Available at: https://digitalcommons.calpoly.edu/theses/3154
Date of Award
2025
Degree Name
MS in Civil and Environmental Engineering
Department/Program
Civil and Environmental Engineering
College
College of Engineering
Advisor
Anurag Pande
Advisor Department
Civil and Environmental Engineering
Advisor College
College of Engineering
Abstract
This thesis builds upon a prior study that analyzed the safety effects of Edge Lane Roads (ELRs) within the U.S. context, addressing gaps related to collision, traffic volume, and speed data. An ELR is a roadway configuration that accommodates two-way automobile traffic within a single center lane while providing edge lanes for vulnerable road users (VRUs), i.e., bicyclists or pedestrians. Although the concept of a single shared lane for opposing traffic often raises safety concerns among the public, ELRs have demonstrated potential to control speeds, optimize roadway space, and improve traffic operations on low-volume roads.
To evaluate safety outcomes, this thesis applies the project-level empirical Bayes before-and-after method, focusing on crash data and exposure measures (traffic volumes and counts) before and after ELR installation. The dataset was organized into six categories: received data; only sites with 5 years before and 3 years after installation; removal of terminal, off-facility, and near installation crashes; pedestrian and cyclist data; and motor vehicle (MV) only crashes. Results from a naïve comparison of crash frequencies were compared with those from the EB approach. Findings suggest that ELRs may reduce crashes involving both motor vehicles and VRUs on low-volume roads where conventional designs often lack sufficient safety features. Prior studies indicate that roadway width is a key factor influencing crash frequency, and that narrower lanes or alternative designs such as ELRs can improve safety by lowering speeds and increasing driver attentiveness. Nonetheless, further analysis is needed to evaluate the performance of ELRs across different roadway types and jurisdictions. A comprehensive database of ELR installations, including site-specific characteristics and treatment details, is maintained at www.edgelaneroads.com. To ensure comparability, this thesis utilized crash data from multiple jurisdictions consistently, with detailed inclusion criteria outlined herein.
The EB before-and-after analysis, conducted using the 2010 Highway Safety Manual–recommended approach, showed that ELR conversions yielded a crash modification factor (CMF) across an 8-year evaluation period at sites throughout the U.S. A CMF below 1.0 indicates a reduction in crashes and a safety improvement. The overall CMF was 0.67, suggesting a 33% reduction in crashes. However, sites with higher volumes and wider center lanes presented challenges, as increased vehicle interaction within the shared center lane may reduce effectiveness. When results from this study were combined with those of a previous study, the aggregated analysis across 28 ELRs demonstrated a 42% crash reduction using EB compared to 24% with the naïve before-and-after method. This consistency indicates that regression-to-the-mean bias is minimal, as ELRs are typically not installed at high-crash locations. Therefore, naïve comparisons may be sufficient for preliminary assessments, though EB remains the more reliable method.
Overall, these findings suggest that ELRs can significantly improve roadway safety. Importantly, no head-on collisions were reported at any ELR site, countering common public concerns that two-way traffic within a single lane would increase such crashes. Nevertheless, past agency experience has shown the importance of public outreach and education to build support and ensure the long-term success of ELRs. This thesis represents the most comprehensive and up-to-date evaluation of ELR safety performance to date and provides guidance on key factors—including roadway volumes, lane widths, and public engagement—that should be considered when implementing ELRs in other jurisdictions.