Study On The Methods Of Service Classification And Evaluation Of The Expressway In Beijing

Service level of road is an indicator, which measures operating conditions of traffic flow and services quality that drivers and passengers experience. In this paper, we will begin with the three elements of traffic flow, create a Beijing expressway classification standard and evaluation method of the level of transport service, then display the transport status.As a result of a lot of external environment interference, there will be some anomalies of the raw data, the more accurate input data should get better output. So In this paper, we will choose and repair the raw data using the method of the data quality of screening and time-related, and provide complete data for the following classification standards and assessment.In this papar, we research a service level classification standard based on the Semi-supervised clustering, which is a new hotspot of Machine learning methods. Through the Delphi method, it can generate seed clusters initiating the algorithm and then aid clustering process to detect known and unknown attacks, then divide the service level.After creating the grading standards of expressway service level, we use the service level evaluation methods based on BP neural network, and build the network with matlab. Besides, we take some mature road as a sample of sections to training the BP neural network, and then give a final result. The results of the FIS evaluation model make a further improvement of the BP neural network's accuracy. At last, we use the above-mentioned expressway by Delphi method to verify the reasonableness of classification.Through the grading service level and evaluation of Beijing expressway using the traffic flow three-parameter, it can provide a scientific basis for the traffic condition and play an important role in traffic decision-making.