PEDESTRIAN PROGRESSION: A VIABLE ALTERNATIVE IN CENTRAL BUSINESS DISTRICTS

Two studies were conducted to determine whether pedestrian arrivals at a traffic signal are affected by the upstream signal. Platooning study data collected at six intersections in St. Louis, Missouri indicated that pedestrian arrivals at these locations were not random but rather were likely affected by the upstream signal. A study on travel time along an arterial, conducted in Columbia, Missouri, showed that 23 of 30 intersections had significant difference between delay expected for random arrivals and delay experienced. Both of the studies indicate that pedestrian arrivals at signals can be affected by upstream signal timing. Therefore, providing for pedestrian progression would be a potential alternative to improve pedestrian flow. Various factors affecting pedestrian progression are illustrated based on field studies and are compared to factors known to affect vehicle and bicycle progression from literature review. TRB 2003 Annual Meeting CD-ROM Paper revised from original submittal. Chilukuri and Virkler 3 INTRODUCTION Pedestrian walkways are important transportation facilities in Central Business Districts (CBDs). Understanding the characteristics of the pedestrian movement along these walkways would be helpful in planning for better mobility of pedestrians. In the conventional method to estimate pedestrian delay, pedestrians are assumed to arrive randomly at signals (1). This may not be true when pedestrians are traveling in a coordinated signal network. Pedestrians arriving randomly at an intersection will move in a group after the signal turns green and might continue as a significant group toward the downstream signal. The objective of this research was to identify the potential for pedestrian platoon progression (movement as group from one signal to the next) in CBDs. Coordination of signal networks in a way that vehicles receive minimum delay is called vehicle progression. Various factors that are known to affect progression of motor vehicles and bicycles are illustrated and, for comparison, factors that appeared to affect pedestrian progression are described based upon observations from this study. Past research of pedestrian flow is described below. Two field studies, a travel time study and a platooning study, were conducted and their results are given. The factors which seemed to affect pedestrian flow are then discussed. Conclusions based on the findings of the studies are followed by recommendations and suggestions for further research. BACKGROUND INFORMATION Virkler (2, 3) studied pedestrian flow characteristics along signalized routes in Brisbane, Australia. No research on pedestrian signal coordination has been reported for American conditions. In an attempt to understand the potential for pedestrian progression, relevant Virkler studies and existing knowledge of vehicle progression and bicycle progression are described below, followed by a description of the assumption underlying the most used equation for pedestrian delay. Signal Coordination Benefits for Pedestrians Virkler conducted studies on ten Brisbane intersections to study the potential benefits of reducing delay through signal coordination. Each approach was one block from an upstream signal that was the source for the majority of the pedestrians on the approach. The signals had cycle lengths from 60 to 90 seconds. The block lengths ranged from 87 to 105 meters. The number of pedestrians reaching the downstream signal was collected in five-second interval for a period close to 15 minutes (depending on the cycle length). The studies indicate that significant amounts of platooning was present due to upstream signals and that platooning could increase or decrease delay at downstream signals, based on the offset from the upstream signal. The study proposed three techniques to determine the best offsets for pedestrians to experience less delay. TRB 2003 Annual Meeting CD-ROM Paper revised from original submittal. Chilukuri and Virkler 4 Prediction and Measurement of Travel Time along Pedestrian Routes Virkler also conducted travel time studies and compared observed travel times with the predicted travel time values based upon an assumption of random arrivals. The data was collected on 17 arterials close to the Brisbane CBD. The arterial lengths ranged from 90 to 1400 meters. The studies indicated that the standard deviation of delay experienced at a signal can be significantly different than expected and that signal coordination along an arterial affects the travel time of pedestrians considerably. Delay Equations The most frequently used equations for pedestrian average delay and the variance in the delay are derived based on the following assumptions (2, 6) 1. Saturation flow rate of pedestrians is infinity. 2. Pedestrians arrive at a uniform rate or randomly. 3. Pre-timed signal control along the arterial. The equations for average delay and for variance of delay: Average delay, D = 2C r (1)