Assessment of Complex Adaptive System Parameters Using a Learning Classifier System Open Access
Downloadable ContentDownload PDF
This study is inspired by the current tendencies of technological changes. The main purpose of this study is to analyze the factors that impact the generation of change requirements, methods of measuring those changes and the impact of those changes on the evolution of a system. Although many books and conference reports have considered the changeability of complex adaptive systems (CASs) and have studied the stability of software system requirements, there has been no in-depth study to assess which factors most impact the generation of new requirements in CASs or the relationships among CAS changeability parameters, such as fitness, agility, flexibility, robustness and adaptability. This study analyzes the impact of CAS changeability when an external agent requires a change and analyzes how such a change affects CAS evolution. The author first reviews the general concepts of changeability and CAS, and then, the author analyzes their relationship. A model using an extension of a learning classifier system (XCSF) is presented and evaluated to meet the objectives of this research, which include 1) analyzing the factors that impact the generation of new requirements, 2) accurately assessing the impact of changeability on telecommunications-based CAS components and their evolution, and 3) gaining insight into the impact of changes on CASs to improve the engineering of such systems in the future.