Product family-based configuration with respect to mass customization has been well recognized. In product configuration, a Product Data Model (PDM) is widely used to capture various semantic and structural information in product configuration tasks. However, PDM is not suitable for computational purpose. Constraint Satisfaction Problem (CSP)-based product configuration is a promising approach for representing and solving product configuration problems. With proper modeling of product data, CSP-based configuration is computationally effective. Although much efforts have been made on PDM and CSP-based configuration, the relationship between data representation in both models is seldom discussed. This paper aims to build the mapping relationship between elements in PDMs and CSP in product configuration. With an analysis of the mapping relationship, subsequent deriving algorithms can be applied to realize the mapping relationship.