Researchers from MUNI’s Faculty of Economics and Administration have developed new software that analyses traffic accident data to identify for the police the most dangerous places in the road network. Photo credit: Freepik.
Brno, April 4 (BD) – Researchers from the Faculty of Economics and Administration (FEA) at Masaryk University have created an app that highlights the places where traffic accidents are more likely to happen. The app will help police officers identify problematic locations and take preventive measures.
Although the number of deadly traffic accidents has been declining, more than 500 people still die on Czech highways every year, and more suffer permanent injuries. This is why Michal Kvasnička and Štěpán Mikula from MUNI FEA, together with Josef Montag and Peter Bolcha from the Faculty of Law of Charles University, have developed unique software which analyses of traffic accident data to identify the most dangerous places in the road network, improving the use of the statistical data available to the Czech traffic police service.
“There are too many accidents on the roads to be analysed with the naked eye, and we are helping the traffic police service solve this problem,” said Kvasnička. “Our application looks for clusters of serious accidents that signal a possible dangerous location. Therefore, we help the Czech Police use the data obtained during standard traffic accident investigations.”
Kvasnička emphasised that the app will assist, rather than replace, the expert knowledge of police officers: “Imagine that the application finds a cluster of serious traffic accidents at a certain location. A police analyst can determine that the accidents happened due to the reconstruction of a road that they know is already finished. In that case, these accidents can be ignored. But if the police officer finds a risky cluster of accidents that does not have such a clear and complete cause, they can then analyse the individual accidents to reveal it, and propose measures to reduce the risk at the given location. So our app can be a useful help, but the role of expert knowledge of police analysts is still irreplaceable.”
Compared to previous similar technology, the new software is innovative in three ways. First of all, it analyses the entire road network in the Czech Republic and is therefore able to identify potentially dangerous places not only on uninterrupted road sections, but also at intersections. The second advantage is that the application allows traffic accidents to be distinguished according to their severity. Without this feature, for example, parking lots would appear very dangerous, even though the accidents that happen there are usually much less serious. Third, compared to other programs, the system does not look for places where accidents are unexpected, but where they are likely.
“Our software is unique in that it uses new scientific knowledge in the field of statistical identification, thanks to which it can find places with an increased number of traffic accidents,” said Josef Montag. “At the same time, it works with real data from the entire transport network of the Czech Republic, which is available to the Czech Police. We believe that our software, which is now available in the public version for anyone to use, will enable the development of similar tools in other countries and data areas.”
The application functions primarily as a warning system. Police officers combine these warnings with their own expert knowledge of a certain location, data on the volume of traffic flows, ongoing road improvements, and so on. It is hoped that combining these information sources will allow police officers to implement quick and effective measures at the most dangerous places on Czech roads.
A sample version of the application is available here: https://trafficacc.econ.muni.cz/