Research On The Optimization Methods Of Large-scale Emergency Rescue Resources Layout And Scheduling

In recent years, under the great pressure of the population, more and more effect ofhuman activities and natural resources overexploitation, global climatic conditionscontinues to deteriorate, and social order has not returned, which lead to many kinds ofnatural and man-made disasters, such as earth quake, flooding, tsunami, war, terroristattack and plague. And the disasters result in inestimable population casualties andeconomic loss. China has a vast territory and a large population, which is one of thecountries that been seriously suffered by natural disasters. In this country, there are manytypes of disasters, which has the characteristic of high-frequency, regional distribute andseasonal property. Besides, due to the weak safety foundation and science and technologyand economic progress level are inferior, the situation of safety production is severe.Faced with this situation, to maintain the sustained, rapid and sound development of thenational economy and stable social order, to strengthen the state capacity for emergencyrescue and relief has been a top priority of the Chinese government's work. During theperiod of11th five-year-plan, many research projects were setup to change the presentsituation of technical equipment lag behind and to improve the emergency rescue andrelief system.In this context, this thesis is on the basis of drawing lessons from the existingachievements about the layout and scheduling of emergency rescue resources, mainlyanalyzes the shortcomings of existing studies on the following key technology problem:estimation of emergency demand, emergency resources location and allocation,emergency resources scheduling, road emergency repair after disasters, trafficorganizations and control under emergency situation. And thoroughly studies them toprovide effective and practical models and methods for emergency rescue and relief work.Firstly, the paper uses uncertain information processing theory to estimate the emergencydemand, and provides good foundation for subsequent work. Then in view of the need ofemergency rescue efficiency, cost and fairness, the paper proposes some multi-objectivedynamic mathematical dynamic models of emergency resources location, allocation andscheduling. Furthermore, to ensure the successful completion of emergency rescue andrelief work, the paper further proposes the road emergency repair optimal method andtraffic organizations and control method under emergency situation. Through theresearch of the paper, the proposed models and methods of emergency need estimation, emergency resources location, allocation and scheduling, road emergency repair, trafficorganizations and control can offers technical guarantee for emergency rescue and reliefsystem. The specific content is as follows:(1) Emergency rescue requirement estimation based on uncertain informationprocessing theoryThe paper classifies emergency resources from different points of view and analyzestheir characteristics. Based on the researches before, the general framework ofemergency need estimation is presented, and the processes of each step are introduced indetail. More significantly, the paper analyzes the uncertainties in emergency needestimation and introduces the concepts of blind number and unknown information tohandle a variety of informaion in emergency need estimation, such as fuzzy information,grey information, random information and unknown information. At last, the emergencyneed estimation of Jiegu Town after earthquake in Yushu is taken as an example tovalidate the proposed methods.(2) Emergency resources location and allocation optimization models underdynamic demandAccording to demands of emergency rescue and relief work, the paper proposedmulti-objective emergency resources location-scale optimization model, temporarydynamic multi-point emergency rescue optimization model and multi-resource dynamicemergency rescues allocation optimization model. The number of human casualtiesduring disaster is vary with time,which has been taken into account in the above threemodels. The proposed emergency resources location-scale optimization model optimizesthe positions of supply centers as well as their scale. In addition, the model contains threeobjects: the maximum of emergency rescue efficiency, minimum of construction cost andfairness of emergency rescue, and designs unit cost utility algorithm to solve it. Theproposed temporary dynamic multi-point emergency rescue optimization model alsocontains three objects: the maximum of emergency rescue efficiency, fairness ofemergency rescue and the minimum number of temporary supply centers and combinemain object method with hill climbing method to solve it. The proposed multi-resourcedynamic emergency rescues allocation optimization model contains two objects: themaximum emergency rescue efficiency and the minimum allocation cost of the overallemergency rescue program in all stages, and combines ideal point method with unit costutility method to solve it. It should be noted that the allocation cost contains acquisitioncost, inventory cost and opportunity loss cost. At last, takes some examples to validatethe proposed models.(3) Road emergency repair model and method after disasterThe paper analyzes the characteristic of damaged road facilities and classifies it.Combined with uncertain information processing theory, the paper proposed a road emergency repair timing optimization method. The emergency path is regarded as theelementary unit of road emergency repair, which should meet tight deadline. And asimple path hierarchical search method under time constraint and a sorting algorithm ofdamaged road importance are proposed to obtain the damaged roads needed to repairduring emergency rescue. Based on this, further proposes an optimization model of roademergency repair resources. At last, take an example to validate the proposed method andmodel.(4) Multi-OD and multi-object emergency rescue scheduling optimization modelsThe paper describes various emergency transportation modes and emergencytransport network. Based on this, the paper proposes multi-OD and multi emergencyresources scheduling optimization model, which contains three objects: the maximum ofemergency rescue efficiency, minimum of transportation cost and fairness of emergencyrescue. And basic efficacy coefficient method and unit cost utility algorithm are used tosolve it. A multi shelters emergency evacuation scheduling optimization model based onvictims' characteristics is proposed and it contains two objects: the maximum of victims'transportation efficiency and transportation cost. Main object method is used to solve themodel. At last, take two examples to validate the proposed models.(5) Traffic assignment, traffic organization and control method under emergencysituationAccording to the characteristics of emergency traffic flow, the paper researches thetraffic assignment, traffic organization and control method under emergency situation.Firstly, based on the existing Logit traffic assignment models and the characters ofemergency traffic flow, a traffic assignment method based on second modified Logitmodel is proposed to allocate emergency traffic flow. Then the paper analyzes on theimplement conditions of contraflow strategy and proposes corresponding contraflowimplementation strategy in emergency rescue and relief work. Thirdly, the paper putsforward emergency traffic organization and control method under contraflow strategywith the aim to minimize total travel time of emergency traffic. At last, take an exampleto validate the proposed method and model.