4/13/2017 | 4:15pm | E51-335
Reception to follow.

Dawn Woodard

Senior Data Science Manager of Dynamic Pricing
Uber

Abstract: Estimates of road speeds have become commonplace and central to route planning, but few systems in production provide information about the reliability of the prediction. Probabilistic forecasts of travel time capture reliability, and can be used for risk-averse routing in mapping services or as a component of fleet vehicle decision-support systems. We propose approaches for probabilistic prediction of travel time on large-scale road networks, and develop systems for use in both mapping services and ambulance fleet management. Our estimates are based on location data from vehicles traveling in the road network; for mapping services this is obtained from mobile phones, while for ambulance fleets it is obtained from automatic vehicle location devices. Our approaches are based on maximum a posteriori estimation in a class of mixture models. These approaches capture weekly cycles in congestion levels, give informed predictions for parts of the road network with little data, and scale efficiently in the size of the road network. We demonstrate greatly improved accuracy relative to a system used in Bing Maps, and show the impact of our methods for improving ambulance fleet management decisions.

Bio: Dawn Woodard leads data science for the Dynamic Pricing team at Uber. Dynamic Pricing creates the pricing systems for Uber, such as surge pricing and next-generation pricing technologies. The team includes specialists in operations research, economics, statistics, and machine learning. Dr. Woodard received her PhD in statistics from Duke University. She was then a faculty member in the School of ORIE at Cornell, where she received tenure. After a sabbatical at Microsoft Research she transitioned to a role at Uber, building and leading their Marketplace Optimization Data Science organization. This is now one of the premier data science teams at Uber and creates Uber’s marketplace-related technologies, such as dispatch, pricing, and incentives, across all of Uber’s products, such as UberX, UberPOOL, and UberEats.