Initiative details
The main road safety challenge addressed in the thesis is the knowledge gap in understanding co-occurring accident conditions. The official road traffic accident statistics in Austria currently focus on identifying a single explicit accident cause for each accident, such as speeding. However, accidents are often multicausal events, involving multiple conditions or factors simultaneously. This thesis aims to investigate and analyze co-occurring accident conditions to gain a more comprehensive understanding of the factors contributing to road accidents.
Initiative date
to
Who was/is your target audience?
Policy makers
Public authorities
Educational staff
Topic
Create awareness
Provide alternative solutions
Organisation details
Vienna University of Technology
School / Research centre
Austria
Vienna
Contact name
Tabea Fian
Telephone number
+436506333777
tabea.fian@tuwien.ac.at
Website
Project activities
If you work together with external partners, list the most important partners and briefly describe their role.
The study was conducted as an independent research project in the form of a doctoral dissertation.
Please describe the project activities you carried/are carrying out and the time period over which these were implemented.
The project activities described in the thesis were carried out between 2012 and 2019. The research focuses on single-vehicle accidents with a single occupant and personal injury that occurred on the Austrian road network outside the built-up areas. The study utilizes the official Austrian road traffic accident database (UDM), which includes over 100 accident-related variables. The data from the UDM is reprocessed and analyzed using various pattern recognition methods.
Evaluation
What has been the effect of the activities?
The activities conducted in the thesis have several effects and outcomes. The primary aim is to detect recurring combinations of accident-related variables, which are referred to as "blackpatterns." By analyzing the data using pattern recognition methods such as logistic regression, decision trees, Bayesian networks, and the developed PATTERMAX-method, the thesis identifies critical variables and blackpatterns associated with severe road traffic accidents. This analysis helps to understand the relationships between accident-related variables and the likelihood of severe casualties.
Additionally, the thesis highlights statistical characteristics of road traffic accident data, discusses existing pattern recognition methods, and compares the applied methods. The combination of the PATTERMAX-method and binomial logistic regression allows for precise detection and comparison of blackpatterns. The findings and knowledge gained from the research can be used to inform the development of targeted prevention measures to address the remaining number of fatal and severe road traffic accidents in Austria.
Additionally, the thesis highlights statistical characteristics of road traffic accident data, discusses existing pattern recognition methods, and compares the applied methods. The combination of the PATTERMAX-method and binomial logistic regression allows for precise detection and comparison of blackpatterns. The findings and knowledge gained from the research can be used to inform the development of targeted prevention measures to address the remaining number of fatal and severe road traffic accidents in Austria.
Please briefly explain why your initiative is a good example of improving road safety.
Comprehensive Understanding of Accident Conditions: By addressing the knowledge gap in understanding co-occurring accident conditions, the initiative provides a more comprehensive understanding of the factors contributing to road accidents. This goes beyond identifying a single explicit cause and considers the complex interplay of multiple variables and conditions that contribute to accidents. This deeper understanding can inform targeted interventions and prevention measures.
Identification of Critical Variables and Blackpatterns: Through the application of pattern recognition methods, the initiative identifies critical variables and recurring combinations of accident-related variables, known as blackpatterns. These blackpatterns represent specific combinations of factors that have a significant impact on the occurrence of severe road traffic accidents. Identifying these patterns allows for targeted interventions and interventions that address the specific conditions and factors that contribute to accidents.
Precision in Detection and Comparison: The combination of the PATTERMAX-method and binomial logistic regression provides a precise detection and comparison of blackpatterns. This allows for a deeper analysis of how different factors, such as driver characteristics or road conditions, contribute to accident severity. Understanding these relationships can help in designing tailored interventions for specific groups or locations, ultimately improving road safety outcomes.
Potential for Prevention Measures: The research outcomes can be used to derive prevention measures that are specifically targeted to address the identified blackpatterns and critical variables. By understanding the combinations of conditions that lead to severe accidents, authorities and policymakers can develop interventions that effectively mitigate the risks associated with those conditions. This targeted approach has the potential to reduce the number of fatal and severe road traffic accidents.
Future Research and Prediction Models: The initiative also highlights avenues for future research, including expanding the investigation to accidents with multiple parties involved and exploring the use of the newly established accident database for accident prediction models. This indicates a forward-looking approach to continuously improve road safety through ongoing research and the development of predictive models that can inform proactive measures.
Identification of Critical Variables and Blackpatterns: Through the application of pattern recognition methods, the initiative identifies critical variables and recurring combinations of accident-related variables, known as blackpatterns. These blackpatterns represent specific combinations of factors that have a significant impact on the occurrence of severe road traffic accidents. Identifying these patterns allows for targeted interventions and interventions that address the specific conditions and factors that contribute to accidents.
Precision in Detection and Comparison: The combination of the PATTERMAX-method and binomial logistic regression provides a precise detection and comparison of blackpatterns. This allows for a deeper analysis of how different factors, such as driver characteristics or road conditions, contribute to accident severity. Understanding these relationships can help in designing tailored interventions for specific groups or locations, ultimately improving road safety outcomes.
Potential for Prevention Measures: The research outcomes can be used to derive prevention measures that are specifically targeted to address the identified blackpatterns and critical variables. By understanding the combinations of conditions that lead to severe accidents, authorities and policymakers can develop interventions that effectively mitigate the risks associated with those conditions. This targeted approach has the potential to reduce the number of fatal and severe road traffic accidents.
Future Research and Prediction Models: The initiative also highlights avenues for future research, including expanding the investigation to accidents with multiple parties involved and exploring the use of the newly established accident database for accident prediction models. This indicates a forward-looking approach to continuously improve road safety through ongoing research and the development of predictive models that can inform proactive measures.
How have you shared information about your project and its results?
Research papers, conference papers, presentations, lectures, teaching (university)