Parameter uncertainty is one of the most concerning issues in manufacturing systems. Information insufficiency and also flexibility in the customer needs are main reasons of the uncertainty. In this study a robust optimization approach has been implemented in order to cope with uncertainty in a cellular manufacturing system. The solution obtained using this robust model remains feasible even optimal in every uncertainty level. Moreover multi0functional machines’ reliability is considered in proposed mathematical model. Machine tool selection is done based on the machine reliability. Other features of proposed model are consideration of inter-intra cell formation, cells’reconfiguration and tools’ install-uninstallation costs. The proposed model is linearized and solved using the Gams optimization package. Based on the obtained results, machine loading volume impacts on the part process routing and also the machine intra-cell layout. Moreover, tool consumption cost is the most sensitive term to the model uncertainty.