Flux Switching Motors (FSM) are gradually emerging in recent years due to their inherent simplicity, rugged construction and low cost power converter. The main focus of this paper is to optimize the design of 8/4 pole FSM by varying its pole arcs for the production of maximum torque. The sensitivity of the pole arcs on the performance of the FSM is performed by cyclically varying the stator and rotor pole arcs. A comprehensive program based on Genetic Algorithm (GA) technique is developed in MATLAB to compute the torque and the results are presented. Further the performance of the machine with the optimal combination of pole arcs is modeled and analyzed using two dimensional transient Finite Element analyses (FEA) for the production of maximum torque. The effectiveness of the results obtained from GA is validated using FEA.