Identification and testing of measures to improve motorway work zone safety

Abstract

Roadway work zones are hazardous, for both workers and motorists who drive through the complex array of signs, delineators and lane changes. Improper lane changing manoeuvres and possible vehicle encroachments in the activity areas may cause injuries to both the car occupants and road workers. Several studies agree that the presence of work zones significantly increases the risk of road crashes. Excessive speeding and high speed variances have also been identified as major causes of a large percentage of work zones crashes, injuries or fatalities. The main objective of this research was to determine safe and effective countermeasures for the reduction of speeds and speed variances within work zones. Furthermore, the influence of work zone layout features on crash occurrences was another important issue addressed with this study. An extensive accident analysis was therefore conducted on the stationary work zones of the Italian motorway network in order to identify the most critical layouts in terms of safety. The Empirical Bayes (EB) before-and-after method was performed in order to estimate crash modification factors (CMFs) associated with the installation of work zones on motorways. The findings of this part of the research have shown that all layout configurations that involve a crossover were associated with the highest values of CMFs. A number of countermeasures have been tested then in virtual reality through nine different configurations of the work zone crossover, in order to evaluate their effectiveness in reducing speeds and speed variances. The experiments investigated the effects of different speed limit sequences and alternative design features, such as wider lanes and median openings. Furthermore, the effects of different channelizing devices and perceptual treatments based on the Human Factor (HF) principles have also been investigated. The results of the experiments, performed on a total sample of 42 subjects, showed that, for all configurations, the drivers’ speeds are always higher than the posted speed limits and decrease significantly only in approaching the crossover by-passes. The implementation of higher speed limits, together with a wider median opening, led to a greater homogeneity of the driving speed. Perceptual countermeasures generally induced both the greatest homogenization of speeds and the largest reductions in mean speed values.