A four-phase conceptual model for supplier selection in agile supply chains (ASCs) is presented. The use of ASCs has become more common in today’s increasingly dynamic markets. However, supplier selection decisions are inherently more complex and difficult under the conditions of uncertainty and ambiguity created as supply chains form and re-form. The four phases of the model comprise Supplier selection preparation, Pre-classification, Final selection and Application feedback. It draws on a range of quantitative and qualitative techniques. These include application of the Dempster-Shafer and optimisation theories, radial basis function artificial neural networks (RBF-ANN), analytic network process-mixed integer multi-objective programming (ANP-MIMOP), Kraljic’s supplier classification matrix and the principles of continuous improvement. The resulting model offers a comprehensive and systematic approach to tackling this increasingly important task.