Dawn B. Woodard is promoted to Associate Professor with Tenure
Dawn B. Woodard, who joined ORIE in 2008, has been promoted to the position of tenured Associate Professor.
Professor Woodard holds M.S. and Ph.D. degrees from the Duke University Department of Statistical Science and a B.S. in Mathematical and Computational Science from Stanford University. She conducts research related to Bayesian statistics and its applications.
Bayesian statistics, dubbed “the theory that would not die” in a recent book by Sharon Bertsch McGrayne, is a class of methodologies for refining estimates of probability distributions or odds by incorporating new data as they become available. [In contrast to so-called frequentist statistics that predominated in the last century, the Bayesian approach does not require full knowledge of the probable causes behind a set of observations, but instead reasons about causes from the observed outcomes].
Woodard has enhanced and applied Bayesian statistics in areas as diverse as taking the right corrective action when a crisis occurs in a data center running thousands of computers, forecasting the rate at which calls come in to an emergency medical service, estimating ambulance travel times, determining the concentrations of nitrates in groundwater, and establishing the relationship between sleep characteristics and health outcomes. Woodard has also investigated ways to improve the accuracy of computational techniques used in Bayesian statistics.
At Cornell, Woodard has been recognized as an excellent teacher of undergraduates, winning a Ralph S. Watts ’72 Excellence in Teaching Award in 2012. In addition to Ph.D. level classes, she has taught the basic engineering probability and statistics course required of OR majors and taken by students in other fields, and statistical data mining, which is a course in “Big Data” and data analytics. A student in the basic course wrote “before taking this class I didn’t like statistics at all. I thought it was just a boring subject but she made this course interesting. Now I find statistics really useful and important. I think every engineer should take statistics.” A data mining student wrote “Professor Woodard is absolutely incredible. Extremely clear lectures and very willing to help when necessary… one of the best courses I’ve taken at Cornell.”
Woodard has advised several Master of Engineering projects, including one to predict how long a patient will spend in operating rooms at New York Presbyterian/Weill Cornell Hospital, another to predict from the attributes of a product sold at Walmart.com whether it can be handled on a standard conveyor built in the distribution center, one to analyze priority accounts at consulting firm Ernst & Young, and one to assist in the development of healthcare securities trading strategy for the Royal Bank of Canada. She has also supervised undergraduate research into the optimal location of franchises and the assignment of IT help desk supervisors.
Woodard’s first Ph.D. student, Bradford S. Westgate, is now a visiting assistant professor in the Department of Mathematics and Statistics at Mt. Holyoke College in South Hadley, Mass. His thesis, “Travel Time Distribution Estimation for Ambulances using GPS Data,” led to a joint publication with Woodard, Assistant Professor David Matteson, and Professor Shane Henderson in the Annals of Applied Statistics
Woodard’s research is currently supported by two grants from National Science Foundation’s Statistics Program. She is an associate editor of the Journal of the American Statistical Association: Theory & Methods and of Stochastics: An International Journal of Probability and Stochastic Processes and referee for many other leading journals. She develops and maintains publicly available software packages that apply the methods she has published for clustering of time series such as arise in identifying crises in distributed computing; for analyzing spatial data to detect such properties as nitrates in groundwater; and for making predictions from the extensive data collected in such varied applications as EEGs from sleep studies, MRIs from brain studies, and spectroscopic data in chemistry