Comparison of Algorithms for Systematic Tracking of Patterns of Traffic Congestion on Freeways in Portland, Oregon

Abstract

Systems for automatically tracking traffic congestion that are integrated into freeway management systems as applications of an intelligent transportation system (ITS) have received a high level of attention in many cities in recent years. Some cities automatically provide traveler information on travel times, congestion activation times, and locations. However, different algorithms are used by professionals and decision makers. Some popular algorithms for automatically tracking congestion include parameter threshold–based algorithms such as those that use speed thresholds and the Automatische Staudynamikanalyse [automatic tracking of moving traffic jams–forecasting of traffic objects (ASDA-FOTO)] models. The rescaled cumulative curve tool has also been used for analyzing traffic features. The advantage of this tool is that it reveals traffic features in more detail, but one disadvantage is that it is a visual tool. When used to process data systematically in an operations environment, this tool can make analysis laborious. This paper proposes two algorithms for automatically tracking congestion by making use of the rescaled cumulative curve tool and compares them with the traditional speed threshold algorithm and the ASDA-FOTO algorithm. The results are represented graphically and numerically.