Minimization of undesirable temperature gradients in all dimensions of a planar solid oxide fuel cell (SOFC) is central to the thermal management and commercialization of this electrochemical reactor. This article explores the effective operating variables on the temperature gradient in a multilayer SOFC stack and presents a trade-off optimization. Three promising approaches are numerically tested via a model-based sensitivity analysis. The numerically efficient thermo-chemical model that had already been developed by the authors for the cell scale investigations (Tang et al. Chem. Eng. J. 2016, 290, 252-262) is integrated and extended in this work to allow further thermal studies at commercial scales. Initially, the most common approach for the minimization of stack's thermal inhomogeneity, i.e., usage of the excess air, is critically assessed. Subsequently, the adjustment of inlet gas temperatures is introduced as a complementary methodology to reduce the efficiency loss due to application of excess air. As another practical approach, regulation of the oxygen fraction in the cathode coolant stream is examined from both technical and economic viewpoints. Finally, a multiobjective optimization calculation is conducted to find an operating condition in which stack's efficiency and temperature gradient are maximum and minimum, respectively.