A MAPE-K-Based Method for Architectural Conformance Checking in Self-Adaptive Systems

Abstract

Self-adaptive systems (SASs) adjust their behavior at runtime in response to internal or external change. The MAPE-K model, which includes Monitors, Analyzers, Planners, Executors, and shared Knowledge, is a reference for structuring feedback loops. As SASs evolve, implementations can drift from the intended MAPE-K architecture, compromising planned quality attributes. Architectural Conformance Checking (ACC) addresses this risk by comparing the current implementation to a specification of the architecture. General purpose ACC techniques are flexible, but lack SAS specific semantics, leading to ambiguous specifications and missed violations. We present REMEDY, an ACC approach designed for MAPE-K based SASs. REMEDY provides three elements: a domain specific language for expressing planned architectures in MAPE-K terms, a tool that extracts the implemented architecture, and a conformance engine that reports violations. By encoding SAS domain rules and reusing MAPE-K abstractions, REMEDY reduces specification effort and lowers error rates relative to general ACC. We evaluate REMEDY through a robotic SAS case study and a controlled experiment with software engineering students. Results show higher modeling productivity and effective detection of architectural drift, supporting more reliable verification of conformance to the MAPE-K reference model.