Ships play the major role in bulk transportation and they need their special energy system. This paper proposes a stochastic programming method for optimal sizing of a hybrid ship power system with energy storage system (ESS), photovoltaic power (PV) and diesel generator. To account for uncertainties, in this study a two-stage stochastic mixed-integer non-linear programming is used to model the optimal design problem of the hybrid system for ships. The uncertainty of the hourly global solar irradiation and its effect on the output power of the PV system is taken into account. The probability density function of the global solar radiation follows a normal distribution. The Monte Carlo sampling approach is used to generate the scenarios with a specified probability and a proper scenario reduction method is used to decrease the computational burden of problem. Three cases are studied and the results are presented and compared.