Data Mining Techniques Applied In SRM: Case Of A Business Company

Young and promising field, data mining or knowledge discovery from data has attracted a great deal of attention in the information industry and in society as whole in recent years. This is due to the wide availability of huge amounts of data and the imminent need for turning such data into useful information and knowledge. The content of a DW is analyzed by the online analytical processing (OLAP) applications for the purpose of discovering trends, patterns of behavior, and anomalies as well as for finding hidden dependencies between data. The outcomes of theses analysis are then the basis for making various business or financial decisions.Data mining is therefore a great tool since in today's extremely competitive environment; companies are investigating new means of increasing profit by implementing so called Customer Relationship Management Systems (CRMs). Data mining techniques have been applied to CRM in order to increase customers'loyalty and from there increase their value. But the needs to continually cut costs and focus on core competencies have led companies to focus on how to improve the supply chain and leverage their supply base. Many have shift from a vertical integration model to outsourcing some or all of their production to gain the necessary competitive edge, but this inevitably creates new needs as coordination costs increases dramatically. In addition, the recent international financial crisis has heavily contributed to question the need for long-term buyer/seller relationship.The goal of this thesis is therefore to investigate solutions companies could put in place to leverage their suppliers using, what is now called:Supplier Relationship Management Systems (SRMs). This will be achieved by illustrating the key functionalities such systems should offer. To avoid reinventing the wheel, we have based our work on the one performed in the CRM field starting from the assumption that both areas are investigating the same problem but in an opposite manner. We will try to apply the data mining techniques such as the Classification and the Clustering in order to generate an efficient data mining method for SRM. As SRM is only an emerging concept with limited implementations and nearly no theoretical nor practical experience, this thesis is not willing to demonstrate or prove anything, it is more a visionary essay of what could be a SRM system and how it could combine with data mining techniques of CRM to create key synergies. To further improve the performance of our approach, we propose to lay the groundwork for future studies that will be done in the same field, by conducting an experiment on a fictive company data set with almost 100,000 records, to show that our strategy significantly improves the performance of the SRM.