Self-adaptive systems (SASs) are controlled by autonomic manager (AM). This ensures the QoS of such complex systems within highly dynamic and unpredictable contexts. However, the massive integration of the adaptation abilities increased drastically the complexity of the AMs. To decrease the complexity and ensure correctness adaptation, scholars propose a subdivision into multi-autonomic entities (AEs) as a design approach. In such a design approach, SASs are controlled through multiple interacting AMs implementing each the well-known MAPE-K Loop. In this article, the writers propose a refinement pattern of interacting multiple MAPE-K Loops to achieve global adaptation without conflict. The authors contribute with a notation to describe the interaction of multiple MAPE-K Control Loops. To ensure the coordinated multi-attributes control, the interaction of the AEs is achieved through the knowledge base of the MAPE-K Loops. To validate the proposed pattern, a case study in the field of Electric Vehicle is presented.
Framing a Knowledge Base for a Legal Expert System Dealing with Indeterminate Concepts
