语义决策过程支撑环境及其语义表示方法

针对决策支持系统缺乏有效的决策过程控制手段的问题,提出了语义决策及其过程支撑环境（ＤＰＳＥ）的概念．ＤＰＳＥ 利用决策问题语义指导系统方案,以决策过程中行为为应激式调度,实现决策支持系统运行全过程中语义形式化．ＤＰＳＥ 将决策环境分为需求组织、资源分配和行为规划３个方面,并定义了语义决策过程中动态语义的表示方法,提供了语义封装的方法及其过程支撑相关的算法．最后给出了系统原型,并通过实验表明ＤＰＳＥ 的有效性．

     

Eiffel语言的语义

本文采用类ＶＤＭ的指称语义技术为Ｅｉｆｆｅｌ（１９８８）语言建立了形式语义模型．该模型首先为Ｅｉｆｆｅｌ语言定义了对象模型和两个语义环境（动态和静态环境）,然后讨论Ｅｉｆｆｅｌ例程的语义．为了描述Ｅｉｆｆｅｌ的意外处理,我们采用了ＶＤＭ的“出口”机制．     

应用模糊集和神经网络的城市评价决策支持系统

本文以决策支持系统（ＤＳＳ）的理论方法为基础，综合运用模糊集（ＦＳ）和神经网络（ＮＮ）理论，提出了城市评价决策支持系统（ＣＥＤＳＳ）的基本结构模式，介绍了ＣＥＤＳＳ的基本功能，并对ＣＥＤＳＳ中评价指标的规范化方法、基于ＦＳ的多层次多目标综合评价模型、基于ＮＮ的系统自学习模型以及ＣＥＤＳＳ的工作原理作了描述。整个ＣＥＤＳＳ的原型系统已在微型机上实现，它可以较好地处理需要定性分析与定量分析相结合的决策问题，从而为城市管理决策者提供从信息、咨询到评价、决策、政策制定等一系列的智能支持     

智能化财务DSS系统设计

财务ＤＳＳ是面向财务、支持管理会计和决策会计，进行企业财务决策的专用决策支持系统．本文重点讨论专用财务ＤＳＳ的主要功能、软件结构及其基本部件的具体设计。包括：财务ＤＳＳ的数据库系统设计；财务ＤＳＳ的模型库系统设计；财务ＤＳＳ的知识库系统设计；财务ＤＳＳ的会话系统设计。     

基于数据库的大型语义网的表示,推理及应用

本文以可扩充的一体化词表概念语义网为背景，分析了语义网络向关系数据库的转换问题及关知识推理的实现技术，并讨论了此方法在智能化大型图书馆管理与决策支持系统ＬＩＭＤＳ中的应用。     

DOSFG中的语义子系统

分布式ＯＳ形成化生成系统模型主要由文法部分ＤＯＳＦＳＧ和语义子系统ＤＯＳＦＳＳ组成。其中ＤＯＳＦＳＧ采用了上下文无关文法。ＤＯＳＦＳＳ则是一个带操作集的语义子系统。     

煤巷锚杆支护设计专家决策支持系统

采用ＶＰ－ＥＸＰＥＲＴ和ＬＯＴＵＳ１－２－３分别构建专家系统（ＥＳ）和决策支持系统（ＤＳＳ）,并用ＤＯＳ批处理命令及ＶＰＢＣＡＬＬ将ＥＳ和ＤＳＳ连接建立成专家决策支持系统（ＥＤＳＳ）,可对煤巷锚杆支护设计进行咨询     

模糊协调决策的理论研究及其智能化计算机辅助系统的设计

本文针对多因素决策问题，提出了一种多专家协调决策方法，给出了该方法的原理、计算过程和应用实例，设计了相应的智能化计算机辅助系统－多用途决策支持系统（ＭＵＤＳＳ）。     

关于Herbrand基上的语义映射

关于Ｈｅｒｂｒａｎｄ基上的语义映射应明生（江西师范大学数学系南昌３３００２７）ＳＥＭＡＮＴＩＣＭＡＰＰＩＮＧＳＯＮＨＥＲＢＲＡＮＤＢＡＳＥＳ￥ＹＩＮＧＭｉｎｇｓｈｅｎｇ（ＤｅｐａｒｔｍｅｎｔｏｆＭａｔｈｅｍａｔｉｃｓ，ＪｉａｎｇｘｉＮｏｒｍａｌＵｎｉ...     

基于知识的多专家决策支持系统

由于决策支持系统（ＤＳＳ）和专家系统（ＥＳ）的不同行为特征,使它们在信息处理中的决策和问题求解方面不能相互兼顾。本文交讨论和ＥＳ结合的可行性,提出一种基于知识的多专家决策支持系统的结构模式,及实现多专家协同求解和决策的方法。     

复杂环境下的平行决策研究

针对复杂环境下决策的不确定性、动态性以及面对大量信息决策不知所措等系列问题,提出了复杂环境下的平行决策研究方法,利用平行决策的分解原理,将复杂决策分解为流决策、平行决策和交叉决策,不仅简化了复杂环境下的决策问题研究,而且从某种程度上能将信息优势适时地转化为决策优势,从而为系统决策获取最终的系统竞争优势提供一定的帮助。     

Team situation awareness measure using semantic utility functions for supporting dynamic decision-making

Team decision-making is a remarkable feature in a complex dynamic decision environment, which can be supported by team situation awareness. In this paper, a team situation awareness measure (TSAM) method using a semantic utility function is proposed. The semantic utility function is used to clarify the semantics of qualitative information expressed in linguistic terms. The individual and team situation awareness are treated as linguistic possibility distributions on the potential decisions in a dynamic decision environment. In the TSAM method, team situation awareness is generated through reasoning and aggregating individual situation awareness based on a multi-level hierarchy mental model of the team. Individual and team mental models are composed of key drivers and significant variables. An illustrative example in telecoms customer churn prediction is given to explain the effectiveness and the main steps of the TSAM method.     

基于multi-agent的企业经营战略群决策支持系统

为了应对迅速变化的市场环境和复杂纷繁的问题,企业有必要建立相应的经营决策支持系统来辅助决策。文章根据企业经营战略的特点,结合决策支持系统领域的多主体系统理论,提出了基于multi-agent的企业经营战略群决策支持系统的模型设计,并从宏观和微观两个部分给出模型的结构及运行过程。     

A study on modeling of human spatial behavior using multi-agent technique

This research follows the design and implementation of an agent-based modeling environment written in Java program language on AnyLogic simulation platform to facilitate observing the human spatial behaviors of electric taxis and passengers. The system is developed with a view to improve decision analysis and support for users or companies. In order to achieve the object of decision support for the user, we provide a graphic user interface (GUI) to support user making decision real time. User can simulate the real condition via GUI in our electric taxi DAR operation system to observe the simulation process and the result to help reasonable decision-making immediately. We validate our multi-agent simulation model with the electric taxi DAR operation system case study. Finally, according to the analysis result, we demonstrate that our multi-agent simulation model and GUI can help users or companies quickly make a quality and accurate decision to reduce the decision-making cost and time. In this condition, users or companies can strengthen their competitive advantage and response ability with the reduction of decision-making risk.     

Can reduced processing decision support interfaces improve the decision-making of less-experienced incident commanders?

Computer-based decision support systems have been proposed as a tool to improve the decision-making of less-experienced personnel by reducing the information processing demands necessary for decision-making. This study investigated the utility of three decision support system interfaces that differed in their capacity for reduced processing. The participants comprised experienced and less-experienced Fireground Incident Commanders who used the decision support system interfaces to identify the most appropriate entry point to extract a victim from a simulated burning building. The results revealed that reduced processing interfaces enabled less-experienced participants to acquire information using a process equivalent to their more experienced counterparts. However, this process did not result in improvements in the accuracy of the decision-making process. Indeed, the accuracy of experienced participants' decisions was consistently greater than the less-experienced participants, irrespective of the decision support system interface. It was concluded that the success of reduced processing decision support systems amongst less-experienced operators is significantly dependent upon their understanding of the relative value of key features associated with the decision-making process.     

基于情景感知的制造组合服务自适应决策机制

针对组合服务面临复杂、动态的业务需求和运行环境,提出基于情景感知的自适应决策机制,保证制造过程的顺利执行,为用户持续稳定地提供高质量服务.从智能制造单元个体视角,构建制造组合服务自适应调整系统架构.根据多变的应用场景,充分考虑决策主体状态和情境因素对服务调整活动的影响,设计情景感知、主体-情景-活动(Subject-situation-action,SSA)2种自适应决策机制,并通过多Agent技术搭建实验环境,应用关联规则方法挖掘决策知识.通过汽车制造云服务实例与现有事件驱动机制对比,验证所提出的2类机制在不同场景下的效果,并从决策准确率、时间和服务QoS两方面证明情景感知、SSA机制的优势和可行性.     

基于e-制造模式的动态工艺规划技术研究

e-制造工艺系统要求及时准确地根据资源动态信息进行工艺路线的实时决策。该文将工艺规划过程划分为零件信息输入、特征工艺选择和工艺路线的决策等3个阶段，采用神经网络BP算法解决了特征工艺链模糊匹配问题，将资源动态信息与工艺规则共同作为约束条件，采用GA算法实现基于资源动态信息的工艺路线优化选择，有效地解决了调度中的资源冲突问题。采用Java技术开发了软件原型系统，通过实例分析验证了设计方法的可行性。     

EXPERIENCE-BASED KNOWLEDGE REPRESENTATION: SOEKS

When managers make decisions, they use previous, similar, or equal experiences to help themselves in a new decision-making situation. Thus, keeping record of previous decision events appears to be of the utmost importance as part of the decision making process.

For us, every formal decision event has to be collected and stored as experienced knowledge, and any technology able to do this will allow us to improve the decision-making process by reducing decision time, as well as by avoiding duplication in the process. However, one of the most complicated issues about knowledge is its representation. Developing a knowledge structure that stores and administers experience from the day-to-day decision processes would improve decision-making quality and efficiency. We are proposing such a knowledge structure and have named it set of experience knowledge structure. A set of experience knowledge structure (SOEKS) is a combination of organized information obtained from a formal decision event. Fully applied, the set of experience knowledge structure would advance the notion of administering knowledge in the current decision-making environment.     

基于改进D-S证据理论的室内环境控制决策系统

针对室内环境因子多且相互作用关系复杂,影响室内环境舒适度的控制精准决策,设计了一种基于改进D-S证据理论的室内环境控制决策系统。首先采用箱线图法和均值替代法检测修复异常采集数据,然后利用距离自适应加权融合算法实现同类传感器数据一级融合,最后利用改进D-S证据理论算法,实现全局融合决策。实验结果表明,改进D-S证据理论算法能够有效避免冲突证据带来的融合决策误差,系统可以实现室内环境控制的精准决策,融合决策精度高,具有一定的推广应用价值。     

Implementation of a group decision support system utilizing collective memory

Collective memory has been characterized as a socially articulated and maintained reality of the past—one form being organizational memory. The collective memory concept can also be specialized to a group level and used to support work groups engaged in repetitive decision-making activities. The advantages of capturing collective memory are many, including simplification of the process, codification of decision strategies, carryover of knowledge when group composition changes, etc. While there has been substantial research in group decision-making, the explicit use of collective memory to support repetitive decision-making has received less attention. This paper describes the development and implementation of a collective memory based environment for support of a multi-attribute, iterative decision process. The environment utilized hypermedia, groupware, and Intranet technologies. Results of its use are encouraging.