Analyzing requirements of knowledge management systems with the support of agent organizations

Knowledge Management (KM) is considered by many organizations a key aspect in sustaining competi-tive advantage. Designing appropriate KM process and enabling technology face considerable risks, as they must be shaped to respond to specific needs of the organizational environment. Thus, many systems are abandoned or fall into disuse because of inadequate understanding of the organizational context. This motivates current research, which tends to propose agent organizations as a useful paradigm for KM systems engineering. Following these approaches, organizations are analyzed as collective systems, composed of several agents, each of them autonomously producing and managing their own local data according to their own logic, needs, and interpreta-tive schema, i.e. their goals and beliefs. These agents interact and coordinate for goal achievement defining a coherent local knowledge system. This paper presents a novel methodology for analyzing the requirements of a KM system based on an iterative workflow where a pivotal role is played by agent-oriented modeling. Within this approach, the needs for KM systems are traced back to the organization stakeholders goals. A case study is used to illustrate the methodology. The relationship of this work with current studies in agent organizations and organizational knowledge management is also discussed. Differently from other works, this methodology aims at offering a practical guideline to the analyst, pointing out the appropriate abstractions to be used in the different phases of the analysis.