摘要
Agent和Multiagent System(MAS)的理论和技术是目前分布式人工智能和计算机科学技术研究的核心内容之一。在实际工作中,笔者参与课题研究和项目开发,产生了设计MAS以促进信息系统建设,提高业务部门信息化程度,在发生危险品泄漏事故时能提供辅助决策,降低事故损失,找寻构建MAS的解决方法,找寻构建应急管理系统解决方法的初衷。在此基础上,做了相应的信息系统分析、设计和开发工作,以及系统中应急资源管理调度模型的研究工作,完成了本学位论文。
本文研究船舶载运危险品运输过程中发生危险品泄漏事故时的应急反应管理的问题,以期提出相关解决方案,并探索建设MAS系统的理论、方法和技术。本文在前人研究成果的基础上做了一些有效的改进和有益的尝试。主要研究内容和研究结果如下：
(1)分析了江苏海事局危险品应急管理业务,提出了相应的业务模型。有害有毒物质水上运输的危险品运量大、品种多、特性复杂,一旦发生事故,救助和环境修复作业都十分困难。SPAERIS项目对江苏海事局危险品应急管理业务进行了完整的业务调研、需求分析,提出了业务模型,设计了系统解决方案,并建设了该信息系统。
(2)提出了基于Agent建模和模拟的应急资源分配模型和机制。通过构造Agent联盟形成解决问题的MAS系统。该Agent联盟形成以Agent间多边谈判为基本机制,以Agent的个体福利和MAS系统的整体福利为导向。文中给出模型的形式化描述和算法描述,并以一个例子探讨了该模型和机制。
(3)采用回归分析和混合关联规则模型,根据港口实际数据,分析出影响危险品事故的因素,得出结论为,对装卸货品影响最大的因素首先是船舶种类,上一港口和下一港口的影响程度远小于船舶种类,二者基本持平。货品的数量与前三者相比基本可以认为实质上不影响货品种类。
Agent and Multiagent System (MAS) theory and technology is now distributed artificial intelligence and computer science and technology research is one of the core contents. In practice, the author participated in a real project. Due to the actual situation of the system with some characteristics of MAS, resulting in a design MAS to promote information system construction, to improve the business sector informatization level, in dangerous when leakage accident can provide aided decision-making, reduce loss of accident, and seek the solution and constructing MAS construct emergency management system for the solution. On this basis, the corresponding information system analysis, design and development, and the system of emergency resource management model research, completed this dissertation.
The main research contents and the results are as follows.
(1) Jiangsu maritime bureau analyzed hazardous emergency management business, and advances concrete model. SPAERIS by systematic project, engineering, the method of jiangsu maritime dangerous emergency management business of business research, a complete requirements analysis, puts forward the concepts of the business model, design the system solutions, and the construction of the information system.
(2) Based on the Agent forward modeling and simulation of resource allocation model and mechanism of emergency. Through the construction Agent alliance formed to solve the problem of MAS system. The alliance formed by Agent between Agent for the basic mechanism, multilateral talks with the Agent of individual welfare and benefits of the whole system of MAS. Given the model and algorithm of formalization by an example, and discusses the model and mechanism.
(3) Using partial least-square regression analysis and mixed association rules based on actual data model, and analyses the harbor dangerous factors of influence, and the accident weight factor.
引文
[1]杨赞.江苏海事局辖区船舶载运有害有毒物质事故应急反应对策研究.2006,大连：大连海事大学出版社.
[2]范维澄.国家突发公共事件应急管理中科学问题的思考和建议.中国科学基金,2007(2)：p.71-76.
[3]Luger, G.S., William. Artificial Intelligence:Structures and Strategies for Complex Problem Solving (5th ed.).2004:The Benjamin/Cummings Publishing Company.
[4]Kendal, S.L.C., M.An introduction to knowledge engineering.2007, London Springer.
[5]维基百科Intelegent Agent. http://en.wikipedia.org/wiki/Intelligent_agent.
[6]石纯一,张伟.基于Agent的计算.2007,北京：清华大学出版社.
[7]维基百科.MAS. http://en.wikipedia.org/wiki/Multi-agent_system.
[8]江苏安全生产网.国务院对中石油吉林石化分公司双苯厂“11·13”爆炸事故及松花江水污染事件作出处理..http://www.jssafety.gov.cn/text_safety/shiguanli/huozaibaozha/sga10611280916.htm.
[9]Mener, A.S. Disaster Response in the United States of America:An Analysis of the Bureaucratic and Political History of a Failing System, in Department of Political Science.2007, University of Pennsylvania:Pennsylvania.
[10]National Response Framework. http://www.dhs.gov/xprepresp/committees/editorial_0566.shtm.
[11]Clare, M. Governance of Disaster Response in Scotland, Resilience Management, and Adaptation to Increasing Climatic Variability.2009.
[12]Schellong, A. Increasing Social Capital for Disaster Response through Social Networking Services(SNS) in Japanese Local Governments.2007.
[13]James Jay Carafano, E., Grassroots. Disaster Response:Harnessing the Capacities of Communities. Backgrounder. Backgrounder,2007(2094).
[14]Ulhrnik, J.A. The Counselor and the Disaster Response Team:An Emerging Role.
[15]Rizwan U. Farooqui, S.H.L. A Structured Approach to Develop an Effective Emergency Preparedness and Disaster Response Plan for Constractors.
[16]Iannella, K.R.a.R. Information standards for disaster response and recovery.
[17]Nathan Schurr, J.M.E. The future of Disaster Response:Humans Working with Multiagent Teams using DEFACTO.
[18]Kodrich, M.L.A.K. On Line Disaster Response Community:People as Sensors of High Magnitude Disasters Using Internet GIS. sensors,2008(8):p.3037-3055.
[19]Department, E.M.S. Disaster Response and Biological/Chemical Terrorism.2001.[20]美国急救医师学会http://www.acep.org/.
[21]董海英.建立应急指挥系统应对公共危机事件.技术与市场,2006(12)：p.80-82.
[22]陈全君,何学秋.事故的定义及其发生原因的理性分析.矿业安全与环保,2004.31(5)：p.10-12.
[23]付跃强,刘卫东.突发事件应急响应系统的结构和功能分析.商场现代化,2006(489)：p.11-12.
[24]张建,何珑.构建环境突发事件应急处置信息系统.计算机与信息技术,2006(5)：p.101-105.
[25]徐富春等.环境事故应急决策管理信息系统建设.环境科学动态,2005(2)：p.11-13.
[26]中国地理信息系统协会.http://www.cagis.org.cn/.
[27]百度百科.GIS. http://baike.baidu.com/view/5094.htm.
[28]张志,徐林.基于ComGIS的重大化工事故应急救援指挥系统.计算机与信息技术,2006：p.101-105.
[29]梁嫒,石剑荣.基于GIS的城市级突发环境灾难事件应急救援信息系统的设计与实现初步研究.中国环境学学会学术年会优秀论文集.2006.
[30]张嵘嵘,曾向东.突发性环境污染事故应急GIS系统框架的建立.云南师范大学学报,2007.27(2)：p.24-28.
[31]李天华等.中小城市应急联动指挥系统构建及分析.测绘科学,2007.32(2)：p.141-142.
[32]维基百科.达特茅斯会议.http://en.wikipedia.org/wiki/Dartmouth_Conference.
[33]Kitano, H. Massively Parallel Artificial Intelligence.1994:AAAI Press
[34]维基百科.DAI. http://en.wikipedia.org/wiki/Distributed_artificial_intelligence.
[35]Bond, A.H. Readings in Distributed Artificial Intelligence.1988:Morgan Kaufmann Pub.
[36]李英.多Agent系统及其在预测与智能交通系统中的应用.2004,上海：华东理工大学出版社.
[37]Dianxun Shuai, X.F. Distributed problem solving in multiagent systems:a spring net approach. IEEE intelligent systems 2005.20(4):p.22-27.
[38]Durfee, E.H. Distributed Problem Solving and Planing, in MultiAgent Systems:A Modern Approach to Distributed Artificial Intelligence.1999, MIT Press.
[39]Answers.http://www.answers.com/computer+Agent+?cat=technology.
[40]Nwana, H.S. Software Agents:An Overview. Knowledge Engineering Review,1996.11(3):p. 1-40.
[41]Minsky, M. The Soeiety of Mind.1985, New York:Simon & Sehuster.
[42]Smith, R.G. The Contract Net Protocol-High-Level Communication and Control in a Distributed Problem Solver. IEEE Transactions on computers,1980.29(12):p.1104-1113.
[43]Shoham, Y. Agent-Oriented Programming. Artificial Inteligence,1993(60):p.51-92.
[44]Michael W, E. Intelligent agents:theory and practice. Knowledge Engineering Review,1995. 10(2):p.115-152.
[45]Hayes-Roth, B. Agents on Stage:Advancing the State of Art of AI Extended Abstract for an Invited Talk. in IJCAI1995.1995. Montreal, Quebec, Canada
[46]A.Newell, H.A.S. GPS:A program that simulates human thought.1961:R. Oldenbourg KG.
[47]R.E.Fikes. STRIPS:A new approach to the application o ftheorem proving to problem solving. Artificial Intelligence,1971.5(2):p.189-208.
[48]S.Russell. Provably bounded-optimal agents. Journal of AI Research,1995(2):p.575-609.
[49]Chapman, D. Planning for conjunctive goals. Artificial Intelligence,1987(32):p.333-378.
[50]Brooks, R. A. A robust layered control system for a mobile robot. IEEE Journal of Robotics and Automation,1986.2(1):p.14-23.
[51]Brooks, R.A. Intelligence without representation. Artificial Intelligence,1991(47):p.139-159.
[52]Brooks, R.A. Intelligence without reason, in IJCAI-91.1991. Sydney,Australia.
[53]Steels, L. Cooperation between distributed agents through self organization. in MAAMAW-89. 1990. Amsterdam,Netherland:Elsevier Science Publisher.
[54]N.R.Jennings. Towards a social-level characterisation of socially repossible agents. IEE Transactions on Software Engineering,1997.144(1):p.11-25.
[55]M.E.Bratman, D.J.I. Plans and resource-bounded practial reasoning. Computational Intelligence, 1988(4):p.349-355.
[56]M.P.Georgeff. Reactive reasoning and planning. in AAAI-87.1987. Seattle,USA.
[57]Bigus, J.P. The Agent Building and Learning Environment, in Proceedings of the Fourth International Conference on Autonomous Agents.2001:Barcelona, Spain. p.596-601.
[58]Agent Builder, http://www.agentbuilder.com/.
[59]Microsoft Agent. http://msdn.microsoft.com/en-us/library/ms695784%28VS.85%29.aspx.
[60]Repast, http://repast.sourceforge.net/.
[61]Amin Milani Fard et, al. Multi-agent Data Fusion Architecture for Intelligent Web Information Retrieval. International Journal of Intelligent Systems and Technologies,2007.2(3):p.201-205.
[62]史殿习,王怀民.基于agent的分布式集成环境.国防科技大学学报,1997.19(4)：p.5-11.
[63]杜荣华.基于Agent的ITS共用信息平台框架及其关键技术的研究,in博十学位论文.2005,国防科学技术大学.
[64]李斌,陈韬略,吕建.一个刻画Agent自主性的框架.南京大学学报(自然科学版),2004.40(2)：p.137-145.
[65]傅松,陆剑锋等.多Agent系统开发方法AMT及其应用.南京大学学报(自然科学版),1997.33(2)：p.215-222.
[66]汤健,郑丽伟,金芝.需求驱动的服务Agent协作及其协商框架,in第二届中国Agent理论与应用学术会议.2008,南京大学出版社：南京.p.657-661.
[67]毛文吉.多Agent交互环境下的社会推理计算模型,in第二届中国Agent理论与应用学术会议.2008,南京大学出版社.p.1138-1142.
[68]曹军海,张和明,熊光楞.多Agent仿真中Agent行为的形式化描述方法.系统仿真学报,2004.16(11)：p.55-61.
[69]刘海龙,吴铁军.基于合同网的多agent任务分配分布式优化算法.2001(5)：p.13-19.
[70]李松仁,景广军.基于multi agent的分布式专家系统原理及应用.中南工业大学学报,2001.32(1)：p.20-24.
[71]王俊普,徐杨等.基于Agent的智能控制系统的实现方法研究.自动化与仪表,2001.16(3)：p.26-29.
[72]FIPA. http://www.fipa.org/.
[73]OMG.对象管理组织http://www.omg.org/.
[74]Wooldridge, J.多Agent系统引论.2003,北京：电子工业出版社.
[75]Russel, S.N. Artificial Intelligence:a Modern Approach.1995, Englewood Cliffs, NJ: Prentice-Hall.
[76]Moore, R.c. A formal theory of knowledge and action, in Reading in Planning, J.H.a.A.T. J.F. Allen, Editor.1990, Morgan Kaufmann:San Mateo, CA. p.480-519.
[77]Hewitt, C.E. Offices are open systems. ACM Transactions on Office Information Systems,1986. 4(3):p.271-287.
[78]何新贵.知识处理与专家系统.1990,北京：国防工业出版社.
[79]Nilsson,N.J.人工智能.2009,北京：机械工业出版社.
[80]Rademache. Cognition in system. Cybernetics and systems,1996(27):p.1-41.
[81]Russel, S.a.S., D. Provably bounded-optimail agents. Journal of AI Research,1995(2):p. 575-609.
[82]石纯一,张伟等译.多Agent系统引论.2003,北京：电子工业出版社.
[83]Stefano Ceri, J.W. Production Rules in Parallel and Distributed Database Environments, in the 18th International Conference on Very Large Data Bases.1992, Morgan Kaufmann:San Francisco,USA. p.339-351.
[84]维基百科.Speech act. http://en.wikipedia.org/wiki/Speech_act.
[85]Austin, J.L. How to do Things with Words.1962:Oxford Press.
[86]Association of Lisp Users, http://www.lisp.org/alu/home.
[87]KQML.知识查询操纵语言.http://www.cs.umbc.edu/kqml/.
[88]Y.Labrou. A proposal for a new KQML specification. External Interfaces Working Group working paper, ARPA Knowledge Sharing Initiative.,1997.
[89]魏晓斌等.Agent通信机制探讨.计算机工程与应用,2002(5)：p.66-70.
[90]Sandholm, T.a.V.L. Leveled Commitment Contracting:A Backtracking Instrument for Multiagent Systems. AI Magazine,2002.23(3):p.89-100.
[91]于永利等编.离散事件系统模拟2003,北京：北京航空航天大学出版社.
[92]钟永光,贾晓菁,李旭.系统动力学.2009,北京：科学出版社
[93]Sichman, J.S. Multi-Agent-Based Simulation Ⅷ2008:Springer Berlin/Heidelberg.
[94]Uhrmacher, A.M. Multi-Agent Systems:Simulation and Applications 2009:CRC Press.
[95]约翰·霍兰.涌现——从混沌到有序.2006,上海：上海科技出版社.
[96]Agent-based model. http://en.wikipedia.org/wiki/Agent-based_model.
[97]Holland, J.H. Hidden order:how adaptation builds complexity.1995, Mass.: Addison-Wesley:Reading.
[98]Gilbert, N.a.K.G.T. Simulation for the social scientist.1999, Buckingham, UK:Open University Press.
[99]Sakoda, J.M. The checkerboard model of social interaction. Journal of Mathematical Sociology, 1971(1):p.119-132.
[100]Epstein, J.M.a.R.A. Growing artificial societies:social science from the bottom up.1996, Cambridge, MA:MIT Press.
[101]Tesfatsion, L. Agent-based computational economics:growing economics from the bottom up. Artificial Life,2002.8(1):p.55-82.
[102]Tesfatsion, L. Agent-based Computational Economics(ACE) 2005 http://www.econ.iastate.edu/tesfatsi/ace.htm.
[103]Kohler, T., G.Gumerman, and R. Reynolds. Simulating ancient socienties. Scientific American, 2005.
[104]Macy, M.a.R.W.. From factors to actors:computational sociology and agent-based modeling. Annual Review fo Sociology,2002.28:p.143.
[105]Bedau, M.A. Artificial Life:Organization,Adaptation and Complexity from the Bottom Up. TRENDS in Cognitive Sciences,2003.7(11):p.505-512.
[106]Gratch, J.a.S.M. Tears and fears:modeling emotions and emotional behaviors in synthetic agents. in the 5th International Conference on Autonomous Agents.2001.
[107]Sallach, D.a.C.M. The simulation of social agents:an introduction. Social Science Computer Review,2001.19(3):p.245-248.
[108]T.W.Sandholm. Distributed rational decision making, in Multiagent Systems:A Modern Approach to Distributed Artificial Intelligence, G.Weib, Editor.1999, MIT Press.
[109]K.J.Arrow. A.K.S., and K. Suzumura, Handbook of Social Choice and Welfare.2002: North-Holland.
[110]H.Moulin. Axioms of Cooperative Decision Making.2988:Cambridge University Press.
[111]A.K.Sen. Collective Choice and Social Welfare.2970:Holden Day.
[112]Maudet, U.E.a.N. Welfare engineering in multiagent systems. in Engineering Societies in the Agent Worlds IV.2004.
[113]A.D.Taylor, S.J.B.a. Fair Division:From Cake-cutting to Dispute Resolution.1996: Cambridge University Press.
[114]H.Moulin. Fair Division and Collective Welfare.2003:MIT Press.
[115]H.P.Young. Equity in Theory and Practice.1994:Princeton University Press.
[116]P.Cramton, Y.S., and R.Steinberg. Combinatorial Auctions.2006:MIT Press.
[117]Sandholm, T. Negotiation among Self-Interested Computationally Limited Agents.1996, University of Massachusetts at Amherst.
[118]Sandholm, T., K.Larson, M. Andersson, O.Shehory, and F. Tohme. Coalition Structure Generation with Worst Case Guarantees. Artificial Inteligence Journal,1999(111):p.209-238.
[119]Fudenberg, D.a.J.T. Game theory.1991, Cambridge MA:MIT Press.
[120]Osborne, M.J.a.A.R. A Course in Game Theory.1994, New York:MIT Press.
[121]Sandholm, T.a.V.L. Coalitions among Computationally Bounded Agents. Special issue on Economic Principles of Multiagent Systems, Artificial Intelligence,1997.94(1):p.99-137.
[122]Sen, S.a.P.S.D. Searching for Optimal Coalition Structures, in Proceedings of the Fourth International Conference on Multiagent Systems.2000:Boston, MA. p.286-292.
[123]Caillou, P., S. Aknine, and S. Pinson. A Multi-Agent Method for Forming and Dynamic Restructuring of Pareto Optimal Coalitions, in Proceedings of the First International Joint Conference on Autonomous Agents and Multi-agent Systems.2002:Bologna, Italy, p.1074-1081.
[124]Li, C.a.K.S. Algorithm for Combinatorial Coalition Formation and Payoff Division in an Electronic Marketplace, in Proceedings of the First International Joint Conference on Autonomous Agents and Multi-agnet Systems.2002:Bologna, Italy, p.120-127.
[125]Shehory, O.a.S.K. Task Allocation via Coalition Formation among Autonomous Agents, in Proceedings of the International Joint Conference of Artificial Intelligence.1995:Quebec Canada. p. 655-661.
[126]Shehory, O.a.S.K. Methods for Task Allocation via Agent Coalition Formation. Artificial Inteligence Journal,1998.101(1-2):p.165-200.
[127]Shehory, O.a.S.K. Feasible Formation of Coalition among Autonomous Agents in Non-Super-Additive Ennvironments. Computational Intelligence,1999.15(3):p.218-251.
[128]Lerman, K.a.O.S. Coalition Formation for Large Scale Electronic Markets, in Proceedings of the International Conference on Multi-Agent Systems.2000:Boston, MA. p.167-174.
[129]Rosenschein, J.S.a.G.Z. Rules of Encounter:Designing Conventions for Automated Negotiation among Computers.1994, Cambridge, MA:MIT Press.
[130]Kraus, S. Strategic Negotiation in Multiagent Environments.2001, San Francisco, CA:MIT Press.
[131]Jennings, N., P. Faratin, A. Lomuscio, S. Parsons, C. Sierra, and M. Wooldridge. Automated negotiation:Prospects, Methods and Challenges. International Journal of Group Decision Negotiation,2001.10(2):p.199-215.