Study On Sustainable Development Oriented Emergency Resource Allocation Decisions In Large-scale Natural Disasters

Strategy design with regard to emergency resource allocation including emergency organization assignment and relief distribution is the most critical task during decision-making for emergency response.In order to alleviate the suffering of the survivors and contribute to the coordinated development of social,economic and environmental sustainability,how to design sustainable development oriented emergency resource allocation strategies is an urgent and critical issue.In this context,this dissertation focuses on the problems including task-oriented emergency organization allocation and relief distribution from the perspective of intergovernmental relations among beneficiaries with the main clue of sustainable development philosophies.Firstly,a novel hierarchical structure model is proposed based on philosophies of sustainable development,semantic bill of X theory,multi-objective and bi-level optimization theory.It fundamentally answers the internal mechanism of establishment of emergency resource allocation model,improves the separated situation between theoretical framework and mathematical model,and fills the gaps between strategic decisions and micro operations.Secondly,social,environmental and economic sustainability is respectively characterized by weighted completion time,carbon emission and emergency cost.Emergency organization allocation problem taking into account eligibility restrictions,workloads and visiting route is formulated as a novel multi-objective 0-1 integer programming model,which improves coordination efficiency from the point of view of horizontal intergovernmental relations.Then,the hybrid global criteria method(HGCM)incorporating branch-and-bound approach(BBA)and a novel improved genetic algorithm(IGA)with elite strategy are designed to solve this model.Results indicate both HGCM and IGA have a better performance,and IGA outperforms GA.Thirdly,a novel bi-level 0-1 integer programming model for emergency organization allocation is built.It aims to minimize total weighted completion times,total carbon emissions and total emergency costs on the upper level,as well as maximize total service utility perceived satisfaction on the lower level.It tries to solve insufficient linkage among decision-agents from different levels.Then,a novel nested genetic algorithm(NGA)with BBA is proposed to solve this model.Computational results show that NGA against GA has a better performance.Fourthly,relief trans-regional distribution problem is formulated as a novel integer programming model to minimize total travel times,total carbon emissions and total emergency costs.It enriches the model library in disaster operations management.Then,both HGCM incorporating BBA and novel improved plant growth simulation algorithm(IPGSA)with elite strategy are designed to solve this model.Results demonstrate that HGCM and IPGSA have the potential advantages on solution quality and computation time.Finally,a novel bi-level integer programming model is used to formulate the task-oriented relief trans-regional distribution problem.It aims to minimize total weighted unmet demand rate,total carbon emissions and total emergency costs on the first level,as well as maximize total survivors' perceived satisfaction on the second level.Such model can benefit the rationality of distribution scheme.And then,a HGCM incorporating primal-dual algorithm and BBA,as well as a novel nested plant growth simulation algorithm(NPGSA)are constructed to solve the proposed model.Computational results indicate both HGCM and NPGSA can obtain optimal distribution scheme,and NPGSA outperforms PGSA.