Addressing uncertainty in pavement performance prediction: a quasi-Monte Carlo simulation-based reliability approach

Predicting the long-term performance of pavement structures is crucial for efficient infrastructure management, but uncertainties in material properties, traffic loads, and environmental conditions pose significant challenges. This research introduces a novel reliability model for pavement performance prediction using Quasi-Monte Carlo (QMC) simulation. Unlike traditional Monte Carlo methods, QMC employs deterministic sequences for enhanced sampling efficiency and accuracy in estimating probabilistic quantities like reliability indices and probabilities of failure. The study investigates the impact of QMC on pavement performance predictions, considering uncertainties in traffic load repetitions, fatigue life, and rutting life. The QMC approach demonstrates superior performance by producing stable mean estimates and significantly narrowing confidence intervals as sample size increases. This research highlights the advantages of QMC simulation in probabilistic pavement design, offering a computationally efficient framework for robust and reliable performance predictions. The findings support QMC as a valuable tool for achieving durable, safe, and cost-effective pavement infrastructure.