An important concern in supply chain management is about network design, involving factories, central warehouses, regional warehouses as well as customers. The best strategy has to be ascer- tained for distributing products within this network. The objective is to select the optimal num- bers and locations of central and regional warehouses such that all customer demands are satis- fied at minimum total costs of the network. An extension of existing approaches is to take into account aspects of postponement, in particular regarding the problem of postponing activities like assembling half-finished goods or packaging them. A mixed integer programming model is provided and it is demonstrated that the resulting formulation can be used to solve realistic problem instances with commercially available mathematical programming software