The Challenge of Traffic Efficiency and Safety

Can Network-Wide Real-Time Safety Evaluation Resolve the Decades-Old Conflict Between Mobility and Safety?

By Dr. Tarek Sayed

Historically, the management of road transportation has involved tradeoffs between two primary considerations: traffic efficiency (mobility) and safety. Even today, these are often considered to be in direct conflict, and departments of transportation typically have entirely different teams managing each, often with little coordination between the two. Increasing mobility through higher volumes and speeds often meant increasing exposure to crash risk and more serious crash severities. The majority of safety countermeasures aimed at reducing the exposure to crash risk did so by reducing mobility (e.g. no right-turn on red, prohibited left turns, etc.).

As we look to the future, there is huge potential that the use of advanced technologies, the availability of ‘big data”, and the capability for real-time safety optimization of transportation networks will resolve the mobility and safety conflict, allowing each to be optimized together.

Real-time crash risk prediction (i.e., real-time safety analysis), is one of the most important components of proactive traffic safety management. The objective of real-time safety analysis is mainly to evaluate the real-time crash risk with dynamic traffic data that are collected and updated within a very short time interval (e.g., 5 min, cycle length, etc.).

Fig 1. Real-time Safety evaluation of Signals at the Cycle Level [4]
Fig 1. Real-time Safety evaluation of Signals at the Cycle Level [4]

A large amount of real-time traffic data will soon become available as a result of advances in connected and autonomous vehicles (CAVs) and sensor technologies. These data can be used to develop real-time safety evaluation models that are based on surrogate safety measures such as near-misses and traffic conflicts. Several of these models have been developed at the University of British Columbia (UBC) and have been used to dynamically identify crash-risk prone locations [1,2]. These models were also used to propose two real-time safety indices, 1) risk of crash (RC) and 2) return level of a cycle (RLC), to measure the safety of locations such as signalized intersections at the cycle level (Fig. 1). For example, a RC value of zero means no crash risk for a cycle, while a RC > 0  indicates a crash prone cycle because of positive risk of crash [3].

The availability of real-time data and safety indices enables the real-time safety optimization of traffic and proposing dynamic operational safety improvements at monitored locations such as signal phasing changes, variable speed limits, ramp metering, etc., or at the network level (e.g. rerouting of traffic) [4].

In the near-term the required data will be available from video and/or lidar sensors installed at continuously monitored locations. In the not-too-distant future, these data will be supplemented by video and lidar data provided by vehicles themselves which will provide coverage of the entire road network.

Currently, research is underway at UBC using data from Waymo vehicles to measure traffic conflicts in real-time across an entire city. These massive amounts of sensor data will be processed at the EDGE, providing low latency between measuring traffic movements and providing actionable data to departments of transportation, infrastructure (e.g. traffic signals, message signs etc.) and road users themselves.

In recognition of this need, there are many vendors now providing or developing EDGE processing technologies. AMAG’s SaaS platform for proactive transport analytics will allow actionable insights to be derived and managed from this massive amount of data.


  1. Essa, M. and Sayed, T. (2018) “Traffic Conflict Models to Evaluate the Safety of Signalized Intersections at the Cycle Level”, Transportation Research Part C, Vol. 89, pp. 289-302.
  2. Essa, M. and Sayed, T. (2019) “Full Bayesian Conflict-based Models for Real Time Safety Evaluation of Signalized Intersections”, Accident Analysis and Prevention, Volume 129, pp. 367-381.
  3. Essa, M., and Sayed, T. (2020) “Self-learning adaptive traffic signal control for real-time safety optimization”, Accident Analysis and Prevention, Vol. 146, 105713.
  4. Zheng, L. and Sayed, T. (2020) “A novel approach for real time crash prediction at signalized intersections”, Transportation Research Part C, 117, 102683.

About the Author

Dr. Tarek Sayed is a Co-Founder of Advanced Mobility Analytics Group (AMAG). Professor Sayed has won numerous awards in recognition of his contributions to road safety, including the UBC Distinguished University Scholar, ITE Wilbur Smith Distinguished Transportation Educator, Canadian Society of Civil Engineering Sandford Fleming, and Transportation Association of Canada Academic Merit, Transportation Association of Canada Gilchrist Medal and several best paper and teaching awards. Professor Sayed is a world-renowned leading authority in the development of video analytic methods for road safety and pioneered much of the early work in this field.