Research On Multi-attribute Traffic Status Evaluation

The development of urbanization has made the pressure of urban traffic more and more.Relying on the original roads is no longer enough to support more and more urban vehicles.However,the construction of transportation facilities can not be completed overnight,so the maximum excavation the potential of urban roads is the most effective,fastest,and low.cost way to reduce urban traffic pressure.How to make the most of the utilizing roads,the first step is to understand the cyclical changes of the traffic status in the depth and comprehensive way.Based on the demand for accurate and comprehensive evaluation of urban traffic conditions,a multi.attribute based traffic status evaluation system was proposed.The description of the traffic state by a single attribute is not comprehensive enough and not accurate enough.Therefore,the four attributes of cumulative flow,average speed,queue length and travel are selected to describe the urban traffic state to prevent special or accidental traffic conditions.The situation in which the evaluation of the traffic status is biased by the single attribute appears.However,it is not generally considered that the importance of each attribute for traffic status evaluation is the identical.Therefore,the most important attribute of multi.attribute traffic status evaluation is the distribution of attribute weights.In this paper,information entropy and analytic hierarchy process are mainly used to determine multi.attribute weights.The most activity vehicles in the city are taxis.They travel through every corner of the city every day,generating a large amount of GPS data.Using the GPS data generated by city taxis,the information entropy is used to calculate the entropy weight of multiple attributes of urban traffic status evaluation.On the other hand,the perception of traffic status comes from human beings,so the expert opinion is very important.Combining the expert opinion with the analytic hierarchy process tocalculate the subjective weight of the traffic status evaluation attribute,and finally normalizing the multi.attribute weight.At the same time,due to the fuzzy characteristics of traffic status evaluation,the fuzzy judgment method is applied to construct the traffic status evaluation model.Finally,VISSIM is used to simulate the traffic state.The comparison with the experienced drivers on the traffic status of the simulated road conditions is found.The model is almost identical to its judgment,and it is almost consistent with the evaluation results of the open source map platform.Therefore,the evaluation results of the traffic status evaluation model in this paper are close to the reality.