The inherent volatility in the maritime freight market and the uncertainties related to the expected transport demand contribute to the complexity of the fleet size and mix problem. This work addresses the problem of the maritime fleet renewal of a Brazilian oil and gas company, for the transportation, in cabotage and international voyages, of oil products. To this end, we present a two-stage stochastic mixed-integer programming model capable of giving recommendations of which chartering contracts to be performed, considering uncertainties in freight market levels and in the forecasted volume movement. The model is able to provide fleet compositions capable of meeting the problem specifications, however, in the evaluated cases, little gain potential was observed by comparing the stochastic solutions to solutions considering expected parameter values. This work highlights a situation in which the evaluation of the computationally less demanding Waitand-See (WS) and Expected Value of Expected Solution (EEV) solutions is useful to calculate the potential gain of the stochastic model solution.