International Journal of Computing and Digital Systems | 2026
Authors: Djeddi C.; Sellami S.; Nasri Z.; Zarour N.
DOI: 10.12785/ijcds/1571146058
Journal: International Journal of Computing and Digital Systems
Year: 2026
Publisher: University of Bahrain
Document Type: Article
Open Access: All Open Access; Gold Open Access
Cited by: 1
This paper presents AIMLKAOS, an extension of the KAOS method specifically designed for Artificial Intelligence/Machine learning (AI/ML-based Systems). The new method aims to enhance the elicitation of requirements for AI/ML-based Systems by addressing challenges unique to these systems, such as dynamic learning, large-scale data processing, and ethical considerations. Despite previous extensions of KAOS, no tailored solution exists that adequately captures the specific requirements of AI/ML-based Systems. The study comprises three key phases: (i) a review of existing KAOS extensions in various fields, revealing the need for an extension that adequately addresses the unique requirements of AI/ML-based Systems; (ii) the development of a systematic, eleven-steps process for extending KAOS, ensuring adaptability across domains by addressing field-specific requirements (e.g., Big Data, AI/ML, and cybersecurity), and (iii) the application of this process to create AIMLKAOS as a solution for AI/ML-based Systems requirements engineering. Following this structured approach, we developed AIMLKAOS. Our results, based on the case study, confirm that AIMLKAOS significantly improves the precision and adaptability of requirements elicitation for AI/ML-based Systems, thereby advancing requirements engineering practices in this domain. © 2026, University of Bahrain. All rights reserved.
AI/ML based-systems; AIMLKAOS; Bigraph-based Formal Verification; KAOS; Requirements Engineering