Combining Experts’ Judgments: Comparison of Algorithmic Methods Using Synthetic Data

Expert judgment (or expert elicitation) is a formal process for eliciting judgments from subject‐matter experts about the value of a decision‐relevant quantity. Judgments in the form of subjective probability distributions are obtained from several experts, raising the question how best to combine information from multiple experts. A number of algorithmic approaches have been proposed, of which the most commonly employed is the equal‐weight combination (the average of the experts’ distributions). We evaluate the properties of five combination methods (equal‐weight, best‐expert, performance, frequentist, and copula) using simulated expert‐judgment data for which we know the process generating the experts’ distributions. We examine cases in which two well‐calibrated experts are of equal or unequal quality and their judgments are independent, positively or negatively dependent. In this setting, the copula, frequentist, and best‐expert approaches perform better and the equal‐weight combination method performs worse than the alternative approaches.     

综合集成研讨厅体系中专家群体行为的规范

首先扼要介绍了国内外在复杂系统、复杂性方面的研究情况 .然后 ,讨论了宏观经济决策问题 .由于宏观经济决策问题属于开放的复杂巨系统范畴 ,解决这一问题的有效方法是采用钱学森教授于 1 992年提出的“从定性到定量的综合集成研讨厅体系”的方法 ,它是将专家群体的定性认识综合后再用适当的建模方法处理 ,最终得出解决问题的结果 .研讨厅体系可以看作是由三部分组成 :群体专家体系、信息与知识体系、计算机软硬件技术 .在研讨厅体系中专家群体是一个非常重要的组成成分 ,专家个人和群体的行为直接影响着决策的结果 .专家本人及专家群体的思维和行为难免有一些不足 ,并且他们的行为也受个体、群体行为规律的支配 .本文通过分析个体、群体行为的缺陷 ,为参加研讨的专家制定一个行为规范 ,使他们克服人类行为的不足之处 ,结合先进的决策技术和工具以及建模方法 ,以得到更好的决策结果     

Using Expert Models in Human Reliability Analysis—A Dependence Assessment Method Based on Fuzzy Logic

In human reliability analysis (HRA), dependence analysis refers to assessing the influence of the failure of the operators to perform one task on the failure probabilities of subsequent tasks. A commonly used approach is the technique for human error rate prediction (THERP). The assessment of the dependence level in THERP is a highly subjective judgment based on general rules for the influence of five main factors. A frequently used alternative method extends the THERP model with decision trees. Such trees should increase the repeatability of the assessments but they simplify the relationships among the factors and the dependence level. Moreover, the basis for these simplifications and the resulting tree is difficult to trace. The aim of this work is a method for dependence assessment in HRA that captures the rules used by experts to assess dependence levels and incorporates this knowledge into an algorithm and software tool to be used by HRA analysts. A fuzzy expert system (FES) underlies the method. The method and the associated expert elicitation process are demonstrated with a working model. The expert rules are elicited systematically and converted into a traceable, explicit, and computable model. Anchor situations are provided as guidance for the HRA analyst's judgment of the input factors. The expert model and the FES‐based dependence assessment method make the expert rules accessible to the analyst in a usable and repeatable way, with an explicit and traceable basis.     

RETURN ON INVESTMENT AS A CRITERION FOR INVENTORY MODELS

In the past most inventory formulations have utilized cost minimization or profit maximization as an optimizing criterion. When viewed from the standpoint of the owner or investor, maximizing the return on investment (ROI) is an appropriate criterion for many types of inventories. This paper proposes ROI as a criterion for inventory models and derives optimal reorder rules for some common assumptions. An economic order quantity that differs greatly from the traditional formulas is discussed. The paper also enumerates the conditions under which ROI is an appropriate criterion and contrasts it to the traditional cost minimization and profit maximization criteria.     

基于信任关系和信息测度的概率语义社会网络群决策模型

设计专家权重和属性指标权重的计算模型已成为近年来备受关注的两个重要研究课题。针对评价信息为概率语义信任函数的社会网络群决策问题,提出一种基于信任关系和信息测度的概率语义社会网络群决策模型。首先,构建基于信任关系的概率语义决策空间,探究专家之间的信任传递模型,通过专家之间信任关系计算专家的权重;其次,引入概率语义信任函数的熵和相似度概念,并运用三角函数设计概率语义信任函数信息熵和相似度的衡量方法;最后,构建基于信任关系和信息测度的概率语义社会网络群决策模型,进而得到合理可靠的决策结果,同时将提出的社会网络群决策模型用于电动汽车供应商的选择实例,对比分析实验验证了模型的合理性和有效性。     

一种基于证据距离的客观权重确定方法

证据理论能较好地处理具有模糊和不确定信息的合成问题,广泛应用于群决策中。但在高度冲突的证据下,合成结果会失真,为解决该问题,提出一种基于证据距离的客观权重确定方法。该方法根据各专家提供的证据间的距离来确定其客观权重,从而重新调整各专家证据的基本概率分配,可以降低与群体意见分歧较大的专家的权重。最后用一项目投资选择案例来验证该方法有效性。     

动态联盟组建与运作中的专家支持

本文针对动态联盟组建与运作中的半结构化和非结构化问题,探讨其决策的专家支持,包括专家选择、调查表设计、专家评分结果处理、专家系统构建,以更有效地组建与运作动态联盟。     

Performance measurement of supply chain management using the analytical hierarchy process

Performance measurement of supply chain management (SCM) is a rapidly growing multi-criteria decision-making problem owing to the large number of factors affecting decision-making. The right choice of performance metrics and measures is critical to the success and competitiveness of the firms in the era of globalisation. Recognising the multiple objective nature of the problem, this paper proposes the use of the analytical hierarchy process (AHP) methodology as aid in making SCM evaluation decisions. For pair-wise comparison in AHP, a survey methodology is used. The methodology presented can help firms to prioritise and formulate viable performance measurement strategies in the volatile and complex global decision environment from different balanced scorecard (BSC) perspectives. A demonstration of the application of this methodology in a real life problem is presented.     

授权决策的期权特征与模型解析

本文利用实物期权方法对授权决策的核心问题——授权时机和授权对象的选择——进行了研究。首先分析了授权决策的期权特征,对授权决策中存在的实物期权进行了总结;然后构建了授权决策的实物期权模型,分别计算出了授权决策的期权价值和授权时员工对公司的人力资本价值,由此得出了一个可比性的授权决策标准,即临界业绩水平,从而简化了决策指标,提高了信息的利用率,使企业可以对特质各异的候选人进行准确比较,并分三种授权决策情形分别给出了决策方案,最后基于数值算例作出了进一步的解析。     

Concepts,Theory, and Techniques DISTRIBUTED COMPUTER SYSTEM DESIGN: A MULTICRITERIA DECISION-MAKING METHODOLOGY*

The design of distributed computer systems (DCSs) requires compromise among several conflicting objectives. For instance, high system availability conflicts with low cost which in turn conflicts with quick response time. This paper presents an approach, based on multi-criteria decision-making techniques, to arrive at a good design in this multiobjective environment. An interactive procedure is developed to support the decision making of system designers. Starting from an initial solution, the procedure presents a sequence of non-dominated vectors to designers, allowing them to explore systematically alternative possibilities on the path to a final design. The model user has control over trade-offs among different design objectives. This paper focuses on the details of the mathematical model used to provide decision support. Accordingly, a formulation of DCS design as a multicriteria decision problem is developed. The exchange search heuristic used to generate nondominated solutions also is presented. We argue that multicriteria models provide a more realistic formulation of the DCS design problem than the single-criterion models used widely in the literature. While obtaining a clear definition of design objectives (single or multiple) is an important activity, by explicitly acknowledging the trade-offs among multiple objectives in the design process, our methodology is more likely to produce a better overall design than methods addressing a single criterion in isolation.     

Selecting IT applications in manufacturing: a KBS approach

The use of the right type of Information Technology (IT) applications or manufacturing systems is expected to usher in a competitive advantage. Selection of the right type of IT application is, however, a challenging task. When a company, with a given dominant process structure, emphasizes two or more competitive priorities, such as quality, product flexibility, etc., an unaided manager faces a complex decision problem in choosing from alternative IT applications available in the areas of product design through distribution. In this paper, we present a Knowledge Based System (KBS) that would assist managers with the identification of IT applications that are consistent with both the competitive priorities and the process structure. Validation of the system illustrates that its performance is consistent with the human experts, and it has the potential to facilitate effective and swift decision-making in the selection of appropriate IT applications that best match an organization's manufacturing strategy.     

Notes and Communications “A DECISION THEORY EXAMPLE IN FOOTBALL”: A COMMENT*

Janssen and Daniel analyzed the choice between a one- or a two-point conversion for a particular game situation in college football. Their decision criteria was maximum expected utility based on a von Neumann-Morgenstern utility function defined over the games outcomes. An alternative approach based on a stochastic dominance criterion is presented that does not rely on knowledge of the relative importance of tying vs. winning; rather, it relies on a notion of consistency in the sequential problem.     

PREDICTING STOCK MARKET BEHAVIOR THROUGH RULE INDUCTION: AN APPLICATION OF THE LEARNING-FROM-EXAMPLE APPROACH*

An artificial intelligence-based rule-induction approach to the analysis of stock market prediction is presented. A single investment analyst was used as the expert for this study. Predicting intermediate fluctuations in the movement of the market for nonconservative investors was selected as the decision to analyze. Commercially available rule-induction software was used to generate rules that predicted the market calls of the market analyst and the actual movements of the market. Rules predicting actual market movement performed better than rules predicting the analyst's calls and better than the analyst himself. Such an approach may prove useful in designing a decision support system for market analysts or in improving the decision-making processes of such analysts. The dynamic nature of the stock market represents a substantially different decision environment than those previously analyzed by learning-from-example (LFE) techniques. This study provides insights into the limits and applications of LFE approaches.     

AN EXPERT SYSTEM APPLICATION FOR DECISION SUPPORT IN LAW ENFORCEMENT

Expert systems offer promise for decision-making support in stressful circumstances such as those that occur when law enforcement officials respond to hostage-taking incidents. These are life-or-death situations in which the costs of decision error are enormous. This paper reports on an expert system being developed and tested to aid police decision makers in hostage-taking incidents. Knowledge is represented by rules embedded in an incident-based decision-tree network. Four different decision makers who have similar but distinct information needs are supported and their efforts are coordinated by the system, which must function in real time as the crisis unfolds. Clear, concise, highly focused dialogue is required at the user interface since frequently only seconds are available in which to make a decision and implement it.     

Linking IT Applications with Manufacturing Strategy: An Intelligent Decision Support System Approach

Research has indicated the importance of matching Information Technology (IT) applications or manufacturing systems with the competitive strategy of a company. Selection of the right type of IT application is, however, a challenging task. When a company with a given dominant process structure emphasizes two or more competitive priorities, such as quality, product flexibility, etc., an unaided manager faces a complex decision problem in choosing from alternative IT applications available in the areas of product design through distribution. In this paper, we develop an Intelligent Decision Support System (IDSS) to assist managers with assessment of the relative importance of competitive priorities in their organization, evaluation of the fit between the competitive priorities and their dominant process structure, and identification of the IT applications that are consistent with both the competitive priorities and the process structure. The IDSS is comprised of an interactive user interface, a knowledge database, a decision model, and a Knowledge-Based System (KBS) that was developed using the 1st Class KBS shell. Validation of the system illustrates that its performance is as good as the human expert, and it has the potential to facilitate effective and swift decision making in the selection of appropriate IT applications that best match an organization's manufacturing strategy. The choice and use of the right type of IT application should provide a company with a competitive edge.     

基于价值-风险双准则的风险决策理论

风险决策问题一直是管理学、心理学和经济学十分关注的重要研究领域。本文对重要的风险决策理论进行回顾,根据已有的研究成果,分析并提出本文研究的理论基础：多重参照点、风险认知和决策多准则。本文将个体决策的参照点归纳为期望参照点和厌恶参照点,并将风险界定为决策事件的未来可能状态与期望参照点发生负偏离的综合,即价值的负偏离和概率的负偏离。在此基础上,论文提出了价值最大化-风险最小化双准则的风险决策理论,给出风险测算方法,及基于期望价值和风险的综合决策值公式。最后,应用该决策理论对经典决策悖论进行了消解,并应用典型的决策实验数据对理论进行验证,结果表明本文所提出的风险决策理论具有很强的解释性。     

区间概率条件下属性值为不确定语言变量的风险型多属性决策研究

针对区间概率条件下属性值为不确定语言信息且属性权重未知的风险型多属性决策问题,提出了一种基于概率理论和不确定语言变量的TOPSIS决策方法。首先建立了区间概率转化为点概率的数学模型,通过期望值将风险型决策矩阵转化为确定型矩阵;然后利用方案与理想解越近方案越优,与负理想解越远方案越优的原则建立属性权重确定模型,并利用TOPSIS方法的相对优属度大小确定方案的排序;最后通过应用案例说明了本方法的决策步骤。     

基于社交网络分析的区间二元语义群决策方法研究

通过基于Steiner点的区间二元语义集结方法,将评价信息通过相关转化规则量化为对应的二维坐标系中的坐标点集,运用植物模拟生长算法获取对应区间点集的Steiner点(专家群体最优结集点,即群体共识点),按照逆映射关系还原为二元语义集结信息,从而提出一种基于社交网络分析和专家自信偏好关系的区间二元语义群决策问题的新方法。结合专家自信程度系数和社交网络结构中的影响力调整专家主观权重,结合专家相对重要系数与群体相似度系数获取客观权重,最终确定专家综合权重,并对方案进行择优排序。通过算例分析说明方法的可行性和有效性。     

面向专家的知识库优化

知识库的质量是影响智能系统性能的主要因素 ,而知识获取一直是设计智能系统的瓶颈问题 ,这是由于目前人类认识的局限性 ,导致知识工程师和专家之间的不协调关系造成的 .为克服上述不利局面 ,本文利用粗糙集等理论 ,得到含有噪声的初始知识库 ,然后采用遗传算法、可视化技术和知识校验等技术对规则库和案例库进行了优化 .从而在知识获取过程中建立了知识工程师和专家之间的新型的关系 ,其中专家处于中心地位 ,知识工程师只是起辅助作用 ,即整个知识获取过程是面向专家的