In this work, a Mixed-Integer Linear Programming (MILP) model for daily routing of trucks in forest industry is presented. The aim is to generate truck routes at minimum cost, while logs supply is guaranteed. Models usually found in the literature assume that the trips that must be made are known in advance or are generated in a previous stage. Unlike those approaches, in this work the configuration of each trip is generated along with the routing decisions. An exact formulation is developed; neither decomposition algorithm nor heuristics are used. The proposed model is evaluated on several cases, showing in all of them that the proposed approach efficiently solves the addressed problem in a short computational time.