Aluminum-based energetic materials exhibit extremely high volumetric and specific energy densities, making them applicable to a broad range of applications ranging from propulsion, pyrotechnics, and material synthesis, to thermal energy generation. In this context, our research aims to provide computational tools that integrate multiscale physical models combined with data-driven surrogates to assist experimentalist in optimizing the material for specific applications. We are developing a set of models across all time scales (nanoseconds to milliseconds) and spatial scales (atomistic level to the centimeter-scale of the experimental device) to describe as accurately as possible, the physics of combustion and dynamics of the reaction front propagation in aluminum-based energetic material. We are developing mechanistic models at the microscale (particle-resolved) to calibrate theoretically the mass and thermal transfer during the combustion process. An N-Euler approach is used to solve the unsteady Reynolds Averaged Navier-Stokes equations in such complex burning environments. We also explore statistical methods of artificial intelligence as a new research avenue to manage the acquisition and processing of data generated by these increasingly complex and computationally costly combustion models.