Over the past 10 years, building on road infrastructure data, crash prediction models (CPMs) have become fundamental scientific tools for road safety management. However, there is a gap between state-of-the-art and state-of-the-practice, with the practical application lagging behind scientific progress. This motivated a review of international experience with CPMs from perspectives of application by practitioners and development by researchers. The objective of the paper is to improve practitioner understanding of modelling road safety performance using CPMs for crash frequency estimation, leading to their greater uptake in improving road safety. In short, why and how should road safety practitioners consider CPMs? Both scientific and practice-oriented literature was retrieved, using academic sources, as well as reports of road agencies or institutes. The selection was limited to English language. From the review it is clear that developing CPMs is not a straightforward task: there are many available choices and decisions to be made during the process without definite guidance. This explains the diversity of approaches, techniques, and model types. The paper explains how some fundamental modelling decisions affect practical aspects of modelling safety performance. There is a need to identify CPM solutions that will be scientifically sound and feasible in practitioners’ context. Together with increased communication between researchers and practitioners, these solutions will help overcome the identified challenges and increase use of CPMs.

Roundabouts are considered the safest intersection design; however, the safety effect may not be satisfactory at each specific roundabout. This is true especially in countries where roundabout design is a relatively new concept, such as in the Czech Republic. Specifically, most Czech roundabout crashes were found to occur on entries. This motivated the presented study to investigate how entry design parameters influence safety on Czech roundabouts and, if possible, use the findings to update current Czech roundabout design guidelines. To this end, the study comprised three analyses: crash-based safety performance functions, speed analysis, and finally safety performance functions which incorporated speed. All three analyses proved that entry design parameters have a statistically significant influence on safety, in terms of crash frequency, severity and speeds. Given the study objective, this fact should be considered in Czech roundabout design guidelines.

The study objective is to compare three approaches to identification of hazardous road locations: (1) Traditional reactive accident-based approach, resulting in identification of accident black spots; (2) State-of-the-art empirical Bayes method using accident prediction model, which identifies critical locations, i.e. both real and potential black spots; (3) Proactive “preliminary” road safety inspection, identifying the risk factors, which may potentially increase accident occurrence and severity. Regional rural road network (approx. 1000km) in South Moravia, Czech Republic was used. The methods were applied in identification and ranking of hazardous road locations in the studied network. It was found that black spot approach is not a suitable method, especially in low-volume road network with scattered accident occurrence. On the other hand risk index, based on road safety inspection, is a valid alternative, with ranking performance comparable to state-of-the-art empirical Bayes method. In addition both empirical Bayes and risk index methods are compared with respect to their data requirements. Their mutual application is recommended as a suitable replacement of traditional black spot management, as well as a step forward to a proactive road network safety management

Roundabout is a popular and safe type of road intersection, which has become widely used around the world, including Central European countries. To conduct statistical comparative study of traffic safety on roundabouts, accident prediction models (safety performance functions) have to be developed. With this aim accident, traffic and geometry data for samples of rural roundabouts in four Central European countries (Czech Republic, Hungary, Poland, Slovakia) were collected and used in statistical modelling, using state-of-the-art generalized linear modelling framework. Both individual and combined models were developed, which enabled insight into relationships between explanatory variables and accidents, as well as comparison with models from international studies. In the end study limitations are summarized and further improvements outlined