Study Of Key Technologies On The Railway Emergency Rescue Intelligent Decision

Conventional rail emergency rescue work adopts emergency response modesbased on the contingency plans. In fact, emergencies incidents are rarely consistentwith pre-designed scenarios. Therefore, researching a more effective emergencydecision method is an important work for ensuring smooth development of therailway emergency rescue. The article adopts the empirical analysis method combinedwith theoretical research and system modeling, simulation calculation and applicationdevelopment. The railway emergency rescue intelligent decision is taked as theresearch object. The distributed group decision method, systems and decision supportsystems of the railway emergency rescue are explored and built. And the key issues ofrailway emergency rescue intelligent decision are studied. What’s more, a distributedmulti-stage Dynamic Group Decision Model and individual decision based oncase-based reasoning of railway emergency rescue is proposed and related issues havebeen studied. Finally, through the development and application of research projects,the method is validated advancing and availability.This paper’s main works are as follows:(1)Studing the railway emergency rescue group decision methods and thearchitecture of support systems.Based on the researching of the foundation on the limitations of contingencyplans decision-making model, the paper proposes the railway emergency rescuedecisions actually should be a decision model which combines experience and rules.On decision-making achievement, it is a distributed multi-stage dynamic groupdecision-making system which is decision-making groups leaded by a leading groupof railway emergency rescue. The features of decision-making model are distributed,multi-stage, dynamic, multi-Agent, group decision-making etc. Then the paperproposes the group’s individual decisions should take the decision-making methodbased on case-based reasoning. Finally,the paper proposes the physical structure ofour railway emergency rescue decision system framework and the railway emergencyrescue group decision support system.(2)Studing the decision making method of individual expert who participate inrailway emergency rescue.Individual of railway emergency rescue decisions adopts decision makingmethod based on CBR and RBR. Reasoning techniques based on case include scase representation, case retrieval, case reuse, case adjustment, case studies and case basemaintenance.(3)Studing the key technologies of the railway emergency rescue distributedmulti-stage dynamic group decision-making.Distributed group decision model under network environment is the key toachieve group decision reasoning. The paper presents a railway emergencyMulti-Agent group decision model based on case. Based on various professions (task)of railway rescue, it is divided into locomotive decisions, works decisions, electricityservices decisions etc. The policy experts group respectively generates respectivegroup decision making program by using Agent to make decisions.As decision conflict may exist between Multi-Agent Decision, resource andobjectives are reasonable arranged through a multi-Agent coordination strategy. Andadjusting their behavior to achieve the greatest degree of individual or system goals.Paper presents a multi-stage groups coordinated strategy model based on the optimalstrategy of optimizing the weights. Through multi-Agent coordination, formingdecision-clustering. Finally, forming group satisfaction decision scheme andgenerating decisions reports to guide emergency rescue of emergency eventsGiven the characteristics(levels, qualitative and quantitative, multi-expertdecision-making, indicators ash etc) of railway emergency rescue decision-makingprogram evaluation system, and for the effectiveness evaluation questions of groupdecision-making program in the implementation, the paper presents gray-level groupsanalysis (GGAHP) method to evaluate decision-making programs.(4)The research focusing on a key technical supporting system of the frameworkon the railway emergency rescue intelligent group decision.The research on the railway emergency rescue intelligent group decisionsupporting system is a key problem of the research on the intelligent decisionsupporting system. A decision for the supporting system framework is built in thedissertation and the J2EE SSH framework is used as the system platform webapplication and designs the system function module. Based on the railway emergencyrescue intelligent group decision support system the system is designed anddeveloped. It has been used in Chengdu Railway Bureau and Urumchi RailwayBureau and the decision support system is working properly.The innovative works of this paper are as follows:(1) By making a more comprehensive literature search and empirical surveys, thepaper establishes a railway emergency rescue decision which the making system anddecision supporting system architecture is based on the case-based reasoning anddistributed multi-stage dynamic distributed group decision; and proposes multi-Agentgroup decision model which is based on task of railway emergencies event; and the proposes for multi-stage groups coordinated strategy is based on Pareto optimalsolution and emergency rescue group decision evaluation methods is based on GreyAHP(2) The key technologies to the case-based reasoning are in-depth researched,and the representation method based on a combination of PETRI network and XML,and the retrieval algorithm based on IPSO-BP neural network, and the adjust themodel based on a combination of CBR and RBR and the learning algorithm based ona combination of BP neural network and the genetic algorithm is proposed.(3) A railway emergency rescue intelligent group decision supporting systemframework is built and is based on technology roadmap and implementation methodof the railway emergency rescue decision-making. Rescue resource management isbased on Web-GIS and railway emergency rescue individual decision is based onCBR as well as group decision-making process is based on distributed multi-stage,both of which are realized. Through the Chengdu Railway Bureau and the UrumqiRailway Bureau trial, the advanced nature of the method is verified and available. Thedistributed multi-stage group decision model is applied in2012National Science andTechnology Support Program "Regional Rail logistics service platform developmentand application"(Topic No.:2012BAH20F05), and China’s first group of distributeddecision-making technology-based regional railway logistics collaboration platform---Urumqi Railway multivariate logistic services platform is built. The platformachieves a logistics supply chain business decisions for high degree of integrationwith business processes and achieves business flow, information flow, logistics,capital flow integration, operation and harmonization, to effectively promote thewestern regional logistics services business conduct.