If you work together with external partners, list the most important partners and briefly describe their role.
Municipality of Rotterdam: User of the anonymous driver behaviour data, and initiated an Artificial Intelligence model to calculate accident risk in a detailed manner at the level of road sections and road crossings. Their development included the provision of a user interface showing the risk of traffic accidents on a detailed road map.
Please describe the project activities you carried/are carrying out and the time period over which these were implemented.
In 2016 we started our ‘ANWB Safe Drive car insurance’ to coach and improve driver behaviour. The data we collected directly from the car or from the sensors of the driver’s mobile phone is used to provide feedback to a driver on his driving behaviour. Detailed feedback and coaching information is provided every 10 days through our mobile app which belongs to this car insurance. Safe driving is rewarded by a discount on the insurance premium, which for sure encourages safe driving behaviour. From the start it has been the goal to use the collected data to improve road safety in general as well. Having this vision from the start resulted in being prepared for the next stage.
After a pilot with road authorities in 2020, a formal data-service was implemented at ANWB early 2021. In March 2021 the car insurance data was provided officially to the first road authority to enable data driven improvement of road safety by assessing the road infrastructure. At the moment driving behaviour data is available continuously from about 75.000 cars in The Netherlands, resulting in about 18 million data events in 2021. Data is historically available from mid-2016. Next to data about driving speed, data is available about harsh braking and harsh cornering, including the precise location, date and time of the event. By including this driving behaviour data into road safety models it was proven to be a strong predictor for traffic accidents.