Designing an Artificial Immune Systems for Intelligent Maintenance Systems

Abstract

Engineering Immune Systems (EIS) are systems able to react against disturbances, to detect anomalous events and to adapt to environment changes in order to keep a stable state. Autonomous Computing and Artificial Immune Systems are biological inspired IT systems that could compose am EIS. Artificial Immune Intelligent Maintenance System (AI2MS) is a system architecture proposal of a distributed Intelligent Maintenance System (IMS) using Artificial Immune Systems concepts. AI2MS is modeled and implemented using multi-agent systems, where every autonomous IMS is mapped to a set of local agents, while the communication and decision process between IMSs are mapped to global agents. Diagnose procedures are performed locally but, if this process find patterns that have unknown meaning, a collaborative diagnose process is started. This paper describes this immune inspired collaborative diagnostic strategy and its implementation by AI2MS agents. Preliminary results are presented, deriving from the application of the proposed approach to a case study.