Analysis of Factors Affecting the Over-Representation of Sequential Crashes in Freeway Tunnels: Using Rule-Based Data Mining Method

The paper provides an empirical analysis of road/tunnel design, traffic volume, and environmental factors associated with the increased likelihood sequential crashes in freeway tunnels. association rule mining decision tree methods are employed since both them capable identifying complicated interactions among variables expressing form rules. Results show that tunnel length, congestion, time day, season, vehicle type significant influencing More importantly, reveal a combination traffic, produces even higher crashes, leading to series hazardous situations. For example, when including long grade ≤ 2%, fourth level, winter combined, proportion is more than twice average complete crash database. Traffic safety management should pay attention monitoring these situations which likely be linked crashes.