Download Presentation (12.9 MB)

Streaming Media




Conventional risk assessment approaches in infrastructure management do not fully capture the system-level impact of structural failure or service disruption. As a result, the priorities of preservation projects may be misidentified, leading to suboptimal maintenance schedules and waste of resources. In this presentation, we will first illustrate why conventional risk assessment is not suitable for transportation structures and networks due to interdependency between assets, and then demonstrate how system-level preservation policies can be devised using novel algorithms adapted from the field of deep reinforcement learning. Results from a series of case studies showcase that the system-level risk management is essential to correctly identifying key assets and work priorities, devising adaptive maintenance policies, and lowering life-cycle costs of transportation infrastructure.

Biographical Information

Dr. David Y. Yang is an Assistant Professor of Structural Engineering in the Department of Civil and Environmental Engineering at Portland State University. His research focuses on structural reliability and risk, and risk-informed decision-making for infrastructure systems. His recent work specializes in unifying risk and resilience assessment in multi-hazard scenarios, devising novel machine learning techniques for infrastructure management, and risk-informed infrastructure adaptation to climate change. He is the recipient of the 2022 Moisseiff Award from the American Society of Civil Engineers (ASCE). He is the author of 29 journal papers, 3 book chapters, and over 10 contributions to various national and international conferences.


Transportation | Urban Studies and Planning

Persistent Identifier

System-level Risk Management of Transportation Structures and Networks