Controlling supply chains involves efficiently managing the flow of materials and information throughout the supply chain, from suppliers to end customers. The main objective is not only to optimise the profitability of the entire process, while minimising costs and guaranteeing a high level of customer satisfaction, but also to increase efficiency, flexibility and responsiveness. Faced with the growing complexity of such systems, most of the control approaches developed are confronted with the thorny problem of modeling. The work presented in this thesis focuses on the development of a methodological framework for the control of supply chains where the use of a model becomes useless. Two major contributions are proposed. The first one is based on the use of differential flatness for the trajectories planning in the open-loop. The trajectories tracking as well as the uncertainties and disturbances rejection are carried out using model-free control. The second contribution employs model-free control in conjunction with recent forecasting techniques based on new interpretations of time series. Several comparative studies and in-depth analyses are carried out on linear, non-linear and delay logistic chains. The products transported may or may not be perishable. A final contribution demonstrates the relevance of model-free control to the dynamic management of hospital logistics.