Empowering road safety engineers with surrogate safety data:  A SCIENTIFIC, SAFE SYSTEMS, & DECISION-BASED APPROACH. 



MicroTraffic’s approach to surrogate safety is focused on science, the safe systems concept, and the practical decision-making needs of engineers. We start with safe systems surrogate indicators that quantify risk according to the potential for severe or fatal injury. We apply benchmarks and context to identify abnormally high risks. We use probabilistic risk predictions to estimate the probability of a fatality during the next five years based on a short window of surrogate data. Finally, we use a smart solutions recommender that leverages AI and countermeasure research to identify context-relevant risk mitigation options for engineers to consider.

MicroTraffic is led by practicing road safety consulting engineers who have Ph.D.'s in road safety, and who have completed hundreds of in-service road safety reviews and road safety audits around the world.

MicroTraffic's approach reflects our founder's commitment to using a scientific approach when empowering engineers with practical data for decision-making.



A large proportion of the research in surrogate safety focuses on Post Encroachment Time, Time-to-Collision, and Evasive Action indicators.

These indicators measure the likelihood of a collision, but they do not do a great job of capturing that collision’s potential for severe injury. We set out to develop indicators that we refer to as safe systems surrogate indicators, focusing on the human body's limited ability to withstand forces during kinetic energy transfer. 

Our safe systems surrogate indicators leverage research on injury biomechanics to categorize an interaction’s risk level according to the configuration of the potential collision, the impact angle, the types of road users involved, the speeds involved, and the time by which a collision was avoided.



We believe that all surrogate data has to be interpreted in context in order to be useful in decision-making. The two elements of context that we work with are exposure and benchmarks.

When we identify a number of critical risk conflict events (e.g. 20 in 1 week), the first piece of contextual information is exposure. If 20 out of 2000 cyclists were involved in a critical conflict with left turning vehicles during that week, then, in the context of exposure, 1% of cyclists were involved in critical risk events.

The second element of context is a comparison to benchmarks. Suppose that the benchmark rate suggests that only 0.2% of cyclists are expected to be involved in critical risk events. In this case, we would say the risk of critical conflicts is 5 times higher than the benchmark (1% vs 0.2%). The relative risk multiple is the ratio of the observed risk rate to the benchmark risk rate.

By using exposure, benchmark comparisons, and relative risk multiples, we can help to identify interacting movements that pose abnormally high risks. This information would suggest that these interacting movements are strong candidates for a risk management intervention by traffic engineers.



In order to make evidence-based engineering decisions in road safety, it is useful to have predictions of future safety levels.


We have adapted the predictive methods of the AASHTO Highway Safety Manual to predict the probability of a fatal collision during a forward-looking time interval based on surrogates observed during a short period of road safety video analytics. 


In our view, one of the most useful pieces of information that we can provide is a statement in the form “The movement type X at location Y is estimated to have a Z% chance of a fatality during the next 5 years.” In this way, we provide a quantitative risk understanding in terms of adverse outcome probabilities.



The most important risk management question is ‘What should we do?’. By leveraging thousands of past studies on collision modification effectiveness, our smart solutions recommender generates a list of context-sensitive risk management interventions that could be effective in reducing the chance of a fatality. The list of interventions typically includes operational and geometric concepts, with both short and long-term options. A local engineer can start with this list, and, with a little additional investigation, can start implementing an optimal course of action.


At the end of the day, MicroTraffic intends to help engineers in four ways. We will provide data on near-miss frequency using safe systems surrogate indicators. We provide contextual information on this data to help you know if risk levels are abnormally high. We help you understand this information in terms of a probability estimate of future fatalities. Finally, we will empower you to make risk management decisions based on a context-sensitive list of effective interventions proven to be effective for the identified risks.


We invite you to book a 30 minute web meeting with our lead road safety engineer, Dr. Craig Milligan, P.Eng., to learn more about the science of surrogate safety and how to integrate it into your engineering work.  

30 Minute Booking Link

MicroTraffic is a leading provider of microscopic traffic data for road safety engineers. Our technology is based on computer vision, video analytics, machine learning, predictive modelling, and the safe systems philosophy. We assist engineers to apply surrogate safety methods for proactive decision making. We produce conflict reports for individual traffic safety studies and also offer a proactive road safety network screening service.

© 2020 by MicroTraffic Inc.