Toward validation of freeway loop detector speed measurements using transit probe data

Abstract

Inductive loop detectors are used throughout the world to report individual and average vehicle speeds along freeway and arterial segments. Most loop detector data are aggregated over an arbitrary period such as 30 seconds, 1 minute or 15 minutes. Loop detectors are usually installed at locations convenient for operating an advanced traffic management system. It is very important to find the optimal spatial resolution for loop detector placement and the optimal temporal resolution for detector data reporting. The objective of this study is to examine this question at one site using a combination of loop detector and automatic vehicle location (AVL) data from a bus fleet. The bus fleet in Portland, Oregon is 100% automatic vehicle location equipped with certain buses running on express routes along freeways with inductive loop detectors. This study uses the AVL data to report speed at several spatial and temporal resolutions. These speeds are compared to speeds reported by inductive loop detectors. The study revealed the potential to improve speed data reported by inductive loop detectors by using the median speed reported by the detectors to represent segment speed during each temporal window. Further research is needed to further develop optimal strategies for loop detector placement and data aggregation/reporting.