Special Session-5

Special Session Organizers:

Prof. Dandan Peng, Northwestern Polytechnical University, China

Assoc. Prof. Zhenzhen Jin, Guangxi University, China

Postdoc. Dandan Zhao, Western University, Canada

Postdoc. Huan Wang, City University of Hong Kong, China

Postdoc. Xiaoxi Hu, Tsinghua University, China

Introduction and Topics:

Transportation infrastructure and equipment, such as railways, highways, bridges, and tunnels, constitute the backbone of modern intelligent transportation systems. Ensuring their safety, reliability, and availability is critical for sustainable mobility. However, transportation assets are increasingly exposed to complex operating conditions, aging effects, environmental disturbances, and imperfect sensing scenarios, which pose significant challenges to traditional operation and maintenance (O&M) paradigms.

Recent advances in sensing technologies, data-driven modeling, and intelligent algorithms have enabled a paradigm shift from reactive maintenance to predictive and prescriptive maintenance. Nevertheless, several open challenges remain, including limited fault data, heterogeneous sensing configurations, cross-domain generalization, sensor reliability, and the integration of physical knowledge with data-driven approaches.

This special session aims to bring together researchers and practitioners to discuss recent advances in intelligent O&M of transportation assets, with a focus on sensing, fault diagnosis, prognostics, and maintenance decision-making. The session emphasizes both methodological innovations and real-world deployment challenges in transportation systems.

Topics of interest include, but are not limited to:

• Advanced sensing technologies and sensor deployment optimization for transportation infrastructure
• Fault detection, diagnosis, and prognostics for transportation equipment
• Intelligent monitoring under imperfect sensing and degraded perception conditions
• Data-driven, knowledge-driven, and hybrid (physics-informed) modeling approaches
• Small-sample, few-shot, and transfer learning for fault diagnosis
• Multisensor data fusion and cross-modal analysis
• Incremental learning and adaptive models for evolving monitoring systems
• Digital twin and virtual sensing for transportation asset management
• Explainable and trustworthy AI for safety-critical O&M applications
• Maintenance decision-making, optimization, and lifecycle management