Product variation and customization is a trend in current market-oriented manufacturing environment. Product configuration is the key technology in this customization environment. But complex customer requirements are still hard to resolve. So the purpose of this study is to offering a new intelligent configuration algorithm to this problem, which is based on Product Family Genealogy Model (PFG). Characteristics of customer group, modules and product platform are discussed to build a PFG model - a hierarchy of reasonable configuration models. Based on it, both the customer requirements and constraints are classified and treated respectively. So the product configuration task can be easily accomplished by four sequenced stages i.e. “PFG position”, “key configuration”, “basic configuration”, and “form configuration”. Finally, an example of desktop configuration is presented to demonstrate this approach.