Artificial Intelligence and Machine Learning for Requirements Prioritization in Complex Adaptive Systems: A Systematic Review and Strategic Research Directions

Abstract

Millions of dollars and valuable resources are being wasted annually due to ineffective requirement prioritization methods in managing the complexities of Complex Adaptive Systems (CAS). Traditional prioritization approaches struggle significantly to adapt to the dynamic and evolving nature of Critical Asset Systems (CAS), especially within critical industries such as aerospace, defense, healthcare, and environmental management. These rigid methods often lead to project delays, budget overruns, and suboptimal performance, ultimately failing to meet stakeholder expectations. This research introduces a transformative approach by integrating Artificial Intelligence (AI) and Machine Learning (ML) techniques into the prioritization process, aiming to dramatically enhance adaptability and responsiveness. Utilizing a mixed-methods strategy, this paper provides empirical validation through detailed mathematical modeling, quantitative assessments, and in-depth qualitative case studies. Notably, AI and ML-driven prioritization frameworks have shown marked improvements in real-time adaptability, operational efficiency, and resource allocation compared to traditional methods. Additionally, the paper systematically addresses critical challenges, including data integrity, algorithmic bias, and the necessity for interdisciplinary collaboration. By offering a robust analytical framework and demonstrating the effectiveness of AI-driven prioritization, this research provides valuable insights and practical guidelines for stakeholders and researchers. Ultimately, this work contributes significantly to the ongoing discourse on requirement prioritization, setting the stage for more resilient and efficient CAS management strategies in complex and dynamic operational environments.