Forecasting Road Safety in Olsztyn

The first step in consciously managing the safety of the road network is to identify the most dangerous sections. With this knowledge, the road manager can choose the most effective actions and monitor their implementation. However, Heller Consult offers something more: forecasting the future road situation, which gives the possibility to effectively prevent road incidents.

Safety performance factor


The Provincial Roads Authority in Olsztyn (ZDW Olsztyn) manages a road network of 1,881 km. It carries out its activities on the basis of the adopted development strategy. In accordance with the priorities set out therein, it carries out a range of activities in the area of modern road management and safety, among others.


ZDW Olsztyn needed to define indicators that would enable accident prediction and thus the introduction of safety improvement solutions before road incidents occur. When they turned to us, they did not have any such solutions in operation, and the activities carried out so far concerned locations where accidents had already occurred.


In 2020, we have determined the potentially dangerous road sections in the Warmia and Mazury region using a traffic safety prediction function. It is based on a model developed at Virginia Tech University.

We had to adapt this function to the conditions on the surveyed network. In our analysis, we took into account not only incident information, but also the characteristics of the network – including information about trees (we determined trees in the verge and tree avenues based on data provided by the client), built-up and undeveloped areas, and intersections. In addition, we took into account data on maintenance works, repairs and rebuilds of sections in order to correctly interpret the results and to properly calibrate the model.

The result of our analysis is the Safety Performance Function (SPF), which indicates the degree of safety on a given road section.

Safety performance factor


The road authority received a Road Safety Analysis from us, which included:

  • recommendations for the sections with the highest road safety risk,
  • a report summarising the measures, with a division of the network into road hazardous sections,
  • thematic maps containing dangerous sections with specific coefficient values and road incidents.

In this way, it was able to effectively prevent traffic incidents and potentially tragic accidents.

The whole process is still being developed, from the potentially dangerous sections we are designating aggregated areas for which audits will be carried out. We are working on a model in which we additionally take speed measurements into account, dividing roads into types according to their location: route sections, built-up areas, areas with urban fabric characteristics.

Find out more about the project – contact us.