Being the largest city in Canada, road safety is the main concern of everyone living in Toronto. Based on Toronto Police Service KSI data, there were 610 fatalities including 351 pedestrians, 109 drivers, 48 motorcycle drivers and 30 cyclists for the period 2008–2018. Since 2017, Toronto city has implemented an ambitious comprehensive five-year (2017–2021) Vision Zero Road Safety Plan and adopted various road safety measures and campaigns aiming to eliminate traffic-related fatalities and serious injuries. The Plan identifies and addresses six emphasis areas, which were determined through collision data analysis, public engagement and Council direction. They include Pedestrians, School Children, Older Adults, Cyclists, Motorcyclists and Aggressive Driving and Distraction.
However, the death toll continues to mount despite well-intentioned measures to make pedestrian crossings safer. Advocates, city councils, politicians and members of the public are demanding Toronto make its streets safer for cyclists and pedestrians after the recent increasing trend of deaths in the last 5 years. Based on the 2016 Ontario road safety annual report, 44.8% of pedestrian fatalities occurred in Toronto. Members of the communities question the efficiency of the Vision Zero Road Safety Plan.
In this study, 11 years of traffic collisions data (KSI: Killed and Seriously Injured) from the Toronto Police Service was analyzed with a view to identify and quantify various factors in space and time on a road network and help location-based decision making to improve Toronto road safety.
Can GIS Analysis Make Toronto Roads Safer?
The Vision Zero Road Safety Plan is anchored in evidence-based analysis of traffic collision data that helps to identify the real issues affecting road safety as seen through long-term trends and location mapping. GIS analysis can help us to target where the issues exist and prioritize the deployment of safety measures. GIS analysis can identify traffic accident hot spots in space and time on a road network and help location-based decision making.
* Which intersections and roadways in Toronto have the highest crash rates?
* When and where do most crashes occur?
* How does the spatial pattern of fatalities differ from the spatial pattern of traffic accidents overall?
* How does the spatial pattern of crash rates occurring during the workweek afternoon commute differ from the overall pattern of crash rates?
* Overtime, which intersections or roadways are persistent problem areas for traffic accidents?
* Where are the hot spot areas for crashes involving elderly drivers, teenage drivers, or alcohol-related accidents?
* When and where do accidents involving elderly drivers, teenage drivers, or alcohol cluster spatially?
The red street segments on the map (please check the web app) show higher traffic accident rates on Toronto streets.
Predicting Street Safety by using Convolutional Neural Networks
It is important to understand the impact of architectural and urban-planning constructs on people’s perception of streetscapes. A machine learning method is trained using image features from more than 200000 Google Street view images to measure the perceived safety of Toronto streets.
When are the most dangerous times to be driving?
The number of car accidents increases with the number of drivers on the road. There are more crashes and fatalities on Friday in Toronto compared to other weekdays.
The graph shows that the number of crashes higher between 1 pm to 7 pm. It reaches the highest peak at 6 pm.
Where do 3:00 to 6:00 PM crashes occur in Toronto?
Our 24 hours data analysis over the last 11 years on the left side map shows that several peaks emerge, but the strongest is associated with the time period between 3:00 pm and 6:00 pm (between hours 15 and 18). The red dots on the map show the locations with statistically significant clustering of high rates of collisions during peak hours (please check the web app).
Traffic Collision Fatalities in Toronto 2008–2018
Transportation Services continuously makes improvements that have proven to be effective in addressing road safety. However, despite the fact that overall traffic collisions in Toronto have been stable with the increasing trend for over a decade, we have seen a recent increase in traffic-related fatalities — most notably pedestrians, cyclists and older adults. The cost associated with traffic accidents is staggering. In addition to the devastation of lost lives and serious injures, collisions involving pedestrians and cyclists in Toronto cost over $60 million each year (2012). The majority of these accidents are entirely preventable. There were average 55 people killed annually for the same period. Pedestrians are the main victims (57% of total fatalities) of road collisions in Toronto.
Road accidents are typically analyzed to address the influences of humans, vehicles, and infrastructure. Many factors contribute to road safety — including roadway design, speeds, human behaviours, technology, and policies. This increasing trend of fatalities over the last decade has emphasized the need for a comprehensive and coordinated road safety strategy that will further protect vulnerable road users and reduce the number of collisions resulting in death and serious injury. The map on the left side shows traffic accident density distribution on Toronto streets. The dark colour indicates higher RTC rates (please check the web app).
Space-time Patterns of Collisions
GIS Analysis can examine space-time trends in traffic collisions. The KSI data is analyzed by using its date field to structure the crash data into space-time hexagon bins (time-step interval one year, distance Interval 1km). The red bins are statistically significant space-time clusters where a large number of crashes occurred. The white bins are statistically significant space-time clusters with very few crashes (please check the web app).
KSI Data Analysis
Based on Toronto Police Service 2008–2018 KSI data, there were 610 fatalities including 351 pedestrians, 109 drivers, 48 motorcycle drivers and 30 cyclists.. Each year, more than 55 people are needlessly killed and thousands more are injured or disabled on Toronto streets. Based on the last 11 years of KSI data, 51.31% RTC (Road Traffic Crash) were intersection related while 48 % RTC happened when there was no traffic control and 7.7 % RTC occurred at stop sign related intersections. 13.29 % RTC (80 fatal) happened under poor visibility or road conditions such as snow (12 fatal), rain (64 fatal), freezing rain (1 fatal), ice, fog and strong wind. 119 fatal crashes happened in loose or packed snow (6 fatal), ice (1 fatal), slush (4 fatal), wet (fatal 108) weather conditions. The map on the left side (please check the web app) shows the fatality density and fatal collisions on Toronto streets.
Traffic Signal Vehicles and Pedestrian Volumes
Vehicle and Pedestrian Volumes at some intersections are higher than at some other intersections in Toronto. The most heavily travelled intersections — the total of vehicle and pedestrian totals — happen to be downtown in Toronto. It can be seen from the map (based on limited data) on the right, both pedestrian and vehicle volume is higher at some major intersections as below:
At some busy spots such as at the intersections of Yonge St with Gerrard St W, Dundas St W, Queen St W, Richmond St W, Adelaida St W, King St W, Wellington St W and Front St W, pedestrian traffic volumes higher than vehicle traffic signal volume based on the 8hr traffic signal vehicle and pedestrian volumes data.
Pedestrians: Major Victims of Fatal Traffic Collisions
Pedestrians (a serious and fatal collision where a pedestrian involved) is the major factor of fatal traffic collisions. In terms of initial impact type, a pedestrian collision takes up 39.85% of the total number of collisions. There were total 2104 pedestrian killed and seriously injured between 2008 and 2018 including 351 fatalities, accounted for 56% of total fatalities for this period.
In terms of pedestrian action data analysis, there were total 1568 KSI crashes related to crossings. Among them, 271 is fatal and accounts for 45.69% of total fatalities. Improper pedestrian behaviours (55%) also play a major role in causing traffic accidents. For example, There were 117 fatal cases (35%) of crossing with no traffic control, 39 fatal cases (11.78%) of crossing without right of way and 14 fatal cases (4.2%) of running onto roadway.
Pedestrian Fatal Collisions
KSI collisions involving vulnerable road users, such as pedestrians, cyclists and motorcyclists are disproportionately higher than all other modes of transportation and accounted for 74% of all KSI collisions in Toronto over the last 11 years.
117 pedestrian fatal crashes happened when there was no traffic control, 81 fatal crashes occurred when pedestrians crossed with right of way and 39 fatal crashes happened when pedestrians crossed roads without right of the way.
Aggressive and Distractive Driving
Aggressive and distractive driving is one of the major causes of fatal traffic collisions in Toronto. Aggressive driving includes speeding, running red lights, tailgating, weaving in and out of traffic, and failing to yield right of way, among other behaviours. There were 290 fatalities in the last decade, which took up 47.5% of total fatalities between 2008 and 2018. Most of them are centred in Toronto Centre-Rosedale, Toronto East York and Toronto Danforth areas. The red road segments of the road network on the right-side map are locations with statistically significant clustering of high collision rates (please check the web app).
Understanding the spatial and temporal patterns of traffic accidents will help to recommend very specific remediation strategies to help prevent future crashes. Knowing which roadways and intersections are associated with persistently high rates of crashes involving alcohol, for example, may direct the locations of DUI checkpoints. For example, there was a total of 488 alcohol-related collisions including 31 fatal crashes and 169 major injuries related between 2008 and 2018. Those accidents mostly happened on weekends between 8 pm and 3 am. In terms of alcohol-related driver behaviours, it includes 375 aggressive driving, 290 speedings, 66 redlight running crashes for the same period. It can be seen from our data analysis that the number of alcohol-related KSI is higher on weekends between 10 pm and 3 am.
Six traffic violations it identified as being the most likely to kill or injure, including:
- Failing to yield to a pedestrian.
- Failing to stop on a signal.
- Improper turns.
- Cellphone use.
- Disobeying signs.
Speeding and Red Light Running
Red-light running is a serious issue in Toronto. There were 36230 red-light-related charges in 2016. Collision statistics reveal that 21% of fatalities and 15.45% of serious injuries involve speeding. Over 40 percent of fatalities at signalized intersections are attributed to red-light running. Summarized speeding data at red lights by filtering fatalities on the map shows that speeding is one of the main factors contributed to fatal traffic collisions. Speeding caused 115 collisions and 79 fatalities and 371 major injuries between 2008 and 2018. There were 16 fatalities and 137 major injuries that occurred at red light intersections in Toronto for the same period.
The City of Toronto is planning to install automatic speed enforcement technology as part of a pair of new Vision Zero initiatives to reduce traffic-related fatalities. In an effort to improve pedestrian safety, the city is expanding the number of intersections where pedestrians will have a head start to cross the street. So far, the city has implemented the program at 12 intersections which gives pedestrians a five-second advanced green signal, including the intersection of Dundas Street West and Mabelle Avenue. The goal, Tory said, is to have the technology implemented at 80 intersections across Toronto by the end of the year. The new intersections were selected based on the number of previous collisions between pedestrians and turning vehicles.
Driver Behaviour Analysis
Driver action (automobile, motorcycle and truck) is the main factor contributing to fatal collisions. There were a total of 4195 killed or seriously injured automobile-related collisions between 2008 and 2018. It includes 449 fatal collisions, accounts for 75.72% of total fatal traffic collisions. Our data analysis shows that improper driver actions (accounted for 24% of total collisions) such as improper turn and lane change or passing, failed to yield right of way, disobeying traffic rules et, played a significant role in contributing to collisions in Toronto.
Cyclist Involved Collisions
There was 1342 cyclist involved collisions in Toronto from 2008–2018. Total 31 cyclists died and 546 cyclists seriously injured for the same period. From data analysis of cyclist behaviour, there were 17 fatal collisions caused by improper riding behaviours such as cyclist turned left across motorists path and cyclist without ROW rides into the path of motorists at intersections.
Existing Safety Measures
- Green Cycling Areas
- Toronto Cyclists Handbook
- Helmet Safety Education Videos
- Space to Cycle Campaign
- “Stay Safe, Stay Back” Campaign
New/Enhanced Safety Measures
- Cycle Tracks
- Automated Cyclist Detection
- Advance Green for Cyclists
- Signalized Crossings for Cyclists
- Enhanced Cycling Facilities
Fatal and Dangerous Intersections in Toronto
Fatal and dangerous intersection-related collisions account for 51.34% of the total road crashes in Toronto. There were 262 fatal intersection-related collisions includes 156 fatal pedestrian collisions in the past 11 years. So It is vital to identify dangerous intersections to improve road safety. Currently, there are over 300 red light cameras at 77 locations across the city operated by the participating municipalities. In an effort to improve pedestrian safety, the city is expanding the number of intersections where pedestrians will have a head start to cross the street ( CP24.com).
Collisions and Age Groups
It can be seen from the graph below that driver of age group between 20 to 29 cause more fatal and major injuries than drivers of other groups. There were 1450 collisions related to the over 55 years old age group in the last 11 years. They include 300 fatal crashes that account for 50.6% of total fatalities.
Improving Road Safety with Crowd-Sourced Data
Smart mapping, mobile apps, and data captured by members of the community can provide real-time road information. Crowd-sourced data (Waze and Live Traffic Data (Here) ) can help to improve public safety and to reduce the tragic impact of road accidents. Esri and Waze partner to offer real-time traffic data to cities. It leverages mobile technology and recommends the fastest routes based on real-time traffic data from millions of users. It allows two-way sharing of publicly available traffic and road condition information and facilitates city transport personnel to make data-driven infrastructure decisions and enhance the efficiency of incident response.
Waze data also includes a couple of attributes associated with the Waze Alert and Jam: a user rating and a report rating. Each of these attributes provides an indication of event validity. All of these tools, in conjunction with roadside Intelligent Transportation Systems, help indicate what is happening, where it is happening, and how traffic is being impacted. Either way, quicker awareness generates faster event response and reporting. Traffic control systems or police departments can begin implementing the appropriate response for traffic incident management and make it scalable based on Waze’s real-time reporting to prevent traffic accidents. Waze user reports may expose locations plagued with accidents but in lacking police coverage. This knowledge may help police departments to improve road safety by relocating the police units to these locations.
The data from Waze can provide real-time road situational awareness. It can be seen from the map on (please check the web app) the side that 2 weeks Waze data hot spots correspond with traffic crash hot spots.
Conclusion: How to Improve Road Safety?
Vision Zero is a strategy to eliminate all traffic fatalities and severe injuries while increasing safe, healthy, equitable mobility for all. It starts with the ethical belief that everyone has the right to move safely in their communities, and that road system designers and policymakers share the responsibility to ensure safe systems for travel. So every citizen needs to take responsibility and follow the rules of the road.
1. Designing a safer street. In the USA, transport organizations started to adopt European-inspired designs (road diet) that change how drivers, cyclists, and pedestrians use the road in order to reduce speeding and encourage safety for everyone.
2. It is up to both behaviour and attitudes of drivers and pedestrians to keep everyone safe on Toronto roads. (improper behavioural Issues pedestrian 55%, drivers 25%, cyclists 56%)
3. Smart Signal Optimization at Intersection Crossing
real-time traffic data and adjusting the signals to accommodate that real-time data — whether it’s pedestrian volumes or vehicle volumes.
4. Red Light Speeding Camera (set up traffic cameras in school zones, traffic hot spots and high volume intersections ).
The City of Toronto is planning to install automatic speed enforcement technology as part of a pair of new Vision Zero initiatives to reduce traffic-related fatalities.
So, do red-light cameras reduce violations, and more importantly, crashes? The simple answer is yes. Red-light cameras can play a role in improving traffic safety for all road users and should be placed where they can benefit a community, like at intersections with high numbers of fatalities. When properly implemented, red-light cameras can help save lives and can serve to supplement law enforcement efforts, rather than generate revenue for governments.
5. Increase traffic control and fine for traffic rule violators
6.A range of measures including generous pedestrian crossing times, wider curbs, medians, dedicated bus lanes, a reduction of two vehicular lanes, bike lanes and more cross-walks. painted pavements
7. Some other methods of preventing road collisions
- Vehicles and Vehicle Systems (Collision avoidance, Collision protection, Commercial vehicle safety, Off-road vehicle safety, Crash investigation and reconstruction)
- Traffic Engineering/Road Design (Impact of road design, Traffic operations and traffic control devices, Intelligent transportation systems, Speed limit setting/monitoring)
- Injury Prevention (Epidemiology of crashes/injuries, Occupant protection, Biomechanics of injury crashes, Automotive medicine)
- Enforcement/Legal Issues (Enforcement strategies, Alcohol and drugs, Legal aspects of crashes)
- Safety Initiatives (Safety promotion programs, Community and partnership engagement, Integrated road safety strategies)
- Policy and Program Development (Societal costs of crashes, Application of policies and programs, Application of legislation/regulations)
- Road Users/Behavioural Issues (Epidemiology of crashes or injuries, Specific road users approaches/research (e.g. pedestrian, cyclist, motorcyclist), Human factors and traffic psychology)
Note: WebApp and Dashboards can be accessed by clicking the following references [1,3 and 4]. Please don’t hesitate to write comments and questions.