In this paper, a new fuzzy bi-objective mathematical model for production–distribution problem under uncertainty for a four-echelon supply chain network design involving of several suppliers, producers, distributors, customers, and a set of transportation modals with different reliability rates are presented. In addition to minimise the total cost of the supply chain, the second objective is maximising reliability rate of the whole proposed system. In order to solve the proposed model, a hybrid two-phase solution procedure is proposed based on possibilistic programming, fuzzy multi-objective programming and an efficient algorithm called self-adaptive differential evolution algorithm. Finally, an extensive comparison by a set of numerical examples with different complexity along with some sensitivity analyses has been reported to explore the efficiency of proposed model as well as the performance of proposed evolutionary algorithm.