This paper is motivated by the case of a forwarder in dealing with inland transportation planning from a seaport, where inbound containers from the sea are filled with pallets, which have different destinations in the landside. Although this forwarder does not have or control any vehicle, he is required to plan the assignment of containers to intermediate depots, where the pallets are unpacked, and, in turn, the assignment of pallets to the vehicles used for the distribution from depots to consignees. We present a mathematical model supporting the forwarder in this two-echelon network to minimize assignment costs, while accounting for a balanced workload among all carriers involved in this distribution scheme. We discuss a tailor-made implementation of the model in a realistic context and present a heuristic method to solve realistic sized instances. Our computational experiments confirm the viability of this method.