Proactive Safety Management Empowered by Big Data

In the last few decades, data-driven methods have been used to assist with key tasks of road safety management like hotspot identification, countermeasure development, and before-after evaluation. These methods have traditionally relied heavily on historical crash data for safety assessment, which can take a long time to collect. Professor Kun Xie will share a more proactive and time-efficient approach based on surrogate safety measures (SSMs), which can assess safety by capturing the more frequent “near-crash” situations. Massive amounts of data from emerging sources like GPS devices, smartphone apps, traffic cameras, naturalistic driving, and connected vehicles (CV) can be leveraged to extract SSMs on a large scale, presenting new opportunities for proactive traffic safety management. Results will show that risk status is a reliable criterion for safety assessment, and promisingly point towards the use CV data for proactive traffic safety management.