conference paper

Extracting traffic patterns from loop detector data using multiple change point detection

Proceedings of the 93rd annual meeting of the transportation research board

Publication Date

January 1, 2014

Abstract

In this paper, we first introduce the Pruned Exact Linear Time (PELT)â??a segmentation approach for detecting multiple changepointsâ??to automatically identify the onset of congested periods of freeway operation using original, disaggregated, 30-second loop detector occupancy data. The purpose of the algorithm is to detect and map phase transitions in the occupancy data, keeping the general features of the traffic pattern while substantially reducing time in computation, retrieving, and presenting data with computation complexity that is only O(n). By using PELT, the start and end of the congestion period is identified automatically. The algorithm is tested on data from over 1000 mainline detectors in Orange County, California, USA both for a single day and for a month. The compression ratio of occupancy data is about 38.5, allowing an opportunity to analyze and monitor traffic in a more efficient way. This research provides an approach to quantify and display both the beginning of the congestion as well as total congestion duration on temporal-spatial maps that could lead to an inexpensive means to improve the quality of ramp metering settings and real time traffic monitoring.

Suggested Citation
Ming-Hsun Yang, Thuy T.B. Luong and Will Recker (2014) “Extracting traffic patterns from loop detector data using multiple change point detection”, in Proceedings of the 93rd annual meeting of the transportation research board, p. 14p.