A company must purchase a lot of diverse components and raw material from different upstream suppliers to manufacture or assemble its products. Under this situation the supplier selection has become a critical issue for the purchasing department.The selection of suppliers depends on number of criteria and the challenge is to optimize selection process based on critical criteria and select the best supplier(s). During supplier selection process initial screening of potential suppliers from a large set is vital and the determination of prospective supplier is largely dependent on the criteria chosen of such pre-qualification. In the literature, many judgments based methods are proposed and derived criteria selection from the opinion of either the customers or the experts. All these techniques use the knowledge and experience of the decision makers. These methods inherit certain degree of uncertainty due to complex supply chain structure. The extraction of hidden knowledge is one of the most important tools to address such uncertainty and data mining is one such concept to account for such uncertainty and it has been found applicable in many scenarios. The proposed research aims to introduce a data mining approach, to discover the hidden relationships among the supplier's pre-qualification data with the overall supplier rating that have been derived after observation of previously executed work for a period of time. It provides an overview that how supplier's initial strength influence its final work performance.