A new study led by Miguel Figliozzi of Portland State University provides a microscopic evaluation of how two advanced traffic control technologies work together.
Powell Boulevard, an east-west arterial corridor in southeast Portland, Oregon, has been the focus of several research studies by Figliozzi’s TTP research lab. The street is a key route for public transit buses as well as pedestrians and cars, but heavy traffic at peak hours often results in delays.
On Powell there are two systems operating concurrently: a demand-responsive traffic signal system called Sydney Coordinated Adaptive Traffic System (SCATS) and a Transit Signal Priority (TSP) system. The TSP in the Portland metro region is designed to give priority to late buses and to boost transit performance.
In previous studies Figliozzi’s lab has analyzed a multitude of factors on Powell Boulevard including traffic congestion, transit times, air quality and cyclists’ intake of air pollutants, and a before/after evaluation of SCATS.
For this study, the researchers used a novel approach to evaluate how well SCATS and TSP work together by integrating three major data sources and video recordings at individual intersections.
Figliozzi’s team worked closely with TriMet and the City of Portland to...Read more
The video begins at 0:33.
There are no archived materials available for this presentation.
The video begins at 1:15.
Traffic conditions were examined along a 30-km section of northbound Autobahn 5 near Frankfurt, Germany, using archived inductive loop detector data recorded at one-minute intervals. By focusing on the spatio-temporal evolution of traffic between freely flowing and queued conditions, it was possible to identify bottleneck activations and characterize reproducible features related to their formation, discharge and dissipation. This was accomplished by systematically probing the excess vehicle accumulation (spatial) and excess travel time (temporal) that arose between measurement locations. It is shown that bottlenecks became active in the vicinity of on-ramps and off-ramps. Further, the evolution of a several shocks of low flow, low velocity, and relatively short duration were traced over an approximately 16 km distance. It is shown that once a bottleneck became active, its measured outflow was reproducible across multiple activations and across multiple days. The analysis tools used in this study were transformed curves of cumulative vehicle count and cumulative time-mean velocity, using loop detector data in their most raw form. These cumulative curves provided the resolution necessary to reveal the spatial and temporal aspects of dynamic freeway traffic flow phenomena. With increasing availability of reliable freeway sensor data, it is important to continue the systematic...Read more
The video begins at 4:29.
Data from two portions of congested freeway have unveiled the mechanisms that cause traffic oscillations (i.e., stop-and-go driving conditions) to form and grow in amplitude over space. Oscillations formed due to driver lane-changing maneuvers. Vehicles that inserted themselves into relatively small spacings in adjacent (target) lanes triggered these formations by inducing temporary decelerations among vehicles immediately upstream. Once they formed, this same lane-changing mechanism triggered oscillation growth. We found no evidence that oscillations formed or grew due solely to driver interactions in single streams of traffic, independent of adjacent streams. Still, car-following behavior might also trigger these formations and growths. (Oscillations have been observed on single-lane roads and this implies that car following plays a role.) Yet on (multi-lane) freeways, whatever the influence of this latter behavior might be on oscillations, it is secondary to that of lane changing. The finding is notable in light of the many attempts to explain oscillations as strictly a car-following phenomenon. Of further note, oscillations often diminished in amplitude when they propagated past merge areas that had become fully engulfed in queues. The oscillations’ growth caused by lane-changing was countered by the inflows from queued on-ramps. This effect is explained with a theory that...Read more
The video begins at 8:36.
Dr. Smaglik is currently working on three separate transportation research projects at Northern Arizona University. This talk will touch briefly on each of the three projects, the concepts behind them, workplans, and expected deliverables. The projects include work with the Oregon DOT on the impact of less than optimal vehicle detection on adaptive control algorithms, development of a ped priority algorithm through a NITC project (as a Portland State subcontractor), and internally funded work on a power harvesting traffic sensor.
Dr. Edward J. Smaglik, P.E. is an Associate Professor at Northern Arizona University (NAU), Flagstaff, AZ, in the Department of Civil Engineering, Construction Management, and Environmental Engineering. Dr. Smaglik has over 7 years of academic research and teaching experience, preceded by 2 years of experience as a post-doctoral research associate. In addition to typical academic teaching responsibilities, he has served as Principal Investigator on transportation related projects on a wide range of topics, including the development and implementation of a pedestrian priority algorithm, the implication of vehicle detection degradation on higher level traffic control algorithms, the analysis of travel time data related to special events, the development of a...Read more
Modeling transportation basically involves development of relationship between the demand for transportation and the land-use, socio-economic and transportation system characteristics. The Indian socio-economic and transportation system characteristics are highly complex and wide ranging and hence, formulation and quantification of appropriate causal variables for modeling is a challenging task. The first part of the talk will focus on this aspect. The traffic on Indian roads is highly heterogeneous and the vehicles move on the roads without any lane or queue discipline. Hence, the commonly adopted procedure to model lane based traffic flow is not applicable for modeling this type of traffic comprising vehicles of wide ranging static and dynamic characteristics. The approach to modeling of this type of traffic flow is distinctly different. An appropriate methodology for modeling heterogeneous traffic flow has recently been developed at Indian Institute of Technology Madras and the same be will discussed in the second part of the presentation.
Dr. V. Thamizh Arasan, Professor and Head, Transportation Engineering Division, Department of Civil Engineering, Indian Institute of Technology Madras, Chennai, India, has been involved in teaching, research and consultancy, in the area of Transportation Engineering for the past two and a half decades. Traffic Simulation and Travel Demand Modeling are the areas of his research interest, and he has guided several Ph. D....Read more
In the Engineering Building, Room 315
Abstract: Signal priority applications in the U.S. tend to be timid about giving priority to buses, because if they interrupt the green period of a competing traffic stream, they have no means of compensating that stream in the next signal cycle (by giving it a longer green period). Common restrictions set up to protect cross streets include preventing a priority interruption in consecutive signal cycles, having short extension intervals, and inhibiting priority when traffic is heavy on the cross street. In addition, most priority applications are limited to one or two simple control tactics, green extension and early green. As a result of these limitations, transit signal priority often falls far short of its potential, saving buses 3 seconds or less per intersection. We show how by using multiple intelligent signal priority tactics, in which traffic is aggressively interrupted but also compensated in the following cycle, large benefits can accrue to transit operations without any undue effect on general traffic. In a simulation study of four traffic signals around a large bus terminal in Boston, we found that average delay per bus could be reduced by almost 20 seconds per intersection with no change in average motorist delay.