Reconfigurable manufacturing systems (RMS) are designed to be able to be reconfigured to produce new items. Nevertheless, reconfigurations of a RMS may be time consuming and costly if they are not considered since early steps of new item design. This work describes a decision support system to automatically generate and test configurations for such RMSs based on a computeraided design (CAD) model of a new product. The proposed methodology consists of two main steps. First, a matrix of possible assembly plans (taking into account resource/tool compatibility, geometric constraints, …) is generated with a skillbased comparison between the new item and the production resources. Second, the assembly plan with minimum reconfiguration cost is found through mathematical optimization. The solution is analyzed by a simulation model in the end. Experiments performed on small use case validate the proposed methodology.