In the design of modern supply chains, integrating supplier selection, order splitting, transportation allocation and inventory control is a challenging issue. Existing optimisation approaches handle the different problems separately and for the sake of solvability, neglect impact of strategic decisions on operational decisions and do not take into account uncertainties. In this paper, a simulation-based evolutionary multi-objective optimisation approach is proposed to deal with this problem. The approach consists of an optimiser and a simulator. The optimiser, based on a multi-objective genetic algorithm, is used to find best-compromised solutions with respect to various criteria, such as the total cost and customer service level. Candidate solutions are evaluated through simulation, which enables realistic evaluation taking into account uncertainties and dynamics along the whole supply chain. A simple case study from the textile industry is presented to illustrate the applicability of the proposed approach for the real-world applications.