Research On Decision-making Method Of Freeway Traffic Emergency Rescue Resource Deployment

Facing sudden traffic incidents,freeway management institution has gradually presented a networked and centralized development trend on emergency rescue resource deployment in the context of network construction and operation.From the perspective of resource space layout,real-time dispatch and process management,it is an urgent problem to be solved in systematic improving emergency response speed and shortening rescue time.Emergency rescue resource deployment has the nature of global and full process.From the perspective of spatial resource layout,emergency rescue resource deployment is based on the freeway network.From the perspective of process management,emergency rescue resource deployment runs through the whole process of traffic incidents,covering facility location,resource allocation,resource dispatching,and case retrieval.Taking freeways in Jiangsu Province as an example,the dissertation analyzes the traffic flow characteristics,traffic event characteristics and emergency rescue characteristics of regional network.The emergency rescue demands of freeway network are revealed.Based on the response process of traffic incidents,the framework of freeway emergency rescue decision-making system is established,and the specific functions of implementing process response were expounded.Applying the process-oriented modeling method,the freeway emergency rescue resource deployment decision-making system is established,and the specific contents of its decision-making objectives,coordination organization,framework,model and support system are discussed in detail.A facility optimization location model based on location cost and demand uncertainty is established.Based on the adaptive comparative analysis of the single-objective location model,a multi-objective location model based on location cost was established considering the location cost and feasibility of project implementation.By optimizing the retention of established rescue facilities,it not only saves the construction cost of the rescue facility,but also reduces the total weighted rescue time and improves the backup coverage.Combining the difference of location cost with the time-varying characteristics of rescue demand,a new multi-objective programming location optimization model is established,and an improved genetic algorithm is applied to optimize the solution.The overall evaluation of location results,program benefits and algorithm effectiveness shows that the model not only solves the problem of optimal combination of new and reserved rescue facilities,but also describes the uncertainty mechanism of emergency events.An optimal allocation model for emergency rescue resources considering secondary accidents is established.Considering the randomness of traffic incidents and the urgency of rescue time,rescue time cost is used as the emergency rescue resource allocation objective,and random variables are introduced in the constraint conditions.Combining the goal of rescue resource allocation with the randomness of allocation requirements,a stochastic programming model based on time constraint is proposed.Considering the concurrency of freeway emergency incidents and the rescue of secondary traffic accidents,an optimal allocation model for secondary traffic accident rescue resources is proposed.The particle swarm optimization algorithm is used to optimize the solution.The optimal allocation method of emergency rescue resources to meet the secondary accidents of the road network is obtained.An emergency rescue resource optimization dispatch model considering the congestion state of the rescue route is established.Based on the analysis of the traditional resource dispatch method that considers the direct cost,not only the current accident but also the potential accident,the emergency resource dispatch model based on the opportunity cost is established to deal with optimization problem of emergency resource dispatch corresponding to multiple rescue facilities.Reflecting the impact of traffic congestion on the rescue route,the rescue vehicle travel time is divided into the sum of the free flow travel time and the queue travel delay time.The delay time of the queue segment is estimated by the relative relationship between the total length of the rescue route and the queue delay time of the rescue vehicle.Based on the estimation of the travel time of the rescue vehicle,a traffic emergency resource dispatch model considering the congestion state of the rescue route is constructed to fully reflect the impact of the traffic operation state on the real-time dispatch decision of the rescue resources,making the rescue resource dispatch decision-making scheme more practical and objective.There are some shortage of the nearest neighbor method,such as that it does not fully consider the difference between the current event case and the historical case structure composition,the feature attribute weight difference,and the ambiguity of some feature attributes.As a result,the freeway feature attributes are reclassified.The feature attribute weight is established combining rough set theory with subjective method.The structural similarity is introduced to solve structural differences caused by missing attributes.The fuzzy feature attributes are calculated by similarity calculation method based on fuzzy set.In order to improve the accuracy and practicality of automatic retrieval of emergency rescue cases,the overall similarity calculation method is established combining structural similarity,feature attribute weights and attributes similarity,which can quickly and efficiently retrieve the rescue plan needed for the current emergency event.In order to improve the reliability of emergency rescue as a whole,and to promote the rationalization and automation of emergency rescue decision-making operations,the decision-making needs of freeway emergency rescue are organically integrated into the various functional modules of the system.Through the analysis of user requirements and system design,key technologies of decision support system are proposed.Then the GIS-based emergency rescue decision support system is designed.