混合不确定性下基于区间证据随机抽样的统一赋权法

由于评价问题的复杂性,以及专家知识和经验的局限性,群决策信息中往往同时包含不同程度的不确定性。在区间证据理论框架下对认知不确定性信息、多区间概率信息、不完全信息等混合不确定信息进行统一表示,进而根据区间证据间的相似度对群决策信息进行融合;采用区间数对评语等级进行定量化,并根据融合结果构造个指标重要程度的概率分布函数;最后通过概率分布函数的蒙特卡洛随机抽样确定指标权重。算例分析结果表明了新方法的有效性。     

Prioritisation of an IT budget within a local authority

This paper describes the prioritisation of an IT budget within a department of a local authority. The decision problem is cast as a simple multiattribute evaluation but from two perspectives. First, as an exercise in group decision making. Here the emphasis is on a shared process wherein the object is to obtain consensus. The use of an explicit evaluation framework and the ability to interact with the evaluation data in real time via a simple spreadsheet model were found to improve the decision making. Second, the prioritisation is made analytically. The motivation is to determine the degree to which the rankings are the result of the structural characteristics of the projects themselves rather than of the differences in importance attached to the achievement of the goals represented by the project attributes. Three methods are used: Monte Carlo simulation of ranks, cluster analysis based on attributes and an approach based on entropy maximisation. It is found that in the case studied the structure inherent in the data is high and so the results of the analyses are robust. Finally, a procedure is suggested for the appropriate use of these analyses via a facilitator to aid prioritisation decisions.     

A simulation study of IT outsourcing in the credit card business

The use of Monte Carlo simulation for evaluation of financial risk of an information technology project selection decision is described. A major Thai bank considered the opportunity to expand credit card operations through information technology (IT). Alternatives considered were in-house development and outsourcing. There were many strategic reasons for the initiative. However, there were also many risks associated with the proposal. A Monte Carlo simulation spreadsheet model was used to model risk parameters, and to analyze key performance variables of financial performance. Key output variables were the number of cardholders expected, project net present value, net profit, and expected return on investment. The spreadsheet model made entry of model elements transparent, and Monte Carlo simulation provided clear visual display of the financial output variables. The bank used this information in its decision to outsource its credit card operations.     

An approach to the valuation and decision of ERP investment projects based on real options

The risks and uncertainties inherent in most enterprise resources planning (ERP) investment projects are vast. Decision making in multistage ERP projects investment is also complex, due mainly to the uncertainties involved and the various managerial and/or physical constraints to be enforced. This paper tackles the problem using a real-option analysis framework, and applies multistage stochastic integer programming in formulating an analytical model whose solution will yield optimum or near-optimum investment decisions for ERP projects. Traditionally, such decision problems were tackled using lattice simulation or finite difference methods to compute the value of simple real options. However, these approaches are incapable of dealing with the more complex compound real options, and their use is thus limited to simple real-option analysis. Multistage stochastic integer programming is particularly suitable for sequential decision making under uncertainty, and is used in this paper and to find near-optimal strategies for complex decision problems. Compared with the traditional approaches, multistage stochastic integer programming is a much more powerful tool in evaluating such compound real options. This paper describes the proposed real-option analysis model and uses an example case study to demonstrate the effectiveness of the proposed approach.     

Heterogeneous uncertainties in cholesterol management

Physicians use clinical guidelines to inform judgment about therapy. Clinical guidelines do not address three important uncertainties: (1) uncertain relevance of tested populations to the individual patient, (2) the patient’s uncertain preferences among possible outcomes, and (3) uncertain subjective and financial costs of intervention. Unreliable probabilistic information is available for some of these uncertainties; no probabilities are available for others. The uncertainties are in the values of parameters and in the shapes of functions. We explore the usefulness of info-gap decision theory in patient-physician decision making in managing cholesterol level using clinical guidelines. Info-gap models of uncertainty provide versatile tools for quantifying diverse uncertainties. Info-gap theory provides two decision functions for evaluating alternative therapies. The robustness function assesses the confidence—in light of uncertainties—in attaining acceptable outcomes. The opportuneness function assesses the potential for better-than-anticipated outcomes. Both functions assist in forming preferences among alternatives. Hypothetical case studies demonstrate that decisions using the guidelines and based on best estimates of the expected utility are sometimes, but not always, consistent with robustness and opportuneness analyses. The info-gap analysis provides guidance when judgment suggests that a deviation from the guidelines would be productive. Finally, analysis of uncertainty can help resolve ambiguous situations.     

Implementing a discrete-event simulation software selection methodology for supporting decision making at Accenture

For large international companies with their own simulation team, it is often hard to make a decision related to selection of new discrete-event simulation software. This paper presents a comprehensive discrete-event simulation software selection methodology that has been successfully used for decision making at Accenture consulting company. Accenture already used a simulation tool at the start of the project, but wanted to find out whether the current tool used still was the most appropriate one for its needs, and to evaluate the latest discrete-event simulation tools. The developed methodology consists of two phases: phase 1 quickly reduces the long list to a short list of packages, and phase 2 matches the requirements of the company with the features of the simulation package in detail. Successful application of the proposed methodology indicates its possible application for decision making in other large organisations, provided that the study is performed by a third party to avoid risks of influencing the outcome of the selection process.     

Monte Carlo simulation techniques for group decision making with incomplete information

In this paper we deal with group decision-making problems where several decision makers elicit their own preferences separately. The decision makers’ preferences are quantified using a decision support system, which admits incomplete information concerning the decision makers’ responses to the questions they are asked. Consequently, each decision maker proposes classes of utility functions and attribute weight intervals for the different attributes. We introduce an approach based on Monte Carlo simulation techniques for aggregating decision maker preferences that could be the starting point for a negotiation process, if necessary. The negotiation process would basically involve the decision maker tightening the imprecise component utilities and weights to output more meaningful results and achieve a consensus alternative. We focus on how attribute weights and the component utilities associated with a consequence are randomly generated in the aggregation process taking into account the decision-makers’ preferences, i.e., their respective attribute weight intervals and classes of utility functions. Finally, an application to the evaluation of intervention strategies for restoring a radionuclide contaminated lake illustrates the usefulness and flexibility of this iterative process.     

基于Copula的VsR度量与事后检验

估计VaR的传统方法有三种:协方差矩阵法、历史模拟法和蒙特仁洛模拟法。通常,文献中认为刚蒙特卡洛模拟法度量VaR有很多方面的优点。但是,本文通过实证检验发现,使用传统蒙特卡洛模拟法估计的VaR偏小,事后检验效果很不理想。本文引入Copula函数来改进传统的蒙特卡洛模拟法。Copula函数能将单个边际分布和多元联合分布联系起来,能处理非正态的边际分布,并且它度量的相关性不再局限于线性相关性。实证检验表明,基于Copula的蒙特卡罗模拟法可以更加准确地度量资产组合的VaR。     

���ھ���ֲ������ĸ�ά��̬�Լ���

??Kolmogorov-Smirnov (KS), Cramer-von Mises (CM) and Anderson-Darling (AD) test, which are based on empirical distribution function (EDF), are well-known statistics in testing univariate normality. In this paper, we focus on the high dimensional case and propose a family of generalized EDF based statistics to test the high-dimensional normal distribution by reducing the dimension of the variable. Not only can we approximate the corresponding critical values of three statistics by Monte Carlo method, we also can investigate the approximate distributions of proposed statistics based on approximate formulas in univariate case under null hypothesis. The Monte Carlo simulation is carried out to demonstrate that the performance of proposed statistics is more competitive than existing methods under some alternative hypotheses. Finally, the proposed tests are applied to real data to illustrate their utility.     

Towards logistics systems parameter optimisation through the use of response surfaces

Logistics systems have to cope with uncertainties in demand, in lead times, in transport times, in availability of resources and in quality. Management decisions have to take these uncertainties into consideration. An evaluation of decisions may be done by means of simulation. However, not all stochastic phenomena are of equal importance. By design of simulation experiments and making use of response surfaces, the most important phenomena are detected and their influence on performance estimated. Once the influence of the phenomena is known, this knowledge may be used to determine the optimal values of some decision parameters. An illustration is given on how to use response surfaces in a real-world case. A model is built in a logistics modelling software. The decision parameters have to be optimised for a specific objective function. Experiments are run to estimate the response surface. The validity of the response surface with few observations is also tested.