The paper describes the results of the Long-Term Pavement Performance (LTPP) study carried out in Finland and Sweden on the GPS-1 experiment (asphalt concrete on granular base). The main results include new pavement deterioration models which are based on a failure time approach using censored data on 64 test sections. A large number of independent variables were examined to identify factors which explain most of the pavement deterioration on wheel paths (traffic related distress) and on a whole pavement surface (traffic and climate related distress). The most important factors explaining deterioration included tensile strain at the bottom of the asphalt layer, or the surface curvature index calculated from FWD measurements and the freezing index. In addition to deterioration models, a new neural network approach to calculate tensile strains on the basis of measured deflection bowl and asphalt layer thickness is introduced.