Navigating the Intersection: How Traffic Control Types Affect Cyclist Right- of-Way using A Mixed Logit Model Analysis

Cycling safety is becoming increasingly critical as both cycling popularity and vehicular traffic continue to rise. Cyclists face heightened risks at intersections and junctions; areas where conflicting traffic flows, ambiguous right-of-way rules, and inadequate signaling often lead to collisions. Enhancing safety at these key points through improved junction design, effective traffic controls, and dedicated cycling infrastructure is essential for protecting all road users. This research investigates the behavior of drivers and cyclists at junctions under various control setup (traffic signals, stop/yield signs, and no control), with a particular focus on how intersection control influences drivers’ failure to yield the right-of-way to cyclists. Utilizing ten years of Michigan police crash investigation data, we employ a Mixed Logit Model to analyze different interaction scenarios at intersections and assess the likelihood of drivers failing to yield. The experimental results indicate that most crash characteristics; such as driver age, weekday, vehicle type, and speed; consistently affect the probability of yielding failure across different intersection control types. On the other hand, driving under the influence of alcohol or drugs shows a distinct impact on crash likelihood. Additionally, the findings reveal a negative association between the probability of a driver failing to yield and the maneuver of driving straight ahead, along with significant heterogeneity in yielding failures across various intersection control types. These insights underscore the need for a deeper understanding of driver attitudes in interactions with cyclists. Strategies aimed at reducing intersection complexity; such as better coordination of bicycle facilities with intersection control types and enhanced driver education on cyclist behavior and traffic rules, should be considered to improve overall safety.