用户名: 密码: 验证码:
基于BDN和Multi-Agent的突发事件应急风险决策方法研究
详细信息    查看官网全文
摘要
突发事件危害巨大,对社会经济发展、人民生活水平以及社会秩序产生了严重的危害。突发事件的应对十分复杂,必须要多个决策主体共同进行决策从而提高决策的准确性;由于突发事件具有不确定性,参与应急的决策主体、决策主体所做的决策、备选的决策方案集以及它们之间的关系是不确定的。本文所要研究的问题就是在应对突发事件时,考虑决策主体、决策以及决策方案之间关系的不确定性特征,如何降低决策的风险,进行应急风险决策。首先,为了描述多主体参与的应急风险决策问题,本文引入Agent思想,用于表示单个决策主体;采用贝叶斯决策网络(BDN),用于描述决策主体、决策以及决策方案之间的关系;其次,考虑决策主体、决策决策方案以及它们之间关系的不确定产生的风险问题,提出多属性的决策方案选择方法。该方法采用BDN推理和多Agent交互以获得满足当前决策目标的多主体的备选方案集合。在此基础上,根据决策方案的多属性计算单个方案在应对突发事件不同情景的损益值,基于方案损益值和方案间关系强度计算各备选方案集合的综合评价值,进而选择最优的方案集合。最后,案例分析的结果验证了本文提出的应急风险决策方法的可行性与合理性。
Emergency events can cause great harm.They can affect social and economic development,people' s iving standards and social order.Emergency response is very complicated.Multi decision-making bodies have to join decisions so as to improve the accuracy of decision.Because the emergency event has uncertainty,participation in decision-making,the decision,the alternative decision plan,and the relationship among them are uncertain.In this paper,the risk analysis method is proposed by considering multi-agent characteristics to solve the risk decision-making problems in emergency response.First of all,the uncertain relationships among multi-agents,the basic characteristics of the decision maker,decision-making,decision-making plan and the relationship among them are analyzed.In order to describe the emergency risk decision-making problem,the Agent theory is introduced,Agent model is used to represent the decision maker;at the same time the Bayesian decision network(BDN) is applied to describe the relationship among decision maker,decision-making and decision-making plan.Then,considering the risk decision problems caused by the uncertainty of decision maker,decision-making,decision-making plan and the relationship among them,multi-attribute decision-making plan selection method is proposed.The method adopts the BDN reasoning and multiple Agent interaction to obtain the collection of multi-agent alternatives that satisfying the current decision target.On this basis,the gains and losses of every plan are calculated according to multiple attribute in dealing with the different scenarios,and the comprehensive evaluation value of each alternative is calculated based on the gains and losses of plans and the strength of relationship among them to choose the optimal solution set.Finally,the results of case analysis verify the feasibility and rationality of the proposed emergency decision-making method.
引文
[1]杜健.应急管理中的多主体协调决策研究[D].大连:大连理工大学,2010.
    [2]王永军.突发事件应急协同决策主体评价与选择研究[D].哈尔滨:哈尔滨工程大学,2011.
    [3]张培勇.突发事件中应急救援方案选择的序贯协同决策方法[D].南京:南京理工大学,2014.
    [4]陈明天.突发自然灾害救灾物资分配优化决策研究[D].大连:大连理工大学,2012.
    [5]汪业凤.突发事件应急决策过程中群体冲突协调机制研究[D].长沙:中南大学,2012.
    [6]孙香勤.三峡坝区突发事件的多主体应急联动组织优化研究[D].武汉:武汉理工大学,2007.
    [7]王慧敏,刘高峰,佟金萍,等.非常规突发水灾害事件动态应急决策模式探讨[J].软科学,2012,26(01):20-24.
    [8]沈荣鉴.考虑决策者行为的突发事件应急响应风险决策方法研究[D].沈阳:东北大学,2011.
    [9]樊治平,刘洋,沈荣鉴.基于前景理论的突发事件应急响应的风险决策方法[J].系统工程理论与实践,2012,32(05):977-984.
    [10]李仕峰,杨乃定,张云翌.突发事件下选择应急方案的风险决策方法[J].控制与决策,2013,28(12):1859-1863.
    [11]刘洋,樊治平,袁媛.突发事件应急响应的多属性风险决策方法研究.运筹与管理[J],2013,22(01):23-28.
    [12]宋琪.基于CAS理论的地震应急协同物流网络演化与仿真研究[D].哈尔滨:哈尔滨工程大学,2012.
    [13]郭航.面向自治服务协同环境的规范多Agent体系研究:表示、实施与推理[D].杭州:浙江大学,2010.
    [14]江时雨,陈建宏,李涛,等.基于Bayes-Monte Carlo的突发事件应急方案风险决策方法研究[J].世界科技研究与发展,2015,37(01):5-8.
    [15]程铁军,吴凤平,李锦波.基于累积前景理论的不完全信息下应急风险决策模型[J].系统工程,2014,32(04):70-75.
    [16]王亮,王应明.基于前景理论的动态参考点应急决策方法研究[J].中国管理科学,2013,21(S1):132-140.
    [17]袁媛,刘洋,樊治平.考虑后悔规避的突发事件应急响应的风险决策方法[J].系统工程理论与实践,2015,35(10):2630-2636.
    [18]Wang Liang,Zhang Zixin,Wang Yingming.A prospect theory-based interval dynamic reference point method for emergency decision making[J].Expert Systems with Applications,2015,42(23):9379-9388.
    [19]Liu Yang,Fan Zhiping,Zhang Yao.Risk decision analysis in emergency response:A method based on cumulative prospect theory[J].Computers&Operations Research,2014,42:75-82.
    [20]Liu Yang,Fan Zhiping,Yuan Yuan,et al.A FTA-based method for risk decision-making in emergency response[J].Computers&Operations Research,2014,42:49-57.
    [21]何华刚,吕奎,王耀琪.基于Agent的长输管道应急指挥决策系统设计[J].中国安全科学学报,2011,21(11):164-171.
    [22]陈登科,张大发,江玮,等.基于MAS的船用压水堆核事故应急决策支持系统[J].核动力工程,2007,28(5):79-82.
    [23]张振海,王晓明,张雁鹏,等.基于分层Agent的铁路应急救援决策方法研究[J].交通运输系统工程与信息,2013,13(02):141-146.
    [24]郭茂耘.基于多Agent的航天发射应急处理系统研究[D].重庆:重庆大学,2004.
    [25]Wang Z,Zlatanova S Multi-agent based path planning for first responders among moving obstacles[J].Computers,Environment and Urban Systems,2016,56:48-58.
    [26]Edrissi A,Poorzahedy H,Nassiri H,et al.A multi-agent optimization formulation of earthquake disaster prevention and management[J].European Journal of Operational Research,2013,229(1):261-275.
    [27]Gomez C,Sánchez-Silva M,Duenas-Osorio L.An applied complex systems framework for risk-based decision-making in infrastructure engineering[J].Structural Safety,2014,50:66-77.
    [28]Ball D R,Deshmukh A,Kapadia N.An options-based approach to coordinating distributed decision systems[J].European Journal of Operational Research,2015,240(3):706-717.
    [29]Hu Zhihua,Sheng Zhaohan.Disaster spread simulation and rescue time optimization in a resource network[J].Information Sciences,2015,298:118-135.
    [30]Mustapha K,Mcheick H,Mellouli S.Modeling and simulation agent-based of natural disaster complex systems[J].Procedia Computer Science,2013,21:148-155.
    [31]Gorbil G,Gelenbe E.Opportunistic communications for emergency support systems[J].Procedia Computer Science,2011,5:39-47.
    [32]濮永仙.贝叶斯决策网在智能决策系统中的应用研究[J].计算机与数字工程,2011,39(08):53-56.
    [33]Greco L,Presti L L,Augello A,et al.A decisional multi-agent framework for automatic supply chain arrangement[M]//Lai C,Semeraro G,vargiu E.New challenges in distributed information filtering and retrieval.Berlin Heidelberg:Springer,2013:215-232.
    [34]Zhu J Y,Deshmukh A.Application of Bayesian decision networks to life cycle engineering in Green design and manufacturing[J].Engineering Applications of Artificial Intelligence,2003,16(2):91-103.

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700