Approach to obtaining weights of uncertain ordered weighted geometric averaging operator

The ordered weighted geometric averaging（OWGA） operator is extended to accommodate uncertain conditions where all input arguments take the forms of interval numbers. First, a possibility degree formula for the comparison between interval numbers is introduced. It is proved that the introduced formula is equivalent to the existing formulae, and also some desired properties of the possibility degree is presented. Secondly, the uncertain OWGA operator is investigated in which the associated weighting parameters cannot be specified, but value ranges can be obtained and the associated aggregated values of an uncertain OWGA operator are known. A linear objective-programming model is established; by solving this model, the associated weights vector of an uncertain OWGA operator can be determined, and also the estimated aggregated values of the alternatives can be obtained. Then the alternatives can be ranked by the comparison of the estimated aggregated values using the possibility degree formula. Finally, a numerical example is given to show the feasibility and effectiveness of the developed method.

Generalized ordered weighted averaging operators based methods for MADM in intuitionistic fuzzy set setting

Multiattribute decision making（MADM） problems, in which the weights and ratings of alternatives are expressed with intuitionistic fuzzy（IF） sets, are investigated.Firstly, the relative degrees of membership and the relative degrees of non-membership are formulated as IF sets, the weights and values of alternatives on both qualitative and quantitative attributes may be expressed as IF sets in a unified way.Then a MADM method based on generalized ordered weighted averaging operators is proposed.The proposed method is illustrated with a numerical example.

Linear goal programming approach to obtaining the weights of intuitionistic fuzzy ordered weighted averaging operator

The multiple attribute decision making problems are studied, in which the information about attribute weights is partly known and the attribute values take the form of intuitionistic fuzzy numbers. The operational laws of intuitionistic fuzzy numbers are introduced, and the score function and accuracy function are presented to compare the intuitionistic fuzzy numbers. The intuitionistic fuzzy ordered weighted averaging （IFOWA） operator which is an extension of the well-known ordered weighted averaging （OWA） operator is investigated to aggregate the intuitionistic fuzzy information. In order to determine the weights of intuitionistic fuzzy ordered weighted averaging operator, a linear goal programming procedure is proposed for learning the weights from data. Finally, an example is illustrated to verify the effectiveness and practicability of the developed method.

Three methods for generating monotonic OWA operator weights with given orness level

Based on the properties of ordered weighted averaging (OWA) operator and regular increasing monotone (RIM) quantifier, three methods for generating monotonic OWA operator weights are proposed. They are geometric OWA operator weights, equidifferent OWA operator weights and the modified RIM quantifier OWA weights. Compared with most of the common OWA methods for generating weights, the methods proposed in this paper are more intuitive and efficient in computation. And as there are more than one solution in most cases, the decision maker can set some initial condition and chooses the appropriate solution in the real decision process, which increases the flexibility of decision making to some extent. All these three OWA methods for generating weights are illustrated by numerical examples.

Some OWA operator weights determination methods with RIM quantifier

Based on the quantifier guided method,an ordered weighted averaging(OWA)weights generating method under given orness level with regular increasing monotone(RIM)quantifiers is proposed.Then the RIM quantifier based OWA weights generating method is modified to make the generated weights be monotonic,which can be used to express the decision maker's consistent preference information.Finally,both of these weights generating methods are extended to their generic forms,so that they can generate the OWA weights for any ordinary elements set with any given aggregated value.

基于IOWA算子的空气质量组合预测

对空气质量指数进行预测,反映数据波动的趋势与规律,提出基于诱导有序加权平均(IOWA)算子的Prophet-EEMD-XGBoost模型。首先利用Prophet算法将数据分解为趋势项、季节项、节假日、误差项;然后集合经验模态分解EEMD将原始数据分为9项IMF以及残余分量。最后,利用XGBoost模型对各分量分别进行预测,预测结果利用MAE、MAPE、RMSE、RMSPE进行评价。结果表明,相较于一些单预测模型与其他组合预测模型,基于IOWA算子的Prophet-EEMD-XGBoost组合预测模型有更好的预测效果。

A Self-determined Evaluation Method for Science Popularization Based on IOWA Operator and Particle Swarm Optimization

With the increase of science popularization, evaluation of science popularization has become an urgent demand. Considering science popularization bases as independent agents, a self-determined evaluation approach for science popularization using induced ordered weighted averaging (IOWA) operator and particle swarm optimization (PSO) is proposed in this paper.Firstly, six factors including science popularization personnel, space, fund,media, activity and influence are selected to construct an index system for science popularization evaluation. On this basis, the absolute dominance and relative dominance of evaluation indexes are used as induced components, and the prior order of the evaluation indexes is determined. Besides, the optimization model of index weighted vectors is established by IOWA operator, index weighted vectors are calculated by particle swarm optimization algorithm, and index weighted vectors and evaluation value vectors are obtain. Finally, the optimal evaluation vectors and evaluation results are given according to the Perron-Frobenius decision eigenvalve theorem .

基于IOWA算子的航班订座组合预测模型

航班订座预测的经典方法包括回归模型、增量模型、指数平滑模型等,在复杂多变的非线性数据环境下,各单项模型存在预测效果不稳定及健壮性不满意的情况。为解决这一问题,本文构造了一个基于诱导有序加权平均(IOWA,induced ordered weighted averaging)算子的航班订座组合预测模型,组合模型采用各单项模型的预测精度作为诱导因子,动态计算单项预测模型组合权重及构建未来的动态预测组合。但传统的IOWA组合预测模型对随机波动较大的数据适应性较差,本文在深入分析航班旅客订座量数据规律的基础上,改进了该模型在预测期的诱导因子,取得了较好的预测结果。

Interval-valued intuitionistic fuzzy aggregation operators

The notion of the interval-valued intuitionistic fuzzy set （IVIFS） is a generalization of that of the Atanassov＇s intuitionistic fuzzy set. The fundamental characteristic of IVIFS is that the values of its membership function and non-membership function are intervals rather than exact numbers. There are various averaging operators defined for IVlFSs. These operators are not monotone with respect to the total order of IVIFS, which is undesirable. This paper shows how such averaging operators can be represented by using additive generators of the product triangular norm, which simplifies and extends the existing constructions. Moreover, two new aggregation operators based on the t.ukasiewicz triangular norm are proposed, which are monotone with respect to the total order of IVIFS. Finally, an application of the interval-valued intuitionistic fuzzy weighted averaging operator is given to multiple criteria decision making.

基于IOWA-VAC的地铁车站暴雨内涝脆弱性评价

为科学评价地铁车站暴雨内涝脆弱性,确定地铁车站暴雨内涝脆弱性的敏感性指标,提高地铁车站暴雨内涝管理水平,提出基于IOWA-VAC的地铁车站暴雨内涝脆弱性评价方法,该方法基于压力—状态—响应(PSR)理论构建地铁车站暴雨内涝脆弱性评价指标,利用IOWA算子按升序原则对指标数据重新排序,同时引入θ系数动态调整区间边界权值,得到指标权重,引入向量夹角余弦(VAC)检验待评价目标向量和理想目标向量一致性的贴近度,实现了代数思维向空间几何思维的转变,得到待评价目标等级。进而以郑州地铁5号线海滩寺站为例,运用单因素敏感性分析确定地铁车站暴雨内涝脆弱性的敏感性指标,评价了海滩寺站暴雨内涝脆弱性。结果表明,郑州地铁5号线海滩寺站暴雨内涝脆弱性等级为Ⅳ级,属于高度风险,为地铁车站暴雨内涝脆弱性管理提供了可行性建议。

基于博弈论组合赋权-云模型的消防员训练损伤风险评估

为降低消防员训练损伤发生率,从个人因素、环境因素和组织管理因素3个方面,提出包含19个指标的消防员训练损伤风险评估指标体系;采用Bonferroni有序加权平均算子(BON-OWA算子)修正层次分析法(AHP)的误差,获得各指标的主观权重;引用基于指标相关性的权重确定法(CRITIC)确定各指标的客观权重;基于博弈论组合赋权法确定各指标的组合权重;引入云模型理论,建立消防员训练损伤风险评估模型,并以河南省某消防救援大队消防员的训练为例,验证该模型的科学性。结果表明:训练装备的战技性能状况指标对消防员的训练损伤风险等级影响最大;该风险评估模型通过数据处理,可以确定消防员训练损伤风险等级;案例中,消防员在5种不同情景下的训练损伤风险等级均为一般风险,这与模糊综合评价法所得结果一致。

基于IOWA算子的赤潮LMBP神经网络组合预测模型

针对现有赤潮单项预测模型不能综合各种理化因子有效信息的问题,提出了一种赤潮组合预测模型,该模型基于诱导有序加权平均(IOWA)算子与Levenberg-Marquardt人工神经网络(LMBP)算法,能够有效集结各种理化因子的信息,其预测精度比单项LMBP神经网络预测模型有较大提高.采用烟台四十里湾赤潮监测数据对该模型进行实验,结果验证了该模型的有效性和实用性.

基于改进IOWA组合模型的舰船装备维修费预测

针对诱导有序加权平均(induced ordered weighted averaging,IOWA)组合模型中,单项预测方法精度相同时诱导值可能存在多种不同排列顺序的问题,提出综合考虑预测方法的历史性能确定其唯一排序;针对实践中单项预测方法较多时,IOWA组合模型构建过程十分复杂的问题,通过构建适合于Matlab软件计算的最优化模型矩阵形式,降低计算复杂性;结合舰船装备维修费预测中样本小、预测方法多的特征,采用各单项预测方法的历史平均拟合精度来估计其预测期诱导值的排序,并以此计算组合预测值。最后实例说明了该方法的有效性。

基于IOWA算子的模糊语言偏好矩阵排序方法

给出了方案之间比较的一种模糊语言标度及其相应的区间数表示形式。运用模糊语言标度构造出模糊语言偏好矩阵。利用导出的有序加权平均(IOWA)算子对模糊偏好信息进行集结,提出了一种基于IOWA算子的模糊语言偏好矩阵排序方法,并把该方法进一步运用于群体决策中的方案排序和择优。最后通过算例说明了该方法的可行性和实用性。

基于区间精度和IOWA算子的变权组合预测方法

现有的区间组合预测模型往往是将单项预测方法不同时刻的预测精度转化为实数,这将导致预测信息的损失。因此,文章提出一种新的区间预测精度的概念,将单项预测方法不同时刻的精度用区间数表示,并利用可能度对其进行排序;将排序后的预测精度作为诱导变量,构建基于区间预测精度和诱导有序加权平均(induced ordered weighted averaging,IOWA)算子的区间变权组合预测模型;实例分析说明了该区间组合预测方法的有效性;最后,对模型参数做了灵敏度分析。

基于GIOWA算子的备件服务绩效评价方法

当前对服务绩效的研究更多的是对服务质量的衡量,但缺少从顾客感知的角度对服务业务的评价研究。同时,由于不同产品备件服务的属性重要性随着服务类型等因素的不同而有较大的差距,很难赋予服务属性固定的权重值。因此,为了全面分析备件服务绩效的影响因素,参考SERVQUAL模型建立了备件服务绩效评价体系。为了从顾客感知的角度对服务业务进行评价,采用GIOWA算子的方法以有效地融合评价体系中的多组模糊和不确定的客户评价信息。实例结果表明,该方法取得了对模糊语言评价的良好集结效果,为备件服务绩效评价提供了一种定性和定量相结合的评价方法。

基于IOWA的我国能源需求组合预测模型的应用研究

基于诱导有序加权平均算子的组合预测方法是近年来发展的具有广泛应用的预测模型。本文首先采用多元回归、基于HP的ARMA模型及灰色预测方法建立了我国能源需求的单项预测模型,随后引进诱导有序加权平均算子的概念,建立了我国能源需求的组合预测模型,并对我国未来的能源需求量进行了实证分析。预测结果表明该模型能提高组合预测精度,是一种较为简便、快捷且预测效果较好的方法。

基于三维空间的新型配电网关键技术时空适应性评估

在新型电力系统发展的背景下,针对配电网规划建设和配电网关键技术在实际应用时与所处时空的适应能力、应用缺陷难定量化等问题,分析新型配电网关键技术的建设条件,提出一种基于电网满足度、时空资源度、成效改善度的三维立体空间的技术时空适应性评估模型.采用连续区间有序加权平均算子对层次分析法主观赋权进行改进,引入标准间冲突性相关性法构建的主客观组合赋权方法,解决了单一赋权法具有偏向性的问题.利用模糊综合评价法确定各指标隶属度,进而得到新型配电网关键技术评价等级.算例结果表明,所提方法能够量化配电网关键技术与所处时空的契合程度和双向辨识关键技术对于不同时空的适应性以及不同时空对于技术应用的满足度,揭示配电网关键技术的薄弱环节,有助于提升配电网投资效率与效益,更好地服务经济社会高质量发展.

基于IOWA-FAHP的物流需求组合预测模型

面向物流需求的预测问题,提出了一种基于诱导有序加权平均算子和模糊层次分析法相结合的组合预测模型。首先,采用诱导有序加权平均算子建立组合预测模型,并基于历史数据解算各单项预测模型的加权系数;然后,针对组合预测模型中各单项模型的加权系数确定问题,引入模糊层次分析法,得到各单项预测模型的公平排序,进而确定各单项预测模型在预测年的诱导值序列,从而预测出相应物流需求量;最后通过应用算例验证了该模型的正确性和可靠性。

基于OWA算子的模糊n-cell数排序方法

针对基于模糊n-cell数的多属性排序问题,提出了一种基于有序加权平均算子(OWA算子)的模糊n-cell数排序方法。该方法首先根据样本数据对评估对象的属性构造模糊n-cell数,其次根据均值将属性按照从大到小排列,然后选取合适的权重向量,应用OWA算子进行信息聚合得到综合模糊n-cell数,接着根据各分量均值得到排序结果。最后,将该方法运用到实例中,并与传统的均值方法进行了比较。结果表明该方法不仅灵活有效,可根据具体情况选择不同的OWA权重来消除部分不合理的情况,使结果更有说服力,还弥补了传统均值方法的不足。