Integrating Neural Networks and Semi-Markov Processes for Automated Knowledge Acquisition: An Application to Real-time Scheduling*

Recently, artificial neural networks (ANN) have gained attention as a promising modeling tool for building intelligent systems. A number of applications have been reported in areas varying from pattern recognition to bankruptcy prediction. In this paper, we present a creative methodology that integrates computer simulation, semi-Markov optimization, and ANN techniques for automated knowledge acquisition in real-time scheduling. The integrated approach focuses on the synergy between operations research and ANN in eliciting human knowledge, filtering inconsistent data, and building competent models capable of performing at the expert level. The new approach includes three main components. First, computer simulation is used to collect expert decisions. This step allows expert knowledge to be obtained in a non-intrusive way and minimizes the difficulties involved in interviewing experts, constructing repertory grids, or using other similar structures required for manual knowledge acquisition. The data collected from computer simulation are then optimized using a semi-Markov decision model to remove data redundancies, inconsistencies, and errors. Finally, the optimized data are used to build ANN-based expert systems. The integrated approach is evaluated by comparing it with the human expert and using ANN alone in the domain of real-time scheduling. The results indicate that ANN-based systems perform worse than human experts from whom the data were collected, but the integrated approach outperforms human experts and ANN models alone.     

Performance measurement with classification information: an enhanced additive DEA model

This paper presents an improved measurement tool for evaluating performance of branches within a major Canadian bank. While there have been numerous previous studies of performance in the banking industry, particularly at the branch level, this study is different in a very significant way: specifically two kinds of data are used to develop the model. The first type of data is that related to standard transactions, available from any bank; such have formed the basis of numerous previous studies. The second type of data, obtained from the site studied, is classification information, based on branch consultant/expert judgment as to good and poor performance of branches. The purpose herein is to present a modified version of an existing benchmarking model, data envelopment analysis (DEA), and to show how this tool is applied in the banking industry. The mechanism used herein to incorporate expert knowledge within the DEA framework is to first apply a discriminant or classification tool, to quantify the functional relation that best captures the expert's mental model for performance. The outcome of this first phase is an orientation of variables to aid in the definition of inputs and outputs. The resulting orientation then defines the DEA model that makes up the second phase of the model.     

企业隐性知识导航方法研究

在对企业隐性知识特性及其研究现状分析的基础上,提出了企业隐性知识导航方法。该方法包括2个方面内容:基于平衡计分卡的专家导航方法,充分考虑了专家之间的能力差别;基于多关系专家网络的专家导航路径查询方法和相关专家查询方法,充分考虑了专家之间的多重关系,该方法为管理者发现及考核专家提供了依据。在此基础上,利用社会网络分析工具Sociom etryP ro对专家网络进行评价,得到了专家个人特征结果,并指出了改善企业隐性知识共享的具体措施。     

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.     

在线负面口碑处理的专家识别方法研究

现有专家识别大多建立在专家知识能力的基础上，然而在负面口碑处理的专家识别中，仅考虑知识能力并不能满足各主体的价值需求。本文从资源角度理解专家识别，建立了专家识别资源映射框架，并依据此框架分析了专家识别的显性资源映射和隐性资源映射过程。除了考虑直接体现用户专业水平的知识能力，还考虑了专家参与的情感抚慰能力和互动程度能力，以此构建的人工神经网络模型在实验中表现出了良好的性能。     

Automating Knowledge Acquisition and Refinement for Decision Support: A Connectionist Inductive Inference Model*

An important application of expert systems technology is to provide support for nonstructured decision making. Usually, nonstructured decision making is characterized by heavy reliance on heuristic knowledge, which is very difficult to articulate or document, and therefore traditional knowledge acquisition approaches are not very successful. The quality and effectiveness of an expert system supporting unstructured decision making is affected when traditional knowledge acquisition approaches are used. To alleviate this problem a model is proposed that combines inductive inference and neural network computing, and an example is presented that illustrates the potential of this model in unstructured decision support.     

Automated Discovery in Managerial Problem Formulation: Formation of Causal Hypotheses for Cognitive Mapping

Development of knowledge acquisition techniques known as automated discovery systems has occurred in deep and narrow domains of knowledge. Automated discovery is the generation of new knowledge by a computer system on its own, without the help of another knowledge source. This paper describes research and validation of an automated discovery system for a wide and shallow domain—business management. The system continues recent advances in expert systems research which have enhanced cognitive mapping, a problem formulation tool. The system perceives the behavior of distal variables in the environment through probabilistic cues-to-causality, and generates previously unknown hypotheses by aggregating the probabilities into a single criterion of causal relatedness. The system is validated against the source code of a simulated managerial environment, and causal relationships posited by decision makers experienced in the play of the gaming simulator.     

A Structured Literature Review for Risk Assessment: EMF and Human Health Risk

The volume of publications on any given topic makes it difficult to select an appropriate subset of publications for review by an expert panel. This paper proposes a method for selecting a subset using a protocol that attempts to order publications based upon good research methods. The protocol was followed by graduate students with no expert knowledge of the area they were reviewing. They reduced 11730 articles on the topic of EMF and Health risk to 68 articles that were reviewed by an expert panel. The interrater reliability was 96% and 94% of the relevant articles were captured by this process.     

An Application of the AXIS Solution Framework to Multiple Objective Aggregate Production Planning

This paper provides data on the first application of a prototype of the AXIS solution framework. AXIS (algorithms combined with knowledge systems in an interactive sequence) is a framework for interactively combining structured algorithms that seek a best solution with knowledge-based expert systems that seek expert heuristic solutions. This paper tests the framework using an interactive multiple objective integer programming algorithm combined with heuristics taken from the domain of aggregate production planning. The results indicate the AXIS framework can be successful in generating high quality solutions, in vastly reduced solution times compared to the structured algorithms, at much lower costs compared to the expert heuristics working alone.     

Expert knowledge,expert systems and commercial interests

Experience to date in building expert systems has brought a general realization of the complexity of the effort required for producing systems capable of operating on ‘real-world’ problems. The most successful knowledge-based/expert systems built to date, e.g. DENDRAL, MYCIN, PROSPECTOR and R1 have demonstrated that a considerable investment in time and dedication on the part of systems designers and experts is required to create a fully operational system. The great majority of development projects attempting to harness this new programming technology do not extend beyond creation of a prototype system largely due to lack of necessary resources, i.e. time, money and know-how. The purpose of this paper is to discuss some of the more important issues associated with development of expert systems and to provide an overview of the commercial and industrial efforts of putting expert systems to work. This paper should be of interest to those who have gained their knowledge of expert systems from the reading of non-specialist publications and thus may have been exposed to somewhat over enthusiastic accounts of the subject.     

决策单元交叉效率的自适应群评价方法

传统交叉效率评价方法因决策单元偏好权重不唯一而难以操作,因交叉效率有效性分值平均化集结而难以被接受。目前的学者通常围绕决策单元指标权重的确定性分配方法、交叉效率有效性分值的去平均化集结等分别开展研究。本文将交叉效率评价方法中自评互评相结合的评价模式看作群决策过程,即每个决策单元既是一个被评对象,又是一个决策"专家",提出了一种决策单元交叉效率的自适应群评价方法,将决策单元偏好权重的确定和交叉效率有效性分值的去平均化集结作为同一个决策过程,根据每个决策单元的评价结果与群体评价结果的接近程度,同步迭代调整决策单元的"专家"权重和决策单元自评产生的、并提供给其他被评价决策单元的一组确定的偏好指标权重。实验验证与实例运用分析表明,该方法收敛效果良好,能得到客观稳定的决策单元交叉效率有效性分值及排序。     

State-of-the-Art Developments in Expert Systems and Strategic Marketing Planning

This paper describes a case history of the development of an Expert System in Strategic Marketing Planning codenamed EXMAR. It traces the evolution of the system from the formation of the DTI club two years ago to the launch of the prototype model. The paper outlines the technical and domain–specific obstacles encountered en route and how these were overcome. A number of conclusions are drawn from the project. The principal one is that there is a bright future for expert systems in the field of strategic management. Professor Malcolm McDonald is the principal expert to the club. Hugh Wilson is a senior consultant with Artificial Intelligence Ltd. He was also the knowledge engineer and the project manager. For readers who would like a simple explanation of Expert Systems, please refer to Professor McDonald's paper ‘Marketing Planning and Expert Systems: an Epistemology of Practice’, Marketing Intelligence and Planning, 7, 7/8, 1989.     

Efficiency and effectiveness of knowledge management systems in SMEs

This paper proposes a 3D fuzzy logic methodology to assess the efficiency and effectiveness of knowledge management systems (KMSs) adopted by small and medium enterprises (SMEs) and identifies a classification bringing together the behaviour of SMEs when adopting KMSs. The proposed methodology was previously tested on an individual SME and then implemented on a sample of SMEs. The results highlight that there is an ample variety of behaviours related to the nature of knowledge and the KMSs used. Specifically, four typologies of behaviour are identified: the efficient and effective SME, the effective but inefficient SME, the efficient but ineffective SME, and the inefficient and ineffective SME. The results show that only 18% of surveyed SMEs are efficient and effective for both KM-Tools and KM-Practices, the remaining 82% are inefficient or ineffective for KM-Tools and/or KM-Practices. Lastly, the paper shows how this methodology may be used by an SME as a managerial tool to suggest appropriate changes to improve the efficient and effective adoption of KMSs supporting SMEs in the knowledge management processes.     

Conceptualizing Structurable Tasks in the Development of Knowledge-Based Systems*

Conventional approaches to knowledge-based system (KBS) development are not appropriate for building KBSs when the application task is structurable (i.e., exhibits a certain degree of ill structure). Building a KBS for structurable tasks requires an understanding of the problem-solving strategies used by an expert to manage the ill structure, while at the same time relying on domain theories to understand the structured parts of the task. This paper presents a methodology for developing a knowledge model for structurable tasks during the conceptualization stage of KBS development. This is equivalent to building a logical model for design during the development of conventional information systems. The methodology relies on prior research on the decomposition and characterization of a task based on its various attributes. The paper also illustrates the use of the methodology in the case of KBS development for financial hedging. The paper concludes with some observations about the potential impact of this methodology on other stages in the KBS development process.     

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     

企业通过联盟进行隐性知识转移的三阶段模型

本文针对隐性知识的特征及知识在联盟中转移的过程和特点,建立了企业通过联盟进行隐性知识转移的微分动力学模型,在此基础上,进一步分析了影响联盟隐性知识转移的渐近解的主要参数控制.模型揭示了联盟中隐性知识转移的规律,指出了企业获得联盟中隐性知识的条件及隐性知识在联盟企业中传播的条件,同时也阐明了一些能改进隐性知识转移效果的途径,文中最后还分析了模型对联盟中知识管理的现实意义.     

Incentive for Peer‐to‐Peer Knowledge Sharing among Farmers in Developing Economies

This study examines the peer‐to‐Peer interactions among farmers when both knowledge learning and sharing are available. We construct a stylized model in which heterogeneous farmers are endowed with their initial production capabilities and can post questions in the platform for help. A representative expert regularly monitors the forum and provides answers to the farmers’ questions, but may be non‐responsive sometimes due to the limited capacity. A knowledgeable core user (farmer) can choose to be silent or responsive, and is allowed to strategically determine the informativeness of her answers. The farmers face the minimum quantity restriction for attracting the buyers, and must make production before the time of sales. We show that in equilibrium the core user never provides answers that are more informative than the expert's, irrespective of her ex ante knowledge level. Redesigning or restructuring the platform does not help eliminate this inefficient knowledge provision. We also find that hiring more staff to frequently monitor the forum turns out to be detrimental for the peer‐to‐peer interactions. Moreover, the competition on knowledge sharing between the platform expert and the core user features strategic complementarity sometimes but strategic substitution at other times. Third, charging for the platform usage may discourage uninformative answers, but it could also discourage the core user from sharing knowledge with other farmers.     

A decision-based approach: application to the automatic design of process planning

The notion of centre of decision presented in this article defines a conceptual and generic model enabling us to represent and use expert knowledge. It is suitable when the problem to be solved can be broken down into a sum of interdependent subproblems. Actions can be planned progressively and previous decisions can he modified by a backtrack mechanism. This principle is iterative. The structure is made up of many independent rules and an action plan. The customization of the application is possible even if the software evolves with future updates. This model has been applied in the LURPA-TOUR software which is a computer-aided process planning system for rotational parts.     

Using Bayesian Networks to Model Expected and Unexpected Operational Losses

This report describes the use of Bayesian networks (BNs) to model statistical loss distributions in financial operational risk scenarios. Its focus is on modeling "long" tail, or unexpected, loss events using mixtures of appropriate loss frequency and severity distributions where these mixtures are conditioned on causal variables that model the capability or effectiveness of the underlying controls process. The use of causal modeling is discussed from the perspective of exploiting local expertise about process reliability and formally connecting this knowledge to actual or hypothetical statistical phenomena resulting from the process. This brings the benefit of supplementing sparse data with expert judgment and transforming qualitative knowledge about the process into quantitative predictions. We conclude that BNs can help combine qualitative data from experts and quantitative data from historical loss databases in a principled way and as such they go some way in meeting the requirements of the draft Basel II Accord (Basel, 2004) for an advanced measurement approach (AMA).