A rule-based approach for evolution of AADL models based on changes in functional requirements

Abstract

The increasing complexity of software systems makes change management costly and time consuming. To ensure the cost-effective system longevity and endurance, it is crucial to apply change management in the early stages of software development. In this paper, we introduce a rule-based approach to make software architecture evolving based on counter examples provided by a model checker for violated, changing functional requirements. The approach works on software architecture in AADL and is based on verifying functional requirements using Maude model checker. Our goal is to provide guidelines to the architect about potential changes. Using an industrial example, we show how our approach helps in determining changes to produce a new version of the architecture.