Nowadays market competition is essentially associated to supply chain (SC) improvement. Therefore, the locus of value creation has shifted to the chain network. The strategic decision of determining the optimal SC network structure plays a vital role in the later optimization of SC operations. This work focuses on the design and retrofit of SCs. Traditional approaches available in literature addressing this problem usually utilize as departing point a rigid predefined network structure which may restrict the opportunities of adding business value. Instead, a novel flexible formulation approach which translates a recipe representation to the SC environment is proposed to solve the challenging design-planning problem of SC networks. The resulting mixed integer linear programming model is aimed to achieve the best NPV as key performance metric. The potential of the presented approach is highlighted through illustrative examples of increasing complexity. where results of traditional rigid approaches and those offered by the flexible framework are compared. The implications of exploiting this potential flexibility to improve the SC performance are highlighted and are the subject of our further research work. (C) 2009 American Institute of Chemical Engineers AIChE J. 55: 1736-1753, 2009