Research On The Emergency Rescue And Evacuation System Based On Dynamic Traffic Assignment

In recent years, natural disasters, accidents, public safety incidents occur frequently,which put the people’s lives and properties into great danger.In the process of disasterrescue and relief, transportation plays a crucial role.The research of this thesis tries tointegrate intelligent transportation systems into emergency management, and researchesemergency response and evacuation decision support systems.First, after analyzing the characteristics of the emergency rescue and evacuationdecision-support problem, this paper studies the dynamic traffic assignment (DTA)theory, and introduces a DTA model that meets the efficiency requirements of real-timesystems.The DTA model is based on traffic simulation, consisting of a traffic demandsimulator and a traffic supply simulator.The traffic demand simulator assigns dynamicOD traffic demands to different paths based on path utility function using discretechoice model.The traffic supply simulator simulates the traveling of the vehicles on thenetwork with an event-based meso-scopic traffic simulator, thus updating the dynamicnetwork traffic conditions.The DTA model is improved in the aspects of networktopology, dynamic OD data structures and event mechanism, which makes it suitable foremergency transportation; by utilizing an event-based meso-scopic traffic simulator, theefficiency of the model is improved.Based on the DTA model, this thesis studies the decision support in emergencytraffic management.ADTA-based emergency route planning method is given.Differentfrom static emergency route planning, this method takes full account of the dynamicnature of the emergency traffic, and considers the interaction between normal traffic andemergency traffic. In order to further optimizing the evacuation traffic, based onSynchronous Perturbation Stochastic Approximation algorithm, an emergencyevacuation site allocation algorithm is given. In order to alleviate the impact of theemergency on the network traffic, a post-emergency dynamic traffic re-equilibriummethod is given. After that, a diversion traffic guidance generation algorithm based onpath traffic volume differences of different scenarios is given.Finally, by performing requirement analysis, the thesis designs the emergencyrescue and evacuation decision support system. A real-time intelligent transportationsystem framework is proposed, and the physical structure, logic architecture and the design of different function modules of the system is given.The prototype system isimplemented on.NET Framework3.5using C#programming language.After that, acase study is performed using Beijing road network and traffic surveillance data. Theresult of the case study demonstrates the feasibility and effectiveness of the system.The results of this study show that by introducing the intelligent transportationsystems in the disaster relief process, and using dynamic traffic assignment model tosolve the emergency rescue and evacuation decision support problems, the efficiencyand quality of emergency transportationcan be improved. The thesis provides newmethods, techniques and tools to promote information technology and scientificmanagement in emergency response,which has important theoretical significance andapplication value.