摘要: |
The objective of this study was to develop models for predicting the cumulative number of load applications to the initiation of cracking for flexible pavements. Crack initiation was defined as the first appearance of cracking in the wheel paths. Data used in the study consisted of 64 in-service pavement sections in Finland and Sweden that were on their first performance period. 21 test sections were part of the Nordic SHRP-LIPP study (Strategic Highway Research Program - Long-Term Pavement Performance), which were monitored between 1990 - 1997. 43 test sections were part of the Swedish national study and these sections were monitored between 1985 -1997.
Test section specific data included information about traffic, climate, pavement structure, surface condition and structural strength (deflection). In addition, pavement responses were calculated for each section. Factors explaining crack initiation were determined in the modeling phase of the study. Pavement response based explanatory factors included, among other things, tensile strain at the bottom of the asphalt layer and other deflection based factors (i.e. delta deflection, deflection ratio, area basin factor, curvature basin factor, surface curvature radius, bearing capacity).
The modeling was carried out using a survival analysis approach. The method utilizes censored data, which can be left censored (occurrence of cracking on a pavement surfacing was earlier than the time it was first observed), right censored (cracking had not occurred during the monitoring period) and non-censored (exact time of cracking is known). The results showed that data came fi-om a Weibull distribution with an increasing risk of failure as pavement age increases.
The most significant models, that predict the cumulative number of traffic loadings to the initiation of cracking in the wheel paths, were based on the model including two independent variables related to pavement response and ageing, which is a relationship between a response variable and the annual cumulative number of traffic loadings. The best model explaining crack initiation was based on tensile strain at the bottom of the asphalt layer together with the ageing factor. In addition, models based on delta deflection (deflection difference) at a distance of 300 mm and deflection ratio at a distance of 1200 mm also explained crack initiation together with the ageing factor.
The delta deflection model developed in the study was compared to the corresponding model derived from the European ®ARIS project The results indicated that the study model predicts a lower number of load applications before crack initiates on a pavement surfacing than would be expected on the basis of the PARIS model for strong pavements with high traffic volume. The reason was that the study data did not include test sections with strong pavements (low delta deflection) and high traffic volume.
|