Effect of Onboard Survey Sample Size on Estimation of Transit Bus Route Passenger Origin–Destination Flow Matrix Using Automatic Passenger Counter Data

Abstract

Origin–destination (O-D) flows at the bus passenger route level have traditionally been estimated from costly and labor-intensive onboard surveys. The availability of automatic passenger counter (APC) data on many bus transit systems offers the possibility of enhancing the quality of the onboard survey data at little marginal cost. This paper investigates the value of estimating route-level passenger O-D flows from APC data and onboard O-D survey data with a focus on the effect of onboard O-D survey sample size. An empirical study using field data collected on three bus routes investigates and quantifies the value of combining APC counts with onboard O-D survey data as a function of survey sample size. Encouraging estimation performance obtained in using APC data with no onboard O-D survey data, previously seen in a smaller study, is confirmed in this more extensive study. In addition, incorporating onboard O-D survey data with APC data produces O-D flow estimates that are better than those produced by using only the APC data, and increasing the sample size of the onboard O-D survey improves the quality of these estimates. However, the magnitude of the improvement depends on the O-D flow structure of the given bus route; increasing sample size results in less appreciable improvement for routes with more concentrated O-D flows than for routes with more evenly distributed O-D flows.