面向非常规突发事件应急管理的临机决策计算模型研究

发布时间：2018-03-15 21:10

本文选题：进化论决策范式　切入点：临机决策计算模型　出处：《天津大学》2016年博士论文　论文类型：学位论文

【摘要】：临机是在时间和资源有限的情况下,决策主体凭直觉创造性的提出决策方案的过程。构建具有临机能力的决策智能系统是提高非常规突发事件应急效率的重要的手段。当前,一方面,尽管已有一些基于即兴机制的临机决策计算方法,然而这些方法对决策主体认知过程的描述性模型研究较少;另一方面,因为缺乏临机的形式化表示,进而很难将临机决策认知过程应用于实践。为了实现具有临机能力的面向非常规突发事件的应急决策智能决策系统,本文以非常规突发事件环境下(个体)决策者为研究对象,展开了以下三项研究内容。1.基于进化论决策范式,构建通用的临机决策描述性模型。以进化论决策范式下的“生态理性”为基础,提出了非常规突发事件的临机决策描述性模型。该模型提供了基于图式的信念表征方法和基于再认启发式的决策方案生成过程。该模型不仅根据进化论决策范式对临机决策认知过程提供了科学的描述,而且可用于解释应急决策、紧急决策、危机决策、风险决策等,具有科学性和通用性。2.基于非公理逻辑,构建一个临机决策形式化模型。该模型能够保证研究内容1的假设前提与非公理逻辑的智能的一致性。同时,基于经验的语义模型和多种推理模式可以为研究内容1所提出的基于图式的信念和再认启发式策略的表征提供规范的形式化表示方法。3.设计统一的临机决策计算模型。基于本文构建的形式化模型,实现一个具有统一的感知、记忆、决策、学习模块的临机决策计算模型,并以此为核心构建一个面向城市消防领域的智能决策支持系统。通过实例表明,本文提出的计算模型能够为应急决策智能系统提供一种具有统一的认知机制框架以及可在计算机中模拟的方法。本文提出的临机决策计算模型为临机决策提供描述性模型和形式化模型,从而为面向(非常规)突发事件具有临机能力的应急决策智能系统提供了理论基础,并为临机决策认知过程应用于实践提供了可行方案。
[Abstract]:Under the condition of limited time and resources, the decision makers creatively put forward the decision plan by intuition. It is an important means to improve the efficiency of emergency response of unconventional emergencies to build a decision intelligence system with the capability of the machine. On the one hand, although there are some improvisational mechanisms based on the temporary machine decision calculation methods, but these methods have less research on the descriptive model of the cognitive process of decision makers; on the other hand, because of the lack of formal representation of the machine, Therefore, it is difficult to apply the cognitive process of decision making to practice. In order to realize the intelligent decision system of emergency decision oriented to unconventional emergencies, this paper takes the decision makers under the environment of unconventional emergencies as the research object. 1. Based on evolutionary decision paradigm, a general descriptive model of machine decision is constructed, which is based on "ecological rationality" under evolutionary decision-making paradigm. In this paper, a descriptive decision making model for unconventional emergencies is proposed. The model provides a schema-based belief representation method and a decision scheme generation process based on recognition heuristics. The model is not only based on the evolutionary decision paradigm. It provides a scientific description of the cognitive process of decision making. And it can be used to explain emergency decision, emergency decision, crisis decision, risk decision, etc. It is scientific and universal. 2. Based on non-axiomatic logic, A formalized model of machine decision is constructed, which can guarantee the consistency between the hypothetical premise of research content 1 and the intelligence of non-axiomatic logic. At the same time, Experient-based semantic models and multiple reasoning models can provide a standardized formal representation method for schema-based belief and recognition heuristic strategy representation proposed in the first part of the study. Model. Based on the formal model constructed in this paper, In this paper, an intelligent decision support system for urban fire protection field is constructed, which has a unified model of perception, memory, decision making and learning, and an intelligent decision support system based on this model is presented. The computational model presented in this paper can provide a unified cognitive mechanism framework for the intelligent system of emergency decision making and can be simulated in the computer. Declarative model and formal model, It provides a theoretical basis for the intelligent emergency decision system which has the ability to face (unconventional) emergencies, and provides a feasible scheme for the application of the cognitive process of the emergency decision to the practice.
【学位授予单位】：天津大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TP311.52;D035

【相似文献】