Estimation of relationships between 85th percentile speeds, speed deviations, roadway and roadside geometry and traffic control in freeway work zones

The Manual on Uniform Traffic Control Devices defines a work zone as an area of highway with construction, maintenance or utility work activities. New federal work zone regulations require states to continually pursue improvement of work zone safety and mobility by analyzing work zone crash and operational data to enhance state processes and procedures. Work zone design guidance was identified as an area of needed improvement. Current guidance is heavily based on desirable speed-related outcomes, but knowledge related to actual speed-related outcomes of design and traffic control decisions is limited. The objective of this research was to investigate relationships between speed behavior, roadway and roadside geometrics and traffic control in work zones. The research objective was accomplished through specification, estimation, evaluation and interpretation of a series of econometric models. Four speed-related performance measures were modeled: 85th percentile passenger car speed, 85th percentile truck speed, passenger car speed deviation and truck speed deviation. Data for model estimation were collected in Pennsylvania and Texas work zones. Three issues were addressed by the work zone speed models: contemporaneous correlation between equation disturbances, contemporaneous relationships between dependent variables and autocorrelation. A simultaneous equation model estimated with three-stage least squares was recommended. Effects of work design and traffic control features on speed were observed but small in magnitude. Additional data collection and modeling activities are recommended before direct implementation of speed findings into work zone design practice, including observation of larger samples with greater variability in geometric design elements and investigation of purely predictive modeling techniques. Selected results and conclusions have near-term value. No geometric or traffic control elements showed a direct effect on 85th percentile passenger car speeds. Passenger car speeds were controlled mostly by truck speeds, which were directly influenced by posted speed, work zone type, type of infrastructure and vertical alignment. Truck speed deviations were lower in work zones with a posted speed reduction of 10 or 15 mph than in work zones with no posted speed reduction. In addition, passenger car speed deviations were higher in work zones with a 70 mph posted speed limit compared to other posted speeds. Both of these findings contradict work zone posted speed guidance in the MUTCD, that a decrease in posted speed causes an increase in speed variance and that posted speed reductions should be avoided or limited to 10 mph for this reason. In addition, passenger car speed deviations were lower in work zone areas with either a temporary concrete barrier or permanent roadside conditions compared to areas with drums, vertical panels or other similar roadside devices.