Researchers at University of Central Florida, in conjunction with the city of Orlando, are utilizing real-time traffic data to reveal strategies for overcoming car crashes and conclusively creating fatality-free roads.
The team at the University of Central Florida has been employing data sources for real-time collision prediction for numerous years. This application incipiently started in cooperation with the Florida Department of Transportation and the Colorado Department of Transportation. The most current research for this work developed out of the U.S. Department of Transportation’s Solving for Safety Visualization Challenge, for which the said work has been selected as a Stage-I semifinalist. As part of that challenge, the team affirmed that by combining real-time and static data, they could reveal predictive analytics to diagnose real-time traffic security conditions.
The development of analytics software and the evolution of a data-rich environment provide to the data-driven analysis concerning traffic safety by investigating the crash, weather, traffic, weather, geometric data. Initially, traffic safety data analyses were conducted based on static and extremely aggregated data, like yearly average daily traffic or annual crash regularity. These aggregated data analyses can only expose the general trend and relationship between crash frequency and few supporting factors, which could result in erroneous findings merely because they are aggregates and cannot embody the real conditions at the time of a crash.
The input data are the framework to conduct real-time data-driven analysis for road safety. In current years, with the progression of big data, abundant data could be utilized for better crash prediction. The team has utilized these data — including Automatic Vehicle Identification, Microwave Vehicle Detection, Bluetooth, and real-time weather — on both arterials and highways, including the common conflict areas to generate accurate algorithms that can foretell the rise in crash risk in real time. Producing all these tools and algorithms below one integrated system will permit operators to monitor safety risk in real-time and develop interventions that decrease the potential problems and predict crashes or at least minimize their severity.
The arrival of big data technologies enables real-time analysis of traffic safety. By combining aggregated data sources, the data could help the team understand the relationship between the appearance of traffic conflicts and real-time supporting factors and quantify the influence of the mentioned factors on real-time crash risk. It is well-known that maximum crashes happen because of the appearance of conflicts between different road users. For example, the conflicts between a head and a resulting vehicle could lead to rear-end collisions. To date, diverse real-time countermeasures have been deployed to avoid conflicts.
The availability of complete microscopic data is a significant enabler and the opportunity for the first time to prevent or decrease the severity of crashes in real time.
Visualization tools that make safety knowledge accessible and credible allow the team to operationalize the insights. Visualization gives them the capacity to go from being reactive to proactive to circumscribe the need for countermeasures before any alarming trends occurring along roadways. The same represents an important step in their ability to deliver a fatality-free roadway system. The department is doing this in connection with the city of Orlando and the University of Central Florida. The team believes this is the direction towards the future.
Traffic data from different detectors could accommodate all the traffic variables such as volume, average speed and lane occupancy, and regular deviation of these parameters during particular time intervals to represent the traffic conditions and turbulence before the crash occurs. Traffic control data could present real-time traffic control statuses and help to predict traffic conditions in the future time intervals. Other environmental knowledge could be obtained from the roadway features inventory database, a local climatological data set.