Searches for new physics with data from the Large Hadron Collider (LHC) have not yet produced any conclusive evidence indicating the existence of interactions beyond the Standard Model. Nevertheless, these searches reveal a challenging and complex structure of the underlying interactions produced alongside the candidate processes of interest. Modern phenomenology models, in particular those that implement the mechanism of multiple partonic interaction (MPI), have proven to be indispensable tools in the study of LHC physics. The effectiveness of these models depends on a continuous development and tuning of model parameters based on the observed data. Using state-of-the-art techniques of Monte Carlo simulation and Neural Networks this proposal has the goal to develop and calibrate simulation models to describe recent measurements from the LHC experiments. This project also aims to provide predictions to guide the search for new physics at the LHC and in future particle colliders.