Abstract
Most state-of-the-art control strategies for coping with arterial congestion provide progression for heavy through-traffic flows. However, such strategies cannot tackle arterial congestion caused by both heavy turning and through-traffic flows, where turning-traffic volumes often spill over their designated bay length and cause link blockage. An effective approach is to offer a progression band to each of those critical path flows that can be identified from the arterial origin–destination (O-D) flow patterns. This study proposes three models for estimating such information from available traffic measurements. The estimated time-varying O-D distributions yield both the number of critical path flows and their respective volume ranks for design of their progression bands. Based on the principle of flow conservations, the first model captures the relationships between link counts and dynamic O-D flows, whereas the second model directly takes turning flows at each intersection as the primary model input. To consider further the impact of traffic signal plans on O-D flow patterns, the third model incorporates a set of additional measurements—the time-varying queue length information—to improve the estimation accuracy. Comparisons of the actual O-D flows and the estimated results demonstrate the effectiveness of the proposed models for identifying the heavy flow paths and their respective volumes.
Get full access to this article
View all access options for this article.