Study On The Prediction Model Of Preventive Maintenance In Highway Asphalt Pavement Based On Neural Network

China’s highway has been into the construction boom since 1980 s, and at the beginning of 2012, the total mileage of highway has reached 85000 kilometers. However, with the use of highway, and under the function of environment and vehicular load, the performance of road decrease. So the conservation task of highway is more and more difficult. In this condition, it is necessary for highway to develop the technology of pavement preventive maintenance.If we want to have a efficient, scientific and economic maintenance work, and meeting the two sides demand of road managers and road users, ensuring the performance of highway, it is important to apply the technology of asphalt pavement preventive maintenance. The pavement preventive maintenance is just applying the suitable function technology in the right time and the right ways. The preventive maintenance can repair the tiny disease, and get effective prevention. And the preventive maintenance can reduce the main disease and extend the highway performance and service time, save much money and time of pavement maintenance.In this paper, the concept of preventive maintenance of expressway asphalt pavement is analyzed, and the important significance of preventive maintenance for asphalt pavement is pointed out. The characteristics and principles of preventive maintenance are also analyzed. Starting from the diseases of expressway asphalt pavement types and causes of diseases, the study of asphalt pavement preventive maintenance standard and method was pointed out in macroscopic and microscopic way. This paper introduced some widely used preventive maintenance method, such as slurry seal and Micro-surfacing, Synchronous Crushed Stone Seal, and study the decision-making methods of asphalt pavement preventive maintenance which include the choice of maintenance time and maintenance measures to protect the performance of asphalt pavement.Finally, according to the advantages and disadvantages of the domestic and international asphalt pavement performance forecast model, the BP network model of neural network is used to establish the forecast model of highway preventive maintenance which get a very good results.