Session VII (May 16, 1:30pm-3:00pm): Experimental Design in Transportation Studies, organized by Feng Guo
Title: Leveraging Statistical Experiment Designs in the Ride-Share Industry: Applications in Road Safety Products
Speaker: Danni Lu, Uber
Abstract: The ride-share industry has experienced steady growth in recent years, with companies continuously seeking ways to optimize their operations and enhance user experience. Statistical experiment designs offer a powerful toolkit for conducting controlled experiments to test hypotheses and make data-driven decisions. In this talk, we explore the practical applications of experiment designs in road safety considerations at Uber. We discuss the general experimentation process to test various aspects of proposed road safety products, including safety improvement, user satisfaction, and marketplace efficiency. With examples of successfully launched products, we showcase how experimental methodologies such as A/B testing and switchback designs have been utilized to identify actionable insights and drive meaningful improvements at Uber. Additionally, we discuss the challenges and considerations specific to implementing experiments in the dynamic and complex environment of the ride-share industry.