The Meaningfulness of Consensusand Context in DiagnosingEvolvable Production Systems

摘要

An Evolvable Production System (EPS) is a complex and livelyentity composed of intelligent modules that interact, through bio-inspiredmechanisms, to ensure high system availability and seamless reconfiguration.The diagnosis of such dynamic systems, characterized by constant change, pre-sents new diagnostic challenges and opportunities that can hardly be tackled bytraditional approaches. On the one hand, given the decoupled nature of the sys-tem, fault interaction and propagation are harder to detect and contain, as is thedevelopment of a global diagnostic model, on the other hand local intelligenceand careful characterization of the interactions, between the modules, can beexplored to emerge the diagnostic functionalities. The impact of simple consen-sus mechanisms (majority voting) and fault context analysis (module and itscurrent interactions states) is assessed in a multiagent-oriented application inthe assembly domain to understand the validity and contribution of this ap-proach in emerging useful self-diagnostic properties in EPS