The progressive deterioration of road pavement structural health necessitates fast, ef- ficient, and automated structural assessment to evaluate the residual life of the pavement. The accurate estimation of the pavement elastic modulus (Ei) plays a pivotal role in this process, also serving as a fundamental factor for designing resilient pavements and opti- mizing maintenance. Strategies and improving overall road safety. However, conventional methods and devices that are used to estimate Ei, such as the Falling Weight Deflectometer (FWD), have several limitations due to operational and se- curity constraints when measuring at the network level. These limitations stem from the need for traffic control during measurement data collection. To tackle these obstacles, the Traffic Speed Deflectometer has emerged as an innovative solution for continuous assessment of pavement bearing capacity, as it eliminates the need for traffic control. TSD runs at an average traffic speed of up to (80 km/h). The continuous measurements of TSD ensure much more adequate data than traditional point measure- ments, which allows for more accurate and cost-effective maintenance planning. TSD’s principle of operation is based on measuring the vertical deflection velocity (DV ) of the pavement instead of the deflection itself. Although the TSD concept holds considerable potential for pavement analysis, effectively processing the substantial data it generates presents another challenge. However, these challenges can be overcome by employing spe- cialized techniques, such as Machine Learning (ML), that excel in learning complex data patterns while offering the prospect of developing a generalized approach. Thus, this thesis presents an efficient and effective ML-based model and methodology for estimating Ei. This novel approach utilizes deflection slope DS measurements instead of the conventional deflection measurement method. The model developed in this research has been built upon data derived from industry-approved software (Alizé-LCPC) and an experimental outdoor simulator, ensuring their alignment with real-world conditions. Fur- thermore, an advanced parametric validation process was implemented to validate their performance. With this model analysis and quantitative sensitivity analysis to investigate and quantify the adverse impacts of some challenges spanning the high-dimensional TSD data, the influence of TSD measurement noise, temperature variations, and the uncertain- ties associated with Ei values. Thus, the ultimate aim is to underscore this thesis’s primary objective: having the potential to make valuable contributions to the field of pavement mechanics engineering