<p>A major challenge that hinders the application of computational models for simulating Additive Manufacturing (AM) processes to predict distortions and residual stresses is the exorbitant computational costs. A recently proposed multi-step incremental simulation approach, closely mirroring the sequential material addition in AM, offers a promising path to reducing the computational cost without sacrificing prediction accuracy. However, the effectiveness of this approach hinges on selecting an optimal step size that balances computational cost against solution error introduced during the solution mapping across the steps. This study presents a numerical optimization scheme to determine the optimal number of simulation steps. The scheme involves formulating a cost function that integrates both computation time and solution accuracy, and minimizing this function for a given AM simulation. The scheme is validated by performing simulations of Direct Energy Deposition (DED) and Wire Arc Additive Manufacturing (WAAM) of 316-L stainless steel components. Results show that assigning equal weight to solution error and computational cost in the optimization reduces simulation time by up to 37% for a complex aerospace part, with only a 3% loss in accuracy compared to a single-step simulation. In addition, the study reveals that Selective Mesh Coarsening (SMC) can introduce significant errors due to limitations in the model-mapping capabilities of current commercial Finite Element Analysis (FEA) software, highlighting the need for further improvements in this area.</p>