Title

EVALUATION OF EFFECTS OF CENTERLINE RUMBLE STRIPS ON LATERAL VEHICLE PLACEMENT AND SPEED

Document Type

Journal Article

Publication Date

2004

Subject Area

operations - traffic, infrastructure - vehicle, infrastructure - bus/tram lane, planning - methods, economics - appraisal/evaluation

Keywords

Width, Two lane roads, Two lane highways, Traveled way, Traffic lanes, Statistical methods, Statistical analysis, Speed, Rumble strips, Roadway, Mathematical statistics, Lateral placement, Data collection, Data acquisition, Centerlines, Center lines, Carriageways, Before and after studies

Abstract

Transportation agencies are using a variety of rumble strip patterns to improve highway safety. Previous studies have documented the efficacy of rumble strips in reducing certain crash types. However, no research has been published on the operational effects of these low-cost safety treatments. An evaluation of the effect a treatment has on operational characteristics can be conducted and reported in much less time than an evaluation based on crash data. The effect of centerline rumble strips on lateral vehicle placement and vehicle speeds on two-lane highways was investigated with a before-and-after observational study. Data were collected using tape switches at two treatment sites and two comparison sites. The lane widths were 11 and 12 ft. Data analysis and statistical testing indicated the centerline rumble strips had a significant effect on the mean and variance of lateral vehicle placement at both treatment sites. The observed change in the mean lateral vehicle location was away from the centerline rumble strips at both treatment sites; the variance of lateral vehicle location decreased at both sites. Data collected at the corresponding comparison sites during the before-and-after periods indicated no change in the mean and variance of lateral vehicle placement. Additionally, the study found that lateral vehicle placement in travel lanes may not be normally distributed as was previously assumed. Further evaluation of lateral vehicle placement distribution should be conducted with larger sample sizes and different roadway cross sections. No relationship could be drawn between speed and the presence of rumble strips for 11- or 12-ft lanes.