Tube model predictive control (TMPC) is a robust control approach that ensures closed-loop stability and constraint satisfaction under uncertainties by restricting system trajectories within an invariant tube. Despite its strong theoretical guarantees, TMPC is often computationally demanding due to complex tube representations and additional optimization constraints. This project aims to develop numerically efficient TMPC design methods based on inner tube approximations. The proposed approaches balance approximation accuracy and computational efficiency, enabling deployment in systems with limited computational resources. The results will contribute to intelligent control solutions, particularly in the process industry.