Crash Analysis of Work Zone Lane Closures with Left-Hand Merge and Downstream Lane Shift

Abstract

The Arkansas State Highway and Transportation Department (AHTD) began the Interstate Rehabilitation Program (IRP) in the spring of 2000 that ultimately rebuilt approximately 380 miles, or 60% of Arkansas' 655-mile Interstate system. While these projects were underway, a new lane closure method, the "Iowa weave," or lane closure with a left-hand merge and lane shift, was introduced to AHTD to effectively save time ensuring that these IRP projects were completed on schedule. Contractors on the projects used the Iowa weave at least 50% of the time for each of the projects. The settings between the Iowa weave and the conventional right-lane closure were rotated on a periodic basis depending on the work to be completed on each side of the lane. Construction officials in AHTD agreed that the use of the Iowa weave helped them to complete the rehabilitation program on time and improved traffic safety, as no major concerns were noted due to this new technique. However, the crash analysis of this weaving pattern as opposed to the conventional right-lane closure has not been assessed to determine which lane closure strategy has an advantage on the basis of actual crash experience. This research examined the work zone crashes in both of the lane closure settings by assessing them with a set of independent variables that have been known to be the factors that influence work zone crashes. In order to properly assess the comparison between the two lane closure strategies, the following primary research objectives were pursued: Examination of crash experience results from the two traffic control methods, and Reexamination of the advantages and disadvantages of the Iowa weave method as compared to the conventional right-lane closure. This study analyzed the crash data between January 1, 2000 and December 31, 2005 on the characteristics of lane closure strategies in Arkansas. The findings of the analyses were organized by: 1) Raw number of crashes (Pearson chi-square and One-Way ANOVA tests); 2) Crash rates (paired t-test); and Crash severity (binary logistic regression modeling). The analysis results of this study may be used for DOTs to reference the impressive rehabilitation work that the AHTD has completed by incorporating the use of the Iowa weave. The results of the statistical analyses showed that there was approximately a 30 percent reduction in crash rate when the Iowa weave configuration was used. Nonetheless, the final results of the logistic regression model found that the safety advantages between the Iowa weave and conventional right-lane closure in changing crash severity were not statistically significant. Traffic volume was found to be the parameter that most significantly affected crash severity in the logistic regression model. Also, the effects of lighting conditions and intersecting streets on the severity of crashes were found to be insignificant.