Sequential decision making in infrastructure planning under uncertain climate change predictions

Abstract

Engineering decisions, as those related to infrastructure planning, face challenges due to the uncertain climate change predictions. Instead of a rigid planning scheme, an alternative is to utilize an adaptive framework, by sequentially collecting and processing information, and revise decisions accordingly. A recently developed probabilistic, physic-based state-space model (SSM) for prediction climate change is utilized in this work, and it is integrated with sequential decision making to facilitate and improve flexible infrastructure planning schemes. This integrated approach is applied to a case study related to the planning and evaluation of water supply infrastructure options. A sequence of information collection, prediction updating (of long-term climate impacts), and cost assessment of different decisions (e.g., investing on projects or taking no further action) is modeled and optimized. By integrating statistical climate modeling and sequential decision making, this work aims to offer insight on flexible planning schemes and to facilitate engineering adaptation to climate change.