In this paper we present the key concepts of an experiences based management system that assists the transformation of individual experiences into usefull knowledge. Such a framework is designed around the notion of capturing and delivering experiences for improving worker's practices and tasks. The study has been done in the area of customer relationships. In this paper we present our approach to modelling and representing individual experiences, as well as our approach to dynamically generating collective experiences, on user request, in order to assist workers in making the best use of existing experiences. The research framework called L2Corp is described in term of its various knowledge components and its associated functional processes for assisting users in authoring the knowledge as well as assisting them in the producing value added knowledge. In this framework, collective knowledge is dynamically generated using broadcasting and self-organisation mechanisms, initally developed in multiagent systems. This approach is at the feasability stage, it is promising as it constitutes a first step towards the formalisation of the know-how using AI techniques. Such know-how is the missing element in existing experiences based management systems, and is what will allow to move from passive systems illustrating the know-how to more active systems which assist its production.
展开▼