In this research study, a comparative examination on the mechanical properties of AA6063 has been carried out after having been processed by isothermal forging, using plane-shape dies and starting from different initial deformation states. It introduces the novelty of employing experimental data obtained from the isothermal forging so as to model the flow rules of AA6063 processed by equal channel angular pressing taking temperature into account and using artificial neural networks to this end. Subsequently, these flow rules are employed to model the behaviour of AA6063 by means of finite element simulation. Furthermore, a validation of the experimental results is made with those obtained from the simulations using the flow rules attained with the neural networks. It is shown that it is possible to achieve higher precision than with traditional fitting methods of flow rules. In addition, this study presents the novelty of carrying out a comparative study between different starting material states, prior to forging, including among these material previously processed by the severe plastic deformation process, which is referred to as equal channel angular pressing. Moreover, the experimental results obtained when processing the aluminium alloy by equal channel angular pressing are compared to those states, which correspond to the traditional way of working on aluminium alloys, which can be quenched and aged for the purpose of improving their mechanical properties.