NSF Grant Boosts Post-Earthquake Relief Study

Professor and researcher Ilya Ryzhov at the University of Maryland’s Robert H. Smith School of Business is leveraging a three-year grant awarded by the National Science Foundation to continue research on predictive and prescriptive methods for humanitarian logistics and disaster mitigation.

Working with Ge Ou and Nikola Marković from the University of Utah, Ryzhov and his co-researchers plan to utilize the $490,000 grant toward their post-earthquake relief research titled “Resource-Constrained Optimal Learning Framework for Post-Seismic Regional Building Damage Inference.”

Ryzhov, associate professor in the Department of Decision, Operations, and Information Technologies  and affiliate researcher of UMD’s Institute for Systems Research, is the principal investigator for the university’s portion of the award, which amounts to $161K.

Recognizing the urgency and necessity of accurately assessing the infrastructure damage after disasters, Ryzhov and his colleagues are developing a mathematical modeling framework to guide inspection teams through post-seismic reconnaissance missions.

According to the researchers, the grant focuses on improving processes toward resource allocation for post-seismic building damage assessment and will also aid inspection crew members in prioritizing the most impacted buildings and optimizing investigation time.

Using real-world benchmarks, which includes data from the 2011 Chile and 2015 Nepal earthquakes, as well as a regional earthquake simulation testbed for the San Francisco Bay, the researchers anticipate that the results can help expedite regional hazard damage assessment, improve disaster management, save human lives, ensure ethical resource allocation, and preserve societal welfare.

By integrating concepts from statistical and optimal learning with models for routing and scheduling — aspects which have been studied extensively separately, but never jointly — the researchers are hoping to close a knowledge gap that inspection teams often face in the field due to resource constraints.

Ryzhov and his colleagues will develop an integrated modeling framework that bridges optimal learning and combinatorial optimization to identify inspection routes and schedules that maximize the predictive power of machine learning models for post-seismic building damage assessment.

The researchers also expect to glean new insights into domains such as health and disease control that face a tight trade-off between data expense and information gain.

The project, they write, will prepare future civil engineers, mathematicians and statisticians with multi-disciplinary knowledge, and will broaden the participation of underrepresented groups in research which positively impacts engineering education.