Research On Two-stage Emergency Materials Scheduling Considering Post Disaster Traffic Congestion

In recent years,various disasters have brought great harm to people’s life and economic development.Therefore,the research of emergency material scheduling under sudden disasters has been paid more and more attention,and scientific and reasonable emergency material scheduling is an important guarantee to reduce casualties and economic losses.The occurrence of disasters often leads to the destruction and congestion of the traffic network.If the destruction factors of disasters on the road network can be considered in the emergency material scheduling,the efficiency of material distribution and the satisfaction of the victims can be improved.Therefore,considering the congestion factors caused by road damage under disasters,this paper constructs a stochastic mixed integer nonlinear programming model of multi commodity twostage scenario of post disaster traffic congestion,which has important theoretical and practical significance for emergency material scheduling under disasters.In this paper,we combine the traffic congestion and BPR function,and establish a twolayer emergency material scheduling model considering the traffic congestion.The generalized Benders Decomposition method is used to analyze and solve the model,and the large-scale emergency material scheduling scheme considering the traffic congestion is obtained.Firstly,a stochastic mixed integer nonlinear programming(MINLP)model based on multi class material two-stage scenario is established by combining traffic congestion delay with BPR function.At the same time,the model in this paper is compared with the traditional model without considering congestion in the existing research.Secondly,based on the characteristics of decomposable structure and nonlinear convex function,the generalized Benders Decomposition is used(GBD)algorithm is used to analyze the model,and single cut GBD decomposition and multi cut GBD decomposition are used to decompose the model,and two different algorithms are obtained.Then,a transportation network is constructed,assuming that there are many different earthquake samples leading to a multi disaster scenario case,and each parameter is set reasonably.For the two models,the first stage material pre layout scheme and the second stage material pre layout scheme are carried out The total cost of the two stage post disaster scheduling scheme and the two stage emergency material scheduling scheme is compared;the four strategies of solving the problem(P)are compared one by one;finally,the three main road arc parameters of the BPR function,the road arc capacity level,the background traffic flow level,the water transportation cost weight and the unmet demand penalty factor,and the BPR function are also compared.The sensitivity of the two correlation factors was analyzed.The results show that,compared with the traditional model,the proposed model can obtain the lowest cost emergency material scheduling scheme under the situation of expected traffic congestion and delay;due to the L-type decomposable structure constrained by the model and the characteristics of its nonlinear convex function,the generalized Benders Decomposition algorithm is more efficient in solving large-scale cases.By introducing BPR function,this paper establishes a two-level emergency material planning and scheduling model considering traffic congestion under disasters,and proposes a new idea to solve large-scale nonlinear programming problems,which expands the research direction of disaster emergency management,and provides theoretical reference and implementation basis for disaster emergency management and large-scale case solving.