The Brazilian transport matrix presents the road dominance over other modes in the movement of cargo in the country. The increase in operational efficiency and cost reduction is a factor that can increase the participation and competitiveness of trains in this context. The application of optimi zation techniques in the process of dimensioning and operational planning of the locomotive fleet has great potential for saving resources, mainly due to its high acquisition and maintenance costs. This work presents a methodology to obtain an exact soluti on for a monthly planning horizon, taking advantage of the cyclical character of the general freight trains timetables and defining the optimum fleet size necessary for its proper functioning. The methodology, implemented in Python language and applied in a scenario of a defined timetable had its resulting fleet compared with the manual sizing process carried out in a Brazilian dealership and was able to indicate the reduction of 2 locomotives in the fleet planning. Associated with this reduction, the est imated capital savings would be between 400 thousand reais and 20 million reais, for the conservative scenario of savings in maintenance costs and for the more optimistic scenario avoiding the acquisition of new locomotives. In addition to reducing the fleet and increasing the tariff competitiveness of general freight trains, the methodology creates, through scientific criteria, objective guidelines for the process of dimensioning and planning the locomotive fleet, replacing the empirical process currently applied in the aforementioned concessionaire.