This paper considers the supply chains' problem for agricultural companies considering multiple scenarios using the methodology of sample average approximation (SAA). We consider an established supply chain, in which the central problem consists of the determination of closure and consolidation of distribution centres. In this work, a stochastic mathematical model representative of the chain has been formulated considering constraints for nodes and variations in customers' demand. The model has been solved using the SAA methodology, which examines the integration of Monte Carlo simulation and optimisation techniques. The efficiency of the mathematical model has been proven with real information obtained from a Colombian multinational company. The results obtained confirm the model's effectiveness and the positive impact on the redesign of the supply chain of companies belonging to the agricultural sector.