A modern optimization algorithm is used for maximizing the performance of solid oxide fuel cell. At first, the cell is modeled using Artificial Neural Networks based on the experimental data sets. Then, a robust, simple, and quick optimization algorithm named radial movement optimizer is used for determining the optimal operating parameters of the cell. The cell parameters used in the optimization process are anode support layer thickness, anode porosity, electrolyte thickness, and cathode interlayer thickness. The optimization obtained results are compared with the previous optimized experimental results and those obtained using genetic algorithm. Two sets of the parameters' constraints are considered during the optimization process. In the first set, the resulting optimal cell parameters are 0.5 mm, 76%, 20 mu m, and 62.26 mu m for anode thickness, anode porosity, electrolyte thickness, and cathode thickness respectively. Under this condition, the cell maximum power density is 1.8 W/cm(2), 2.25 W/cm(2) and 2.72 W/cm(2) for experimentally, genetic algorithm and the proposed strategy, respectively. This implies that using the proposed method increases the power density by 33.8% and 17.28% over the experimental and genetic, respectively. In the second set, the proposed optimizer increases the maximum power by 28.85% compared with genetic optimizer. (C) 2019 Elsevier Ltd. All rights reserved.