The aim of this research project is to improve control and process strategies for hot strip flatness. Objectives are: - the improvement of flatness control strategy through improved measuring devices which introduce self-learning pattern recognition methods to rolling processes, and the integration of closed loop, on-line control in the set-up calculations for shape and profile; - the evaluation of effects and their interactions on strip flatness after run-out from the finishing mill up to entry into the subsequent processing units, by investigating input variables with the aid of artificial intelligence methods; - the definition of improved set-point selections and alternative process strategies able to minimize negative modifications to strip flatness.