The role and limitations of quantitative techniques inteh strategic design of Global Logistics System (GLS) are discussed in this paper, with emphasis on the use of Mixed Integer Programming (MIP) models. Factors in GLS design are classified into three categories, according to their attainable modeling accuracy : factors that can be modeled with a high degree of accuracy; factors that can be modeled with a high degree of accuracy; factors that, to date, can be modeled adequately by making some assumptions, and factors that, to date, are very difficult to include in those formulations. The importance of data acquisition, data management,and data consistency is also illustrated, focusing on the impact of uncertainties on the impact of uncertainties on the design and potimization of GLS. A MIP model is presented to illustrate these ideas and the role of this kind of mathematical models in the global supply chain design process. Exchange rates and suppliers' reliability are included in the model, as examples of the system fluctuations. By solving a small instance of the problem, the importance of uncertainties and the usefulness of MIP models are demonstrated.