For my Ph.D. dissertation I evaluated the relationship between access to essential services (e.g. supermarkets, libraries, pharmacies) and household recovery in Southwest Florida following Hurricane Irma in 2017. I demonstrate that measures of access are more valuable as hazard-specific, place-based measures of resilience and vulnerability than other typical resilience and vulnerability measures. Journal article in progress.

I developed a Novel Bayesian belief network-based anomaly detection method applied to LBS data to estimate household and workplace recovery periods in Florida following Hurricane Irma. This work won the Society of Risk Analysis Engineering and Infrastructure Specialty Group Student Merit Award in 2022 and the American Association of Geographers Jeanne X. Kasperson Student Paper Competition for the Hazards, Risks, and Disasters Specialty Group in 2023. This paper is currently under review.

I developed an unsupervised learning approach applied to LBS data to estimate closure periods of essential services facilities in Florida following Hurricane Irma in 2017. This work was published in Risk Analysis in July 2023.

In consultation with University of Michigan's College of Engineering, we developed a risk-based simulation model to estimate transmission risk associated with returning to in-class learning in Fall 2021. This work was published in PLOS One and is available open source here.

A lack of consistency in the reporting and availability of disaggregated, detailed data on COVID-19 in the US has limited the application of artificial intelligence methods and the effectiveness of those methods for supplementing mechanistic approaches for projecting the spread and subsequent impacts of this disease in communities. These limitations are missed opportunities for AI methods to make a positive contribution, and they introduce risk of inappropriate use of AI methods when not proactively acknowledged. Published in the Journal of Risk Research.

For my Urban Informatics course, I used openly available data and tools to visualize healthcare access in Detroit by travel mode and provider type. These results showed that when only considering walking and transit, far more areas of Detroit stand out as healthcare shortage areas based on DHHS defined ratios of providers to population. Urban health care access varies by travel mode and service need, so any interventions for providing better access should include consideration of those variables. Find the GitHub repository and full poster here.