A simulation study in Seattle found that if 6 to 10 percent of travelers started using pre-trip traveler information during severe weather conditions, there would be a small positive impact on roadway system efficiency and mobility .
Weather data obtained from the National Climatic Data Center’s Local Climatological Data Report for the Seattle-Tacoma Airport reporting station included observations of weather event occurrences and hourly precipitation data. As part of the initial MMDI evaluation, Washington State’s Department of Transportation (WSDOT’s) traffic web site utilization was tracked using WebTrends™ log analysis software. The data files provided usage information in three forms: hits, page views and user sessions. The page view, which is a combination of hits providing the type and amount of information accessed by a user, was selected to measure web site activity. Battelle analyzed a cross section of web site usage by examining three months: May 1998, December 1998 and April 1999. For weather data, the daily analysis periods were the AM peak period between 6:00 AM and 9:00 AM, the PM peak period between 4:00 PM and 7:00 PM, and the AM and PM peak period combination. To define weather and snow events, the peak period categories were further broken down by water equivalent – 0.01 inches (0.25 mm), 0.05 inches (1.27 mm), 0.10 inches (2.54 mm), 0.15 inches (3.81 mm) and 0.20 inches (5.08 mm) and snow. A weather event was defined as the combination of AM and PM peak periods where the measurable water equivalent rainfall was 0.01 inches (0.25 mm) or greater. Using this criteria, 16 of 64 days were deemed weather event days and one of the 64 days was deemed a snow day.
Analysis of the weather and web site usage data revealed that (1) during a weather event, web site activity – measured in average page views – increased by 26.52 percent, (2) that there was a 69.09 percent increase in average page views during a snow event and (3) that ATIS penetration is not evenly distributed as originally assumed (i.e., when there was some type of weather event, web site activity increased). These results were used to develop a non-uniform distribution of ATIS users, while maintaining an average of six percent. Non-uniform values were found to be 5.70 percent during non-weather events, 7.22 percent during weather events and 9.66 percent during snow events. The comparison analysis of non-uniform distribution results to MMDI baseline data included the percent difference and statistical significance for five measures of effectiveness for "Freeway Only" and "Freeway and Arterial" scenarios. The measures of effectiveness were Total Number of Stops, Adjusted Travel Time, Coefficient of Variation, Vehicle Throughput and Vehicle Kilometers of Travel. Most of the comparison analysis results were small, with minimal impact on the overall system. Only four of the ten comparisons had a strong statistical significance. For the "Freeway and Arterial" scenario, the Total Number of Stops decreased by 6.39 percent, the Adjusted Travel Time decreased by 0.74 percent, and the Vehicle Kilometers of Travel increased by 0.15 percent. The Coefficient of Variation for the "Freeway Only" scenarios decreased by 0.62 percent indicating that the travel time is more reliable. The results of the network analysis show that a non-uniform ATIS utilization rate related to severe weather has a small positive impact on roadway system efficiency.
Metropolitan Model Deployment Initiative Seattle Evaluation Report: Final Draft, May 2000.
Impacts of Supplementing Web-Based Urban Freeway ATIS With Parallel Arterial Travel-Time Data, November 2000.
ITS Impacts Assessment for Seattle MMDI Evaluation: Modeling Methodology and Results, September 1999.
Analyzing the Effects of Web-based Traffic Information and Weather Events in the Seattle Puget Sound Region: Draft Report
Author: Hardy, Matthew; James Larkin; and Karl Wunderlich
Published By: Federal Highway Administration, U.S. DOT
Prepared by Mitretek for the U.S. DOT
Source Date: October 2000
Average User Rating
Related Metropolitan Integration Links
Typical Deployment Locations
Metropolitan Areas, Statewide