|

Automated Evaluation of the Effectiveness of Curb Radii Adjustments on Turning Vehicle Speed: A Case-Study in Toronto

Author(s): Sohail Zangenehpour, Charles Chung, Sheyda Saneinejad

Slidedeck Presentation:

4B_2_Zangenehpour

Abstract:

Wide curb radii at intersection corners increase drive sight angle, reduce pedestrian visibility and can contribute to high speed turning movements. These can lead to serious accidents with pedestrians. Therefore, one way to increase the safety of pedestrians at intersections is to reduce the speed of turning vehicles. To address this, as well as to reduce the crossing distance for pedestrians, City of Toronto is adjusting curb radii at intersections on an ongoing basis. However, thorough evaluation of this type of intersection treatment is needed in order to quantify the safety benefits. The objective of this work is to apply an automated video based method for evaluating the effectiveness of curb radii adjustments on reducing turning vehicle speed as well as reducing conflicts between turning vehicles and pedestrians at two intersections in Toronto . Two locations were selected following several complaints from the public on vehicles speeding through the turns, not yielding and not stopping (at the stop sign or red light). In addition to complaints from the public, in the past 5 years, one of the intersections has had two cases of collisions resulting in fatalities or serious injury. Two intersections in City of Toronto, Davenport / Christie and Yorkwoods / Driftwood are the subjects of this study. The speed of turning vehicles and frequency and severity of conflicts between turning vehicles and pedestrians are measured through an automated software that is capable of tracking and identifying different road users in traffic video and extracting their speeds. Using speed and conflicts as surrogate measures for safety at intersections is an efficient way to determine the safety effect of curb radii treatments which does not require many years of waiting to observe accidents and possibly injuries or deaths. To date, the before-study videos from these two intersections have been recorded and processed. The curb radii at these two intersections will be adjusted in Fall 2016. After the adjustment, videos will be recorded for the after-study and then processed to evaluate the effectiveness of the treatment. The intersection of Davenport and Christie has a steep slope at the north leg; we expect to see a noticeable difference in turning vehicle speeds after the adjustment. On the other hand, the intersection of Yorkwoods and Driftwood will have a drastic change in the physical curb radius; we expect the impact of the adjustment on turning vehicle speed at this location to also be significant. To improve the safety of pedestrians at intersections, cities are constantly building new or adjusting their current ¬infrastructure. However due to the difficulty of data collection, very few studies have thoroughly investigated the effectiveness of these countermeasures. The speed of turning vehicles and frequency and severity of conflicts between turning vehicles and pedestrians can be good surrogate measures for safety at intersections. In this paper, using a before-after study and an automated technique for extracting speeds and conflicts, we targeted two intersections in City of Toronto to evaluate the impact of curb radii adjustments on reducing the speed of turning vehicles and their conflict with pedestrians.