Ambulence Travel Time Estimation Using GPS Data

• Activity Number: 250
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Bradford Westgate*+ and Dawn Woodard and David Scott Matteson and Shane G Henderson
• Companies: Cornell University and Cornell University and Cornell University and Cornell University
• Keywords: Reversible jump ; MCMC ; Gibbs sampler ; GPS ; ambulance ; EMS

Abstract

Estimates of ambulance travel times on road networks are critical for effective ambulance base placement and real-time ambulance dispatching. In previous work, we introduced a Bayesian method for estimating the distribution of travel times on each road segment in a city, using Global Positioning System (GPS) data recorded during ambulance trips. Due to sparseness and error in the GPS data, the exact ambulance paths and travel times on each road segment are unknown. The Bayesian method estimates the ambulance paths and road segment travel time distributions simultaneously. In this work, we propose approximations to this model that are more computationally efficient, and therefore able to handle large GPS datasets and road networks using reasonable computing resources. We test the predictive accuracy of this method on the city of Toronto, using data from 2007-2008. We compare the performance of this method with simpler methods using the GPS data and a method proposed by Budge et al. (2010) that uses only total travel distance and duration of the ambulance trips.