, Former Professor, Civil, Environmental and Geo-Engineering
The entire fleet of Metro Transit buses has been outfitted with Automatic Vehicle Location (AVL) systems to monitor vehicle location and track its schedule adherence, and about ten percent also have on board Automatic Passenger Count (APC) systems which collect ridership data for route planning, schedule frequency analysis and quality of service evaluation. Metro Transit collects significant amounts of data from these systems daily, and while examining AVL or APC data individually can provide understanding of individual issues, such examination does not offer system-wide perspective. The goal of this project was to develop a database model and a data mining and fusing algorithm that would combine these two separate data sets for systematic analysis. The resulting database model, by generating performance measures, allowed a transit manager or operator to identify problematic locations in the network and to implement appropriate strategies to maintain overall system performance, and could also be used to improve data quality through data mining and fusion from other data sources.