Safety analysis of freeway interchange speed change lane facilities

[EMBARGOED UNTIL 6/1/2023] The entrance speed change lane is an uncontrolled terminal between the entrance ramp and freeway, with the primary purpose of creating a merging area for entering vehicles. The exit speed change lane is an uncontrolled terminal between the freeway and exit ramp, with the primary purpose of creating a diverging area for exiting vehicles. Due to the merging and diverging maneuvers, the entrance and exit speed change lane segments experience more conflicts than a basic freeway segment. One study found that 36 percent of all crashes at a freeway interchange occur as drivers enter the freeway, and 53 percent of all crashes at a freeway interchange occur as divers exit the freeway. The same study also reported that freeway interchanges witness higher crashes per mile than other freeway segments. A better understanding of the factors that cause crashes is critical to improve traffic safety within the entrance and exit speed change lanes. This study uses two approaches to better understand crash causation within the entrance and exit speed change lanes. The first approach uses SHRP2 Naturalistic Driving Study (NDS) data, which provides an unprecedented opportunity for drivers' interaction with their environment in safety-critical events. Three kinematic measures were established to analyze rear-end near-crash events. These are the critical time to collision (TTC), advance warning time to collision (AWTTC), and speed reduction of lead vehicles. The analysis was conducted for all traffic densities and separately for each level of service. The study also uses inverse TTC profiles to identify the highest crash risk. The average critical TTC upon applying the brakes for entrance and exit speed change lanes was found to be 1.65 seconds and 1.61 seconds, respectively. In addition, the average AWTTC for entrance and exit speed change lanes was found to be 4.77 seconds and 7.26 seconds. The exit speed change lane has a greater AWTTC than the entrance speed change lane, which could be related to a higher reduction in speed of leader vehicles. AWTTC can be useful for designing collision warning systems such as forward collision warning. For entrance speed change lane, it was also found that the lead vehicle's speed reduction while transitioning from safe to unsafe conditions (based on TTC values) were higher for lower traffic densities. The percentage speed reductions were 60 percent, 30 percent, 30 percent, 31 percent, and 23 percent for LOS A, B, C, D, and E, respectively. For exit speed change lane, it was found that the lead vehicle's speed reduction while transitioning from safe to unsafe conditions were higher for LOS D. The percentage speed reductions were 35 percent, 51 percent, 53 percent, and 27 percent for LOS B, C, D, and E, respectively. The second approach uses observed crash data from two states -- Washington (five years) and Missouri (three years) to fit regression model and develop crash frequency factors that account for the effect of traffic volume and roadway geometrics. Crash frequency factors were calculated for freeway AADT, entrance ramp AADT, exit ramp AADT, number of lanes, and speed limits. These factors can be used by safety practitioners in allocating resources for safety improvements. Taken together, these microscopic and macroscopic approaches provide a comprehensive framework to evaluate the safety of speed change lane facilities.