Bus Routing Model For Transit-based Evacuation With Emergency Traffic Organization

Emergence events have been occurred frequently all around the world in recent years,causing lots of casualties and property losses.Therefore,governments from various countries formulated a series of measures and advocated "bus priority",in order to make the large capacity transportation,such as buses,the leading transportation of emergency evacuation.However,the existing studies concentrate on the target of minimum evacuation time,ignoring the problems over the process of evacuation,such as safety,other requirements and so on.Also,the lack of in-depth analysis and quantitative modeling lead to the low practical value.Therefore,it is necessary to make efficient evacuation management strategy in time and arrange reasonable bus evacuation route.Only in that way can we achieve the ultimate goal of improving the evacuation efficiency,maintaining evacuation order and guaranteeing the safety.Firstly,this paper summarizes emergency traffic organization strategies from three aspects:intersections,segments of roads and cities.These organization strategies include intersection-channelizing design,reversible bus lane and bus evacuation by steps.Establish the constraint models based on the constraint conditions of these strategies.Secondly,evaluate on the shortest-path evacuation model,the shortest-time evacuation model and minimum-risk evacuation model from safety,efficiency,dynamic characteristic and applicability.According to the results,we select the minimizing sum of the risk and path as the objective function.Establish the path of bus evacuation model by combing the objective function and constraint condition.Thirdly,design algorithm to get the approximate optimal result by analyzing the multi-objective model.Finally,we take Beilun District,Ningbo City as an example.It is assumed that people in the district need to be evacuated before hurricane comes.We verify the model and the algorithm by adopting the established bus evacuation path-planning model to solve the problem.