A capacity fade prediction model has been developed for Li-ion cells based on a semi-empirical approach. Correlations for variation of capacity fade parameters with cycling were obtained with two different approaches. The first approach takes into account only the active material loss, while the second approach includes rate capability losses too. Both methods use correlations for variation of the film resistance with cycling. The state of charge (SOC) of the limiting electrode accounts for the active material loss. The diffusion coefficient of the limiting electrode was the parameter to account for the rate capability losses during cycling.