The Research Of Multi-Agent Based Analytical CRM

With the deepening of the economic globalization and ever-increasing competition among enterprises, the difference of products and services are getting smaller and smaller. Business operation philosophy had changed significantly. The most prominent one is that the business practice has shifted from"products and marketing-centered"to"customer-centered". Nowadays, companys which can manage customer resources effectively, know the demand of customers, keep sound relation with customers, and customize the products, will take the upper hand in competition. It is under such circumstances that Customer Relationship Management comes into being.Customer Relationship Management (CRM) is a new management mode. It makes use of information technology to form an automatic solution by reorganizing and redesigning of the"customer-centered"practice. From the system application angle, CRM is divided into three categories which provide different functions respectively. Collaborative CRM provides customer access service, Operational CRM provides business's flow management, and Analytical CRM provides customer analysis. Analytical CRM supports the other two categories of CRM and is the core of the whole CRM system.Multi-Agent System (MAS) is a loose coupling Agent network, and these Agents can solve problems exceeding each single Agent's ability by cooperation. Multi-Agent System is autonomic, distributed and coordinative, and has the ability of self-organizing, learning and reasoning. So we can make full use of Agent's characteristics such as intelligent, cooperative and adaptive to build a reliable, flexible, and extensible Analytical CRM with Multi-Agent technology.By analyzing the advantage of Analytical CRM based on Multi-Agent, we introduced MAS technology to the architecture of Analytical CRM.The main content of the paper contains:1) From the management angle, we set up a generic customer analysis management framework, it takes the analytical tactics of customer analysis into a uniform framework.2) From the system architecture and design angle, we design a system architecture of Analytical CRM based on MAS, and make a detailed discussion of the core Agents'model structure and their working principle. 3) The research focuses on customer segmentation which is the most important part of Analytical CRM. Put forward the methodology of customer value segmentation based on customer value, and present the detailed implemental methods.4) On the MAS platform JADE, we design and develop a prototype system of customer value segmentation based on AHP method. Making use of Agent's characteristic such as autonomous,collaborative and self-adaptive, this prototype system can improve the flexibility of the methodology, and initially demonstrate the feasibility and effectiveness of MAS technology introducing into Analytical CRM.