|

Evaluation of an Automated Speed Enforcement Program in Toronto, Canada

Author(s): Zubair, Schwartz, Rothman, Cloutier, Macpherson, Howard

Slidedeck Presentation:

CARSP 2022 Conference Sudburry_SaroarZ_20220617

Abstract:

Background:

Speed is a major contributing factor in road traffic collisions, increasing the risk of collisions and risk of severe injury and fatalities. In observational studies, Automated Speed Enforcement (ASE) has been found to be an effective method for reducing collisions, injuries, and deaths. Yet, no studies have included controlled, randomized experiments of safety impacts of ASE. In December 2019, the City of Toronto began an ASE program as part of Toronto’s Vision Zero strategy to achieve safety around schools. Fifty mobile cameras are currently being rotated through Toronto community safety zones (designated roadways around schools) in six phases of 4-5 months (300 locations total). Three phases are currently completed.

Aims:

This study aims to examine the impact of ASE cameras on motor vehicle speeds in safety zones around schools in Toronto, Canada.

Methods:

Pre-installation vehicle speed and volume data were obtained from the City of Toronto with measurements from pneumatic tubes in Spring 2018 and 2019. Speed and volume data were collected by ASE devices post-installation. Percent of cars over the speed limit and 85th percentile speeds were compared pre-installation and during the ASE intervention for the first 3 phases. Generalized estimating equation models will be used to estimate the longitudinal impacts of ASE cameras, controlling for seasonality, speed limit changes, and school closures due to the COVID-19 pandemic. Forty-five control sites, consisting of future ASE sites, will also be included. ASE sites will be stratified based on location on 30km/h or 40km/h speed limits roads.

Results:

For the first phase of the intervention (50 sites), the percent driving over the speed limit dropped from 49% to 28% on 40km/h roads and 55% to 44% on 30km/h roads after ASE implementation. The 85th percentile speed similarly decreased from 58 km/h to 46 km/h on 40 km/h roads and from 42 km/h to 39 km/h on 30 km/h roads. Multivariate modeling is ongoing to estimate impacts (including relevant controls) on excess speeding and PMVC collisions.

Discussion:

Early results indicate that ASE can effectively reduce dangerous speeding on the road. Yet, speeding in these community safety zones was still common at levels that could pose a serious danger to children. Promising findings suggest that ASE may be an effective program to improve safety in community safety zones. This program could be used in tandem with other Vision Zero strategies to better eliminate excessive speeding. The first phases of data collection for this program were complicated by widespread school closures and decreased volumes due to the COVID-19 pandemic.

Conclusions:

We will continue updating this analysis examining the impacts of ASE on speed and will model impacts on pedestrian motor vehicle collisions, as data becomes available at all 300 sites.