Perceiving moisture damage of asphalt mixes containing RAP using survival analysis based on Kaplan-Meier estimator and Cox proportional hazards model

Abstract

Moisture induced stress tester (MIST) has been recently developed which aims to simulate stripping due to repeated pore pressure generation. In the current work, asphalt mixes containing recycled asphalt pavement (RAP) (in proportion of 0, 10, 20, 30 and 40% by weight) were subjected to varying stress cycles of MIST (0, 1000, 2000, 3500, 5000, 10,000 cycles) and indirect tensile (IDT) strength test was conducted on such mixes. AC30 was used for control mix whereas AC10 was adopted for RAP mixes. The results of IDT strength tests assisted in computing the failure times and such times were analysed using survival analysis based on Kaplan Meier (KM) estimator and Cox Proportional Hazards (CPH) model. The results indicated that survival analysis can be effectively utilised to analyse the IDT strength test results. It was observed that increasing stress cycles of MIST increase the time to failure whereas RAP is expected to reduce the time to failure. Moreover, KM estimator indicated that RAP reduces the survival of asphalt mixes. Hazard ratio (HR^) computed using CPH model shows a decrease with the increase in RAP proportion. Both MIST tensile strength ratio (M−TSR) and HR^ revealed that RAP leads to an improvement in moisture damage resistivity of asphalt mixes. Hence, HR^ has the potential to be an effective indicator of moisture damage susceptibility of asphalt mixes. Also, “Good” and negative correlation was observed to exist between M−TSR and HR^. The current study is expected to further enhance our understanding regarding moisture induced damage of asphalt mixes and assist in effectively screening good mixes from poor ones.