Pavement design computer programs typically use three-dimensional finite element (3D-FE) models to calculate critical pavement stresses. However, these models are computationally intensive, making their direct use in design programs that require numerous iterations of stress calculations impractical.

This webinar introduces a fast-running model, developed using machine learning (ML) techniques, to estimate critical stresses in concrete pavements. The ML-based model serves as an efficient substitute for 3D-FE analysis, providing stress estimates for top-down cracking in airport rigid pavement design with significantly reduced computation time—up to two orders of magnitude faster—while maintaining comparable accuracy.

This webinar will discuss:

• Overview of the Federal Aviation Administration (FAA) rigid pavement design procedures.
• Top-down cracking failure in airport rigid pavement.
• Technical challenges in designing for top-down cracking.
• Application of ML is physics-based modeling.
• Benefits of incorporating ML-based models in pavement design.