Estimating the Efficiency of Urban Transit Agencies that Provide Multiple Types of Service

The authors develop a Data Envelopment Analysis (DEA) procedure for identifying the efficiencies of the different types of service provided by a transit agency, and the effect of each service type on the efficiency of the agency as a whole. The authors show how to estimate the efficiency with which an agency allocates resources among its service types, and provide a method for identifying reallocations that will increase agency efficiency. The authors illustrate the procedure by applying it to three large United States urban transit agencies using data from the National Transit Database.