— In this paper, the problem of optimizing the performances of a distributed supply chain is faced by means of a decentralized architecture. In particular, a general scheme for an optimization algorithm, based on lagrangian relaxation of the precedence constraints, is presented. In this framework, the trade-off between the production cost of each node of the supply-chain and the cost of the whole network, which consists of earliness and tardiness of the received orders, is taken into account. Finally, a study case is analyzed in detail and some simulation results are presented.