Research On Emergency Logistics Scheduling Model Based On Uncertain Traffic State

China has a vast territory and is seriously affected by natural disasters.At the same time,with the progress of science and technology,disasters caused by human accidents often occur in our production and life.In order to reduce the damages caused by disasters,three stages are used to deal with such emergencies: preparation stage,process stage and recovery stage.Among the stages,the process stage is the most important step in emergency rescue,and the efficiency of dispatching would directly affect the rescue effect.In order to better deal with the emergency events caused by disasters in production and life,this paper designs the corresponding emergency logistics scheduling model for the problems in the scheduling.Due to the unknown of the disaster,the timing of the disaster is uncertain.In the urban rescue,the traffic state of the road at different times will have a great influence on the scheduling plans.The traffic state not only changes over time,but it is also highly uncertain in itself.Many factors will have an impact on the traffic state,such as weather and traffic restrictions.In order to inference prediction and reduce the impact of the uncertain traffic status on the scheduling,this paper uses the cloud model to deal with the uncertainty.According to the numerical characteristics of traffic spatial-temporal data,several different strategies are designed to extract the numerical characteristics from the data,then the algorithms are used to inference and prediction.To verify the proposed model,the taxi spatial-temporal data of Beijing are used in this paper.The results show that the hybrid cloud model algorithm based on data separation has superior performance in dealing the road traffic state,and the total error of the proposed algorithm is 13.55% lower than that of other algorithms.Secondly,the uncertain traffic state and real road network are introduced in the scheduling model,and considering the two most important problems in rescue process,rescue time consumption and rescue cost.A multi-objective emergency logistics scheduling model based on uncertain traffic state is proposed in this paper.According to the characteristics of the proposed model,a new individual coding and evolutionary operator are designed,and three different improved multiobjective evolutionary algorithms are used to solve the proposed model.Experiments show that the proposed model can provide effective scheduling plans for decision-makers in three different rescue times.Finally,aiming at the road insecurity and driving safety in the scheduling process,a many-objective logistic scheduling model based on uncertain traffic state is rebuilt.In the many-objective scheduling model,new individual coding and evolutionary operators are used to improve the many-objective evolutionary algorithms,and the improved NSGA-III,MOEA/D,SPEA2,SPEA2-SDE,Gr EA and RVEA are used to solve the proposed model.The experimental results show that under three different rescue times,the model can still give a series effective scheduling plan.