Driver violations at traffic signals are a major cause of intersection vehicle crashes. The yellow interval is used to inform approaching drivers of an upcoming change in the traffic signal indication from green to red. Current yellow-interval durations are currently calculated to accommodate for dilemma zone protection for passenger cars only. Buses with different vehicle, driver, and occupancy characteristics behave differently at the onset of a yellow indication. The research presented in this thesis characterizes the difference between bus and passenger car driver behavior at the onset of yellow-indication. A revised set of yellow timing procedures are presented to address the requirements for bus dilemma zone protection. A dataset of 864 stop-go records were collected as part of the research effort using a school bus approaching a traffic signal on the Virginia Smart Road facility. The experiment was conducted at an instructed speed limit of 57 km/h (35 mph) approach speed where participant drivers were presented with yellow indications. A total of 36 participating bus drivers were randomly selected from three age groups (under 40 years old, 40 to 64 years old and 65 and above) with equal number of male and female for each age group. Using the data collected as part of this research effort, statistical models were created to model bus driver perception-reaction times (PRTs) and deceleration levels considering driver attributes (age and gender), roadway grade, vehicle approach speed, and time to intersection (TTI) at the onset of the yellow indication. A Monte-Carlo simulation was conducted to develop appropriate yellow indication timings to provide adequate dilemma zone protection for buses. Lookup tables were then developed for different reliability levels to provide practical guidelines for the design of yellow signal timings to accommodate different bus percentages within the traffic stream. The recommended change durations can be integrated within the Vehicle Infrastructure Integration (VII) initiative to provide customizable driver warnings prior to a transition to a red indication. / Master of Science

Driver violations of traffic signals are a major cause of intersection vehicle crashes. The duration of yellow intervals is highly associated with driver yellow/red time stopping behavior. Rainy weather and wet pavement surface conditions may result in changes in both driver behavior and vehicle performance. The research presented in this thesis quantifies the impact of wet pavement surface and rainy weather conditions on driver perception-reaction times (PRTs) and deceleration levels, which are used in statistical models for the design of yellow intervals. A new dataset with a total of 648 stop-run records were collected as part of the research effort during rainy weather and wet pavement surface conditions at the Virginia Department of Transportation's Smart Road facility. This experiment was conducted at a 72.4 km/h (45 mi/h) approach speed where participant drivers encountered a yellow indication initiation. The participant drivers were randomly selected in different age groups (under 40 years old, 40 to 59 years old, and 60 years of age or older) and genders (female and male). Combined with an existing dataset that was collected by the same research group under clear weather conditions during the summer of 2008, statistical models for driver PRT and deceleration levels are developed, considering roadway surface and environmental parameters, driver attributes (age and gender), roadway grade, and time to the intersection at the onset of yellow. Using the state-of-the-practice procedures with the modeled PRT and deceleration levels, inclement weather yellow timings are then developed as a function of different factors (e.g., driver age/gender, roadway grade, speed limits, and precipitation levels). The results indicate that an increase in the duration of change interval is required for wet roadway surface and rainy weather conditions. Lookup tables are developed with different reliability levels to provide practical guidelines for the design of yellow signal timings in wet and rainy weather conditions. These recommended change durations can be integrated within the Vehicle Infrastructure Integration (VII) initiative to provide customizable driver warnings prior to a transition to a red indication. / Master of Science