Optimization Of Emergency Evacuation Route Considering Random Damage Of Road Network

As my country enters a new era of socialism,the rate of urbanization is accelerating,and the urban population density is also increasing.Once an emergency occurs,a large number of people in the region need to obtain better route information for evacuation in a short period of time.Establishing an effective emergency evacuation route in advance helps to organize and allocate emergency rescue vehicles in an orderly and reasonable manner after an emergency,and evacuate people efficiently.A reasonable and efficient emergency evacuation path can evacuate as many people as possible from the disaster area as quickly as possible to ensure the safety of the evacuees.Under emergencies,the destruction of the road network is random.When formulating the optimal emergency evacuation path under emergencies,the randomness of road network damage is considered,so that the formulated emergency evacuation path is more reliable and more in line with the actual evacuation needs.Firstly,this paper analyzes the research status of emergency evacuation paths at home and abroad,expounds the research methods,contents and models of domestic and foreign scholars on emergency evacuation paths,and proposes the lessons learned from the current research and the deficiencies in the existing research.By analyzing the characteristics of evacuation traffic flow and the road section capacity model and road section impedance model under evacuation conditions,this paper lays a theoretical foundation for the subsequent optimization model of emergency evacuation routes for road network damage.Secondly,this paper adopts the basic theory of graph and network in operations research,and abstracts the traffic road network structure into a network graph.This method can display the evacuation road network structure valuably and is also convenient for the design of subsequent algorithms.In the modeling process,the traffic capacity of the road section is stochastically constrained,In this paper,the random capacity of road section under emergency is expressed by θC,which is the θ different random value between [0,1].and the constraints of the model include flow balance,capacity constraints,sub-cycle constraints,and vehicle service point constraints.Taking the minimum total system evacuation time as the objective function of the model,in order to consider the reliability of the evacuation plan,this paper improves the objective function of the model by referring to the Bertsimas robust optimization model.In addition,according to the characteristics of the model,the genetic algorithm is used to design the algorithm of the model.In order to keep the path with the smallest evacuation time,the algorithm chooses the reciprocal of the objective function as the individual fitness function.The randomness of road traffic capacity caused by emergencies should be fully considered in the model algorithm design,and the algorithm program design is carried out by using the method of random constraint of road traffic capacity in the built model.The algorithm design also includes the design of encoding,decoding,selection strategy,crossover operator and mutation operator.Finally,two examples are used for verification.Finally,this paper selects the area south of the Yellow River in Chengguan District,Lanzhou City as the emergency evacuation area for case study.It starts from three stages: data investigation,model solution and result analysis.Important data such as the road network structure,length of road section,and road traffic capacity in the selected area are obtained through data investigation;the model solving part solves the actual data surveyed by algorithm to obtain the optimal solution of the model;analyzes the results according to the operation results,and finally determines An emergency evacuation path with high evacuation efficiency and strong robustness under random damage of the road network in the south of the Yellow River in Chengguan District is obtained.The calculation results show that the built model has strong applicability to the actual network.