AbstractThis study integrated reliability-based pavement performance prediction with scheduling of rehabilitation, considering the imperfect effect of repair. Traffic and pavement structural response data were collected from the national highways of India and were analyzed statistically. Pavement reliabilities for traffic and construction quality uncertainty levels were estimated and modeled using the Weibull function. The timing and number of required rehabilitation treatments were identified using a hybrid hazard rate model coupled with a cost per unit time function. A case study was evaluated using the proposed framework. The hybrid hazard rate model captured the system’s faster degradation as the rehabilitation intervals were reduced by 2–3 years. The realistic imperfect repair cost applied was higher than the cost involved according to the traditional age-T policy. The associated risk, in terms of both cost and time, increases as the uncertainty in traffic and construction quality increases. Furthermore, the effect of traffic uncertainty was found to be significant in the case of pavement structures with lower variation in construction quality. However, beyond the 10% coefficient of variation (COV) level in construction quality, increasing traffic uncertainty from 5% to 25% COV did not influence the reliability trends considerably. Even at a higher uncertainty level of traffic, lowering the construction quality variation resulted in a substantial cost reduction and extended the cycle lengths by 2 years. The framework evaluated with a case study emphasized that although traffic uncertainty is inevitable during operation, strict quality control and assurance programs to reduce variability in pavement production minimize the maintenance and rehabilitation needs substantially.