Analysis of airport runway pavement reliability considering temperature variation: The case of Sao Paulo-Congonhas international airport

TL;DR

This study evaluates how seasonal temperature variations affect pavement reliability at Sao Paulo Congonhas Airport, revealing that traditional methods overestimate damage and suggesting the need for climate-calibrated design approaches.

Contribution

It introduces a reliability analysis incorporating temperature variation into pavement design, demonstrating significant differences from traditional methods and highlighting the importance of local climate considerations.

Findings

Reliability under temperature variation is higher than predicted by FAA methods.

Pavement can support 2.5 times more traffic at 95% reliability.

FAARFIELD overestimates damage, increasing costs.

Abstract

Airport pavement design methods, such as those of the FAA, assume general climatic conditions, although regional temperature variations significantly affect pavement behavior, stress strain distribution, performance, and reliability. With global warming, evaluating how temperature influences pavement materials and resilience has become increasingly important. This study conducts a reliability analysis of pavement designed by traditional methods while incorporating seasonal temperature variation for Sao Paulo Congonhas International Airport CGH. After designing the pavement based on the airports traffic mix, a Monte Carlo Simulation was used to assess reliability under temperature changes. Results show that the cumulative damage factor from MCS is 79% lower than that obtained with the FAA method, indicating that aircraft cause less damage than predicted. The pavement could support 2.5 times more traffic at 95% reliability and 5.0 times more at 50% reliability. These findings suggest that, in Brazilian airports where fatigue governs design, FAARFIELD overestimates damage and increases construction costs, highlighting the need to calibrate design methods to local climatic conditions. Limitations include the deterministic temperature model and the exclusion of factors such as precipitation, thickness variability, and temperature dependent fatigue testing. Future work will aim to address these gaps and refine performance equations.