Surface array electrode (SAE) provides high performance, which improves the stimulation selectivity and controllability. The electrode design and stimulating waveform have a marked effect on the performance of functional neuromuscular stimulation (FNS). For motor rehabilitation of hand, the simplified forearm model with different layers was established in this paper, and finite element method (FEM) was used to simulate the distribution of electric fields within the forearm while a SAE with all large contacts was separated by small ones outputs direct current signals to achieve cathodic stimulation. Moreover, the activation function (AF) indicates the effect of applied electric field on the activity of the nerve axon, and then the ratio of the maximum to the multiplication of halfwidths of AF for the deep axon in the forearm was used to evaluate the stimulation selectivity. In order to achieve a perfect performance, the particle swarm optimization (PSO) algorithm was used to optimize the design of SAE. Results show that the optimal performance is achieved for SAE when the sizes of large square contacts and small ones are 9.80 mm×9.80 mm and 10.72 mm×10.72 mm, respectirely, while the maximum of performance index for simulation selectivity is 11 252.68 V/m^4. In addition, different stimulation waveforms are investigated based on the timevarying maximum of AF for the deep axon in the forearm. Compared with other stimulation waveforms, the biphasic chargebalanced rectangular pulse produced a slightly larger maximum of AF, i.e. 3.448 V/m², which has a positive effect on the activation of axons. The FEM simulations provide foundation for the design of SAE and clinical application.