Human-in-the-Loop Artificial Intelligence for Asset Management: A Systematic Review of Applications, Methods, and Research Challenges

The novelty of this research lies in the development of a conceptual framework, known as the Human In The Loop (HITL)-AssetOps Framework, which explicitly integrates human roles into the Artificial Intelligence (AI) based asset management cycle to enhance transparency and accountability in decision-making. This study focuses on the application of HITL-AI in vital sectors, including water and infrastructure, energy systems, transportation and logistics, and structures and the built environment. The proposed framework integrates human involvement into the AI-based asset management cycle, from data acquisition to strategic decision-making. This approach allows humans to act as validators, decision overrides, or interactive assistants in supporting AI models. Thus, the system not only produces accurate predictions but also provides accountable recommendations. The study employed a Systematic Literature Review approach, following the PRISMA protocol of the Scopus database. Of the 97 identified articles, 30 were selected for further analysis based on their applications, methods or algorithms, evaluation metrics, and future research challenges. The analysis results show that techniques such as digital twins, federated learning, deep learning, and genetic algorithms are more effective when integrated with human involvement.