Participatory Statistical Inference of Interdependent Critical Infrastructure Recovery Times

点击翻译按钮获取中文摘要

This research will develop new methods to estimate disaster recovery times of interdependent critical infrastructures, such as electricity and water systems, after damaging hazard events. Current recovery estimation is typically more ad hoc than systematic and statistically rigorous. This project will address this important challenge systematically and with rigor. It will lead to open-source tools for developing disaster recovery time estimates and planning for hazard-impacted critical infrastructure systems, with a focus on ensuring ease of use. Project case studies of the framework will underpin development of the tools, expand our understanding of infrastructure recovery after disasters, and help establish best practices that can be emulated by other communities. To build capacity of young researchers and future practitioners, the project team will integrate undergraduate and graduate students throughout the project. To engage traditionally underrepresented students in STEM education, the project team will leverage two existing programs targeting high school and incoming college students. This scientific research thus supports NSF's mission to promote the progress of science and to advance our national welfare and prosperity with benefits that will facilitate future planning initiatives to improve the resilience of United States communities and their critical infrastructure systems.The project develops a new methodological framework, as well as software tools to support this framework, for estimating post-event interdependent critical infrastructure recovery times. The core of the framework is a participatory process for eliciting recovery estimates from topical experts. The framework will include tablet-based and web-based software tools to facilitate the elicitation. A human-centered approach will be used to develop the software to maximize user experience and elicitation performance. The methodological framework will use Bayesian inference to integrate available empirical data with expert estimates. Experts will estimate recovery functions or trends, rather than points or probability distributions. This will enable calculation of common resilience metrics, such as the area under a recovery curve. The framework and tools will be evaluated based on case studies in Seattle, WA and Portland, OR.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.