Santiago de Chile has one of the largest electric bus fleet in the world. One of the main problems that the public transport system must address is optimal dispatch of these buses, taking into account a given schedule of departures and travel times that are considered stochastic. In this work, we plan the dispatches of two depots in a full day of operation, divided into certain time windows, in which there are certain behaviors of travel times and availability of buses. A multistage stochastic optimization model is formulated and a particular instance is solved in order to determine the respective sequence of dispatches and times when the battery of each bus will be charged in order to meet the initially proposed departure plan.