A multiperspective systems-based framework for decision support systems design

This paper reviews the dominant approaches to the design of decision support systems with reference to recent developments in information technology. The complexities and difficulties of the various developments are also discussed and their implications explored. A multiperspective, systems-based framework is presented which aims to lessen these difficulties. The framework is based on the theory and concepts of critical systems thinking.     

Validation for decision support systems: Recent developments and findings

This paper reviews earlier work on the validation of mathematical models and modeling within Decision Support Systems (DSS), in particular where the DSS were developed to support operations management. The authors found that little validation was carried out in practice. With this work as a base, reasons for the paucity of validation are advanced and comparisons with DSS developed to support marketing management are made. The case for a more appropriate methodology of validation is advanced.     

Multiple-criteria decision support systems: The powerful tool for attacking complex,unstructured decisions

The purpose of this research is to introduce the basic concept of multiple-criteria decision support systems (MCDSS) and to provide a general framework for building an effective MCDSS, a powerful tool for managing complex, unstructured decisions. Although various techniques for multiple-criteria decision making (MCDM) have been developed, this paper investigates the functions and properties of goal programming model-based MCDSS, the role of decision makers and MCDM models in the decision-making process, and the dynamic interaction between the model and the decision maker. The potential benefits of the MCDSS are also explained.     

固定收益决策支持系统机理建模与数据挖掘的协同研究

以固定收益决策支持系统(FIDSS)的机理建模和数据挖掘的协同研究为重点,对FIDSS的结构分层设计、协同联结模式与协同作业功能进行了分析,对FIDSS的确定性关联、选择性关联和动态关联等三个层次上的机理建模与数据挖掘的交互方式进行了建构,以便进一步提高FIDSS的决策质量和对金融经济复杂系统管理的支持水平.     

Method for designing organization decision support system framework

1. INTRODUCTION Decision Support System(DSS) was presented by Scott Morton in his paper entitled as “Management Decision Systems” in the early 1970’s. Until the 1980’s, DSS had become a new interdiscipline involved computer science, operational research and managerial science. The definition for DSS presented by Gorry and Scott Morton is that DSS combines personal resources with computer’s abilities so as to improve the quality of decisions. And it is an assistant system bas…     

决策支持系统中的模糊神经网络研究

面对复杂系统中许多需要决策选择的对象和方案难以满足人们对决策精度的要求,论述了模糊神经网络的概念及特点,在此基础上提出了基于模糊神经网络的决策支持模型。介绍了决策支持模型运行计算的过程与方法,并通过农机选型配套的决策支持实例验证了该模型,说明该模型对复杂的决策问题具有简单方便、快速准确地作出反应的优点。     

Study of component-ware & application software framework based decision support system development

1.INTRODUCTION Thedevelopmentofthedecisionsupportsystem (DSS)wasanarduousandcomplicatedwork.Asa consequence,software reusewasextremelyempha sizedinthecourseofDSSdevelopment.Traditional methodemploysDSSgeneration(DSSG)tobuilda targetDSSsystem.Itwasin1982thatSpraguepro posedafamousdevelopmentflowofDSSasFig.1. NowadaysitisstillinfashioninDSSdevelop ment[1,2]. IntraditionalDSSdevelopmentprocess,itwas abovealltobreakdowntherequirementQintothe followingfunctionorderedpairs…     

一种新的基于Web的开放式决策支持系统模型

给出了基于 Web的开放式决策支持系统的形式规范 ,更加明确了系统中的各种实体及其相互关系 ,在此基础上提出了一种新的基于 Web的开放式决策支持系统模型 ,并详细分析了系统运作的事务处理过程 ,最后用教育规划决策支持系统实例说明了这种模型的合理性和实用性.     

基于多agent系统的舰艇编队防空辅助决策系统

提出基于多agent系统(multi-agent system,MAS)技术的分布式编队防空辅助决策系统基本框架。在目标信息共享和平台间实时作战参数交换的前提下建立了基于平等协商的分布式目标协商算法,混合运用Ta-bu搜索算法和基于规则的计划算法。在目标分配协商中,采用~Brown算法通过agent之间平等协商进行目标分配。在本舰防空计划中,采取软/硬防空计划并行比较消解冲突,实现本舰软/硬防空武器防空计划的协调。     

村庄布局决策支持系统研究

介绍了县级政府进行平原地区村庄规划布局的决策支持系统的设计,该系统运用系统工程的思想,以决策支持的理论为核心,把整个决策过程分成若干步骤,每为个步骤提供若干基本模型,采用定性和定量相结合的决策方法,用０－１目标规划和知识推理两种方法科学合理的村庄搬迁合并方案,。软件把数学模和地理信息技术相结合,大大增强了决策支持系统的功能。     

Knowledge-based decision support system for tool management in flexible manufacturing system

Tool management is not a single, simple activity, it is comprised of a complex set of functions, especially in a flexible manufacturing system (FMS) environment. The issues associated with tool management include tool requirement planning, tool real-time scheduling, tool crib management, tool inventory control, tool fault diagnosis, tool tracking and tool monitoring. In order to make tools flow into/out of FMS efficiently, this work is aimed to design a knowledge-based decision support system (KBDSS) for tool management in FMS. Firstly an overview of tool management functions is described. Then the structure of KBDSS for tool management and the essential agents in the design of KBDSS are presented. Finally the individual agents of KBDSS are discussed for design and development.     

基于Agent的农业经济智能决策支持系统研究

将分布式人工智能中的Agent技术融于现代农业经济管理决策支持系统的研究与开发中,提出了基于Agent的农业经济智能决策支持系统(AEIDSS)的结构和组成,引入界面Agent、管理Agent、数据Agent、模型Agent、知识Agent以及各类决策职能Agent构成系统的主要部件,分析了基于分类器系统的Agent实现方法,探讨了在多Agent通讯与合作环境下实现农业经济系统的动态分析、评估、预测以及优化等决策支持功能的工作原理。图2,参13。     

结合Agent技术的飞行任务规划GDSS研究

如何更好地利用领域专家和计算机软系统相互协调实现飞行任务规划成为近年来对飞行任务规划研究的一个热点。在飞行任务规划系统功能、结构和特点分析的基础上,借鉴了群决策支持系统(GDSS)理论和人工智能领域的Agent技术,将飞行任务规划系统设计成以各个领域专家为核心,利用Agent定义的全新飞行任务规划群决策支持系统,并给出其描述模型。该设计以具有领域专家为核心,各个Agent相对独立,系统结构灵活等特点。     

发电市场不同决策规则三寡头博弈模型研究

在三寡头垄断的电力市场中假设生产者分别采用静态预期及有限理性预期规则,建立了一个基于不同决策规则的动态博弈模型.对该模型进行了较详细的理论分析和参数仿真计算,在一定范围内得出了系统出现混沌的参数值,分析了不同参数组合条件下系统表现出来的分岔,混沌和密度周期性等复杂性态.所得结果对于一般寡头电力市场中的生产者制定产量计划具有一定启示作用.     

Perspectives on executive information systems

Executives who consider the acquisition of Executive Information Systems (EIS) may encounter several major problems already in their selection and introduction stages. Key issues are the following: Can the EIS's really match our managerial needs and support decision making in our organizational contingencies? How widely in the leadership levels can they be introduced? Are they easy-to-learn, easy-to-use and easily customizable? Do they utilize the newest information technology, e.g. graphics and modern interface technology? Can they work in our computer network? In order to facilitate elegant solutions of the selection problems, we construct an evaluation framework which guides the selection of an EIS product to match the managerial needs according to the organizational contingencies. This framework is useful immediately after the specification of the decision making support requirements of the executive. The effectiveness of the framework is shown by applying it and analyzing 13 EIS products. We also offer introduction and use experiences from four English and four Finnish large companies covering the opinions of 132 EIS users. In this paper we give general answers to the questions above and discuss other relevant issues.     

基于分离BDD的通用多阶段任务系统可靠性分析

静态的通用多阶段任务系统的不完全覆盖、跨阶段依赖性极大地增加系统可靠性分析的复杂性和难度.基于SEA思想给出两级分离的组合方法将任务和阶段的不完全覆盖分割, 通过新增的阶段代数规则和向后阶段依赖运算式处理跨阶段依赖性,并且充分利用高效的多阶段系统双值决策图,实现模型规模和运算的简化.最后,通过案例分析, 将此分离BDD方法与三值决策图和多状态双值决策图对比, 验证其精确、高效和模块化的优点.     

Establishing formal state space models via quantization forquantum control systems

1.INTRODUCTION Quantumtheoryisoneofthemostimportantachieve mentsinthe20thcentury,anditdisclosesthestruc tures,characteristicsandmotionlawsofmatterinthe microcosmicfieldandleadsusfromthemacroscopical worldintomicrosystems.Aseriesofnewphenomena whicharedifferentfromclassicalsystemshavebeen discovered,suchasquantumentanglement,quantum coherenceanduncertainty.Meanwhile,quantum theoryandquantummethodsarealsowidelyusedin differentfieldssuchaschemicalreaction,geneticen gineering,atomicphysicsandi…     

Parameter selection of support vector machine for function approximation based on chaos optimization

The support vector machine （SVM） is a novel machine learning method, which has the ability to approximate nonlinear functions with arbitrary accuracy. Setting parameters well is very crucial for SVM learning results and generalization ability, and now there is no systematic, general method for parameter selection. In this article, the SVM parameter selection for function approximation is regarded as a compound optimization problem and a mutative scale chaos optimization algorithm is employed to search for optimal paraxneter values. The chaos optimization algorithm is an effective way for global optimal and the mutative scale chaos algorithm could improve the search efficiency and accuracy. Several simulation examples show the sensitivity of the SVM parameters and demonstrate the superiority of this proposed method for nonlinear function approximation.     

基于决策树的快速SVM分类方法

为提高支持向量机（support vector machine, SVM)算法对大规模数据的适应能力,加快SVM算法的分类速度,提出一种基于决策树的快速SVM分类方法。该方法的重点在于构建一棵决策树,将大规模问题分解为相对简单的子问题,树中节点由线性支持向量机组成,每个节点包含一个决策超平面,分类过程取决于节点的数量。此方法在分类复杂样本时避免了使用非线性核函数。并且由于使用线性核函数,则不用进行模型选择,进一步加快了样本的分类速度。实验表明,针对大规模多特征数据的非线性分类问题,该方法比传统方法具有更高的速度。