Doctoral Defense - Sagar Keshari

About the Event

The Department of Civil & Environmental Engineering

Michigan State University

Ph.D. Dissertation Defense

July 22nd, 2025 at 10:00 AM EST

3540 Engineering Building (Seminar Room)

Abstract

DRIVER RESPONSE TO WINTER WEATHER WARNING SYSTEMS AT BRIDGE OVERPASSES AND HORIZONTAL CURVES

By: Sagar Keshari
Advisor: Timothy Gates

Transportation agencies often deploy warning signs to warn motorists of potentially icy road surfaces during winter travel conditions. However, prior research has shown that standard warning signs tend to have little to no effects on driver behavior or safety performance. The conspicuity and message recognition of traditional winter weather warning signs may be improved by adding a flashing beacon or flashing LED sign border to attract attention, but these systems have limited effectiveness if they flash continuously, regardless of the pavement surface condition. To counter this, some transportation agencies have begun to implement condition-responsive winter weather warning systems such as bridge deck warning systems (BDWS), and slippery curve warning systems (SCWS). Such systems improve upon the “always-on” warnings by providing warning alerts or messages only when warranted based on environmental sensors that provide real-time weather and pavement conditions. Typically, the warning alert consists of an LED sign border or flashing beacon(s) on a standard warning sign or weather warning messages displayed on a changeable message sign (CMS). However, due to the limited implementation of condition-responsive systems in the United States, the effectiveness of these systems on driver behavior during winter travel conditions has not been well established. To that end, a series of field evaluations were performed at multiple freeway bridge overpasses and rural horizontal curves to assess the effectiveness of winter weather warning systems as a speed reduction countermeasure during winter weather conditions. Bridge overpasses and horizontal curves were selected for this study due to their susceptibility towards icing, creating critical safety hazards for speeding drivers. Hand-held LiDAR guns were utilized to track the speeds of free-flowing vehicles traversing each study site during periods when the pavement surface was wet and temperature was below 32F. Several measures of effectiveness were utilized to assess the effectiveness of the warning system, including the magnitude of the speed reduction, the likelihood of a speed reduction, and compliance with the speed limit or advisory speed. The data were analyzed using either linear regression (magnitude of speed reduction) or logistic regression (likelihood of speed reduction or speed limit compliance). Separate analyses were performed for passenger vehicles and heavy vehicles and for faster drivers (top one-third), average drivers (middle one-third), and slower drivers (bottom one-third) based on their speed upstream of the warning system.

Considering all phases of field evaluations, it was concluded that the winter weather warning systems have a significant impact on reducing driver speeds approaching bridge overpasses and horizontal curves during winter weather conditions. Specifically, a BDWS with a flashing light-emitting diode (LED) border reduced motorist speeds when encountering a bridge during winter weather conditions compared with the standard “Bridge Ices Before Road” warning sign. The speed reduction effect was consistent between passenger cars and heavy trucks but was somewhat stronger at night compared with daytime. Greater speed reductions were observed when the BDWS sign was combined with a dynamic speed feedback sign (DSFS) that displayed a “SLOW DOWN” or “ICY ROAD” message to approaching motorists, with the strongest effects observed when the DSFS message was pulsed between high and low brightness intensities. Regarding the CMS winter weather warning messages, the most substantial impacts, both in terms of the speed reduction likelihood and magnitude, were observed when the "SLIPPERY ROAD CONDITIONS / REDUCE SPEEDS" message was displayed on the CMS. Additionally, the winter weather warning messages were found to have the strongest speed reduction effects on the fastest one-third of drivers, which is the most vulnerable group from a crash severity risk standpoint and, thus, are considered the most highly targeted driver group for speed reductions. Furthermore, the evaluation of SCWS also showed a significant speed reduction effect for drivers approaching a horizontal curve during winter weather conditions, and the strongest speed reduction effects were observed among the fastest group of drivers. These findings were subsequently utilized towards the development of guidelines and recommendations for future use of winter weather warning systems.