Next-Generation Computational Intelligence in Tunneling and Subsurface Infrastructure

Tunneling and subsurface infrastructure development are essential to modern urbanization and sustainable transportation networks. However, the complex and heterogeneous nature of underground environments presents significant challenges in design, construction, and long-term performance. Traditional analytical and empirical approaches often fall short in capturing the intricate interactions among geological, structural, and operational parameters. In recent years, computational intelligence, encompassing artificial intelligence (AI), machine learning, and bio-inspired optimization algorithms, has emerged as a powerful tool to address these complexities. These advanced methods enable data-driven predictions, real-time monitoring, risk assessment, and automation in tunneling processes. As the tunneling industry increasingly embraces digital transformation, the integration of next-generation computational intelligence techniques promises to revolutionize underground engineering practices, enhance safety, and improve decision-making across the project lifecycle. This Research Topic aims to highlight recent advancements and foster innovation at the intersection of computational intelligence and underground infrastructure.

Despite significant advances in tunneling and underground infrastructure engineering, the industry continues to face persistent challenges related to uncertainty in geological conditions, cost overruns, construction delays, and safety risks. Traditional modeling and design approaches often rely on simplified assumptions that fail to capture the dynamic, nonlinear, and highly variable nature of subsurface environments. As underground projects become more complex and data-rich, there is a pressing need for intelligent, adaptive solutions that can integrate diverse datasets, learn from historical trends, and optimize decision-making in real time. Recent developments in computational intelligence - such as deep learning, evolutionary algorithms, reinforcement learning, and hybrid AI models - offer promising avenues to address these issues. These techniques can be employed for ground condition classification, tunnel boring machine (TBM) performance prediction, risk analysis, structural health monitoring, and automated design optimization. This Research Topic seeks to bring together cutting-edge studies that explore and demonstrate the application of next-generation computational intelligence tools to enhance efficiency, resilience, and sustainability in tunneling and subsurface infrastructure projects.

This Research Topic invites original research articles, reviews, case studies, and technical notes that explore the application of computational intelligence in tunneling and subsurface infrastructure. We encourage contributions addressing a wide range of themes, including but not limited to: AI- and ML-based ground behavior prediction, tunnel boring machine (TBM) performance analysis, geotechnical data interpretation, structural health monitoring, intelligent sensing systems, optimization in tunnel design and alignment, risk assessment, and digital twin technologies. Manuscripts may focus on theoretical advancements, algorithm development, or practical implementations using real-world or synthetic data. Studies integrating computational models with sensor technologies, big data analytics, and real-time decision support systems are particularly welcome. Submissions should clearly demonstrate how the proposed techniques contribute to improved accuracy, safety, cost-effectiveness, or sustainability in underground engineering. Both academic and industry-led research that bridges the gap between innovation and practical application are encouraged.