ONGOING PROJECTS
The Intelligent Management of Destructive Wildfires:
Endeavors to employ advanced analytic methods to detect, predict the spread, and control ecological disturbances such as destructive wildfires. Destructive wildfires result in billions of dollars in damage each year and are expected to increase in frequency, duration, and severity due to increasing drought, excess biomass in the forests, and infestations of insects and disease. The overall research objective of the project is to develop intelligent management models that incorporate the spatial distribution and severity of wildfires over time and space, and to design a control plan.
Case Study: The dataset on California’s wildfires provides an excellent opportunity to test developed models. The California dataset includes geospatial features and allows the ArcGIS online service to display parameters for current, active wildfires from the National Incident Feature Service. There are currently 2,761 rows of data in the California dataset. Once a fire has been extinguished, the data for that fire is removed from the “active” category and archived for study and analysis. The dataset on California’s wildfires provides an excellent opportunity for the PI to test the models that she is developing.
Women and Minority Investors in the Cryptocurrency Market
... under construction
ACCOMPLISHED PROJECTS
Dynamic Reserve Network Design Problem
Developed innovative integer programming models to support decision-making on the reserve network design problem as it relates to biodiversity loss from ecological stressors such as rapid climate change, invasive species/plants, urbanization, and other human-caused environmental changes. The proposed models also select a subset of sites which are fully connected in order to maximize the utility, given a limited budget.
Case Study
Worked with the Georgia Department of Natural Resources (DNR) to develop a decision framework for reserve actions intended to conserve the statewide population of the gopher tortoise.
The tortoise, a candidate for listing under the Endangered Species Act, has declined over its southeastern U.S. range. Its habitat requirements include sandy soils that permit the tortoise to dig its characteristic burrows, a sparse forest overstory that allows light to penetrate to the forest floor, and periodic ground fires that encourage the growth of vegetative forage. These habitats are largely unavailable on agricultural and urbanized lands, which now account for most of the tortoise’s original range.
**Reserve network design is the problem of selecting parcels of land such that the assembled set maximizes some criterion pertaining to the conservation of species or natural communities with consideration of spatial constraints.
Invasive Species Management Problem
Invasive species pose a significant threat to global biodiversity. Managing invasive species often involves modeling the species’ spread pattern, estimating control costs and damage costs due to the invasion, designing control efforts, and accounting for uncertainties in model parameters. Proposed a robust spatial optimization model in order to select treatment sites in a way that maximally reduces an invasive population, given a constraint on financial resources.
Case Study
Studied an invasive tree from Australia (melalenca) that has spread in Florida’s Everglades National Park. Melaleuca is able to invade most natural community types found in south Florida and can rapidly take over areas due to their prolific seed production ability. It replaces native species (e.g., cypress, mangroves, sawgrass) and forms dense, homogenous stands that exclude other vegetation. Melaleuca stands decrease hydroperiods, intensify fire regimes, and decrease habitat quality for native fauna.
Demand Forecasting
Case Study
Conducted research in time series forecasting (multi-variate) using quantitative techniques, focusing on the demand forecasting. Among quantitative forecasting approaches, econometric, statistical, and artificial intelligence approaches are well known.
US domestic airline passenger demand forecasting: this studyinvestigated COVID-19’s impact on the US domestic air passengers demand, identify the most influential features on air passenger demand, and design more accurate forecast models.