Exploring Relationships Between Crash Patterns and Vehicle Fleet Age: Are There Empirical Clues in Favour of Partial Driving Automation?

In recent years, most newly registered vehicles have been equipped with a combination of advanced partial automation driver assistance systems (ADAS) on board (i.e. at least level 2, according to the Society of Automotive Engineers SAE). The gradual shift towards automation hides several issues, also given the long period of mixed traffic (vehicles belonging to different automation levels) and the impact on road safety. In this study, two datasets of crashes on Italian roads (2014 and 2021) provided by ISTAT and relating to multi-lane rural roads were analysed. These datasets include several information about crash-, driver-, and vehicle-related factors. The data analysis aims to reveal any changes in crash patterns (including crash- and driver-related factors) identified at a National level, considering different classes of vehicles according to their year of registration. Results show that the percentage distribution of the type of crash was significantly changed from 2014 to 2021, with a significant reduction in run-off crashes and an increase in the hit of obstacles and stopped vehicles. Analysis of each dataset revealed that newly registered vehicles (likely equipped with updated ADAS) in both 2014 and 2021 may be associated with a reduction in run-off crashes compared to other vehicles. Although this is an indirect approach to infer clues about the effect of partial driving automation, it could nevertheless provide a new perspective based on real traffic crash occurrences, useful for further research.