Case-Based Reasoning (CBR) is an AI technology based on the premise that problems can be solved by adapting previous solutions to new requirements. The basic idea is to retrieve problem-solving experience that has been stored in a case base, adapt and reuse it to solve new problems and, if unsuccessful, learn from failures. Current process models present CBR as a low maintenance AI technology and ignore the effort required to develop and use the system. However, a successful CBR system must be integrated into an organizational structure and used by more than one person, which means that processes for continuous knowledge acquisition, utilization and maintenance must be put in place. In this presentation, I give an overview of the development of HOMER, a case-based sustem for CAD/CAM help-desk support at DaimlerChrysler in Sindelfingen. I describe the knowledge-modeling approach and its influence on the system architecture, as well as the different user roles and the help-desk tool itself.  Based on the experiences made during HOMER's development, I analyze the shortcomings of classical case-based process models, and I consider the managerial, organizational, and technical processes related to the development and utilization of case-based systems.
 

Return to the seminar schedule