Fuzzy值函数的Riemann积分。在此基础上,本文将建立Fuzzy值测度理论的框架,所有未定义的符号均可在[2]、[3]中找到。     

A-E方法的多目标车间排产方案精选决策

分析车间排产方案决策阶段企业各部门利益关系,构建基于层次分析和熵理论(A-E)及欧氏范数的多目标车间排产方案决策模型.具体利用AHP法建立车间排产方案层次化的决策体系,采用层次分析法和熵理论(A-E)方法确定各决策因素的权重.在车间排产决策中,利用欧氏范数来描述备选排产方案与理想方案的距离大小作为精选的依据.并结合实例进行数值演算和验证,计算表明该模型能根据生产实际数据帮助决策者快速的从多个车间排产方案中精选,得到令各部门最满意的唯一方案.     

基于灰靶决策对BFCC力学性能及抗渗性能的评估

试验研究了不同纤维长度和不同应力比条件下BFCC抗压强度及抗渗性能的变化规律,运用灰色系统理论对BFCC的抗压强度、电通量与孔结构参数进行灰色关联分析,运用多目标加权灰靶决策模型对不同应力比下的BFCC综合性能进行评估.研究表明:随纤维长度增加,BFCC抗压强度先增加后降低,BFCC电通量值持续减小;相同应力比条件下,延长养护龄期,BFCC的抗压强度增加,电通量值减小;养护龄期相同时,增加应力比,BFCC电通量值先减小后增大,且在应力比为0.2时为最小,而其抗压强度则呈持续减小的趋势.根据灰色关联分析结果,采用气泡间距系数从微观角度表征抗压强度的变化,采用气泡比表面积表征电通量的变化,运用多目标加权灰靶决策模型定量评估不同应力比下BFCC的综合性能,随应力比不断增大,BFCC相对综合性能指标不断减小,且其综合性能不断劣化.     

医疗资源配置的动态生成系数灰色预测模型北大核心CSSCI

针对人口密集型都市的医疗资源配置缺乏准确性预测作为实时决策依据的难题,利用灰色理论的小样本建模优势进行预测方法拓展优化,建立等维递补灰色预测方法以提高灰区间白色度和淡化灰平面灰度;挖掘灰色生成系数与外部影响因素间的内在映射关联,提出了动态生成系数优化的灰色理论医疗需求预测方法;动态拟合人口总量变化与灰色生成系数以实现预测模型实时重构,解决了传统灰色预测方法的纯样本序列建模局限,显著提高了预测算法的准确性,输出的医疗需求趋势可有力支撑医疗资源配置决策。     

密闭空间爆炸温度动态不确定度评估方法研究

融合了灰色模型GM(1,1)、Bootstrap方法以及不确定度评定理论,建立了密闭空间内爆炸温度动态测量不确定度的灰自助评估模型GBM(1,1),选取某次爆炸试验300s的温度数据作为分析数据,运用估计真值、估计区间和平均不确定度等参数表征其估计结果。实验结果表明,GBM(1,1)模型融合了灰色模型GM(1,1)和Bootstrap方法的优势,可以准确模拟动态测量数据的概率分布,并实时跟踪被测量瞬时值的变化趋势。相比于灰色模型GM(1,1)和Bootstrap方法,灰自助模型GBM(1,1)具有更高的真值估计准确度和区间估计可靠度,其估计误差分布区间为［-12.62, 13.58］,标准差为8.69℃,最大相对误差为1.2%,区间估计可靠度高于90%,估计区间能够较完整地包络被测量的动态波动范围。由此证明GBM(1,1)模型能够对密闭空间内爆炸温度的动态测量不确定度做出准确评估。     

Waste treatment: an environmental,economic and social analysis with a new group fuzzy PROMETHEE approach

Most complex decisions involve several stakeholders and therefore need to be solved using a group multi-criteria decision method. However, stakeholders or decision-makers often have divergent views, especially in the environmental sector. In order to integrate this divergence, a new group fuzzy PROMETHEE approach is introduced to combine the traditional environmental criteria of life cycle assessments with social and economic criteria. The modelling of uncertainty within the group of decision-makers using a fuzzy approach makes this method unique. The proposed fuzzy approach differs significantly from the standard one. The decision-makers express their judgments in crisp forms. In order to take into account the intrinsic dispersion of judgments within the group, a posteriori fuzzification procedure is applied. The crisp values are not simply aggregated; they are converted into a triangular fuzzy number based on the given evaluations. As a consequence, the definition of fuzzy membership functions, as required in standard fuzzy logic, is not required, which simplifies the process and makes it more reliable. The new approach is illustrated with a real case study concerning the selection of the best waste treatment solution in a natural park from among a traditional incinerator and an innovative integrated plant.     

Quality-driven recovery decisions for used components in reverse logistics

Reverse logistics has emerged as a promising strategy for enhancing environmental sustainability through remanufacturing, reusing, or recycling used components. It is crucial to pursue quality-driven decision-making for component recovery because quality is a dominant factor for component salvage value and its recoverability. To maximise the profit from component recovery, a quality-driven decision model was proposed in this study. Remaining useful life (RUL) was utilised as a measure of quality in the proposed model, where conditional RUL distribution was predicted by utilising both the failure data and condition monitoring data based on a proportional hazard model. Under RUL uncertainty, an interval decision-making approach was developed to suggest recovery strategies for the decision-makers to identify a satisfactory solution according to their risk preferences. Compared to the existing approaches for quality-driven recovery decision-making based on RUL prediction, this work provides a more accurate and powerful approach to managing and mitigating decision risk. Numerical experiments demonstrated the effectiveness and superiority of the proposed model.     

A bi-objective robust model for emergency resource allocation under uncertainty

Emergency resource allocation constitutes one of the most critical elements of response operations in the field of emergency management. This paper addresses an emergency resource allocation problem which involves multiple competing affected areas and one relief resource centre under supply shortage and uncertainty in the post-disaster phase. In humanitarian situations, both the efficiency and fairness of an allocation policy have a considerable influence on the effectiveness of emergency response operations. Thus, we formulate a bi-objective robust emergency resource allocation (BRERA) model which tries to maximise efficiency as well as fairness under different sources of uncertainties. To obtain decision-makers’ most preferred allocation policy, we propose a novel emergency resource allocation decision method which consists of three steps: (1) develop a bi-objective heuristic particle swarm optimisation algorithm to search the Pareto frontier of the BRERA model; (2) select a coefficient to measure fairness; and (3) establish a decision method based on decision-makers’ preference restricted by the fairness coefficient. Finally, a real case study taken from the 5 December 2008 Wenchuan Earthquake demonstrates the effectiveness of the proposed method through numerical results. The solution and model robustness are also analysed.     

Incorporating Confidence Intervals on the Decision Threshold in Logistic Regression

This paper discusses an application of confidence intervals to the threshold decision value used in logistic regres]sion and discusses its effect on changing the quantification of false positive and false negative errors. In doing this a grey area, in which observations are not classified as success (1) or failure (0), but rather ‘uncertain’ is developed. The size of this grey area is related to the level of confidence chosen to create the interval around the threshold as well as the quality of logistic regression model fit. This method shows that potential errors may be mitigated. Monte Carlo simulation and an experimental design approach are used to study the relationship between a number of responses relating to classification of observations and the following factors: threshold level, confidence level, noise in the data, and number of observations collected. Copyright © 2015 John Wiley & Sons, Ltd.     

基于petri网装配序列灰色聚类决策研究

利用petri网直观图形化特点建立装配模型,并根据petri网原理得出可行装配序列。影响可行装配序列的大部分因素是定性的、模糊的、非数值的,将装配序列看作灰色系统,采用灰色聚类决策方法对可行序列进行评估。分析了影响因素的灰色分类并且研究了灰色聚类决策步骤。实例分析表明,该方法为petri网原理所得可行序列进行正确的评估并且得出决策向量。     

疫情演变不确定情境下考虑多参考点的应急医疗用品生产决策模型

针对疫情演变不确定情境下考虑多种决策信息和企业社会责任的应急医疗用品生产决策问题，提出一种基于累积前景理论的考虑多个参考点的应急医疗用品生产决策模型。综合考虑企业经济目标和社会责任对生产决策的影响，基于决策者的风险规避、参考依赖和损失厌恶等心理特征，建立多个参考点以表征疫情演变方向不确定时决策者期望的不确定性。将含有多种评价信息的决策矩阵归一化处理，利用不同演变方向的参考点，建立益损决策矩阵。根据决策者面对收益和损失的态度，建立主观价值矩阵，依据主观概率函数得到扭曲后演变方向的概率。计算各个生产方案的累积前景值并对其进行优先排序。通过算例验证模型的有效性，进行对比分析和敏感性分析证明模型的优越性。     

Investing in lean manufacturing practices: an environmental and operational perspective

Lean manufacturing practices (LMPs) and corporate environmental sustainability are becoming inextricably linked. Throughout the lean and green debate, many organisations have recognised that LMPs have implications for their sustainable development and competitive positioning. Not only LMPs are complex on their own, but when perceived from an environmental sustainability perspective, the decision to implement an LMP can become even more intricate. Although general tools exist, the lack of effective decision-making tools to help in the implementation of LMPs with an environmental sustainability dimension is palpable. Thus, this study tackles the aforementioned decision problem by incorporating environmental and operational performance outcome expectations as these expectations are viewed in light of the ease of implementation of various LMPs. A novel multi-criteria decision-making (MCDM) model for evaluation of LMPs is developed in this respect. The model integrates a three-parameter interval grey number with rough set theory and the TODIM method. The model is run using empirical data from six manufacturing organisations. The findings facilitate the identification of a ‘locus of investments’ for a better selection of LMPs. The robustness of the decision support model developed is assessed through sensitivity analysis.     

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。