Self-explaining roadways bringing safety up to speed

Over the past decade, speeding has contributed to about one-third of the more than 11,000 traffic fatalities in Illinois.

Seeking to increase traffic safety are the Illinois Center for Transportation and IDOT in a joint project, “Advancing Self-Explaining Streets Phase 1: The Relationship between Roadway Environment and Crash Severity". Stephane Seck-Birhame, IDOT bureau chief of Safety Programs and Engineering, represented the department on the project, which was led by Alireza Talebpour, University of Illinois Urbana-Champaign professor with help from Victoria Barrett, Chicago Metropolitan Agency for Planning senior transportation planner.

The three aimed to improve understanding of self-explaining roads, which use visual cues and physical layouts to control speeds, as well as identify opportunities and barriers to their use in Illinois.

“Just one fatality is unacceptable, but the fact remains the number of them continues to be stubbornly high in Illinois and across the nation, as people drive at higher speeds, more recklessly and more distracted,” Seck-Birhame said. “We are committed to the ambitious goal of driving the number down to zero through a holistic, safe-system approach that takes into account the need to build safer roads. The work that’s happening as part of this study is an innovative approach to explore together with our safety partners, with the potential to apply design elements to help reduce speeding.”

Features of self-explaining roadways include speed feedback signs, narrower or reduced number of travel lanes, speed humps, and curves or shifts in the roadway, among others.

“Rather than design to allow the highest possible speeds safely, self-explaining roadways use geometry and design features to constrain speeds to what is believed to be safe for the context,” Barrett said.

The researchers examined national and international studies on self-explaining streets and conducted a preliminary analysis of the impacts of roadway elements on crash severity in Illinois.

The team examined data from more than 1.15 million crashes in Illinois between 2019 and 2022 and used Google Street View to obtain geospatial data from those sites, with the goal of developing a methodology for analyzing relationships between roadway geometry and features and crash outcomes.

They applied image segmentation — an AI technique that trains computers to interpret images by classifying pixels into specific categories — to determine trends in crash severity and contributing factors.

From the segmentation results, they developed models that predict probabilities based on multiple possible outcomes. The models found a relationship between crash severity and road and vegetation characteristics.

“Higher percentages of visible road surface were strongly linked to more severe crashes, with fatal crashes becoming more likely as road surface increased,” Talebpour said. “Severe crashes were more common in darker conditions and during weekends.”

The next steps involve studying additional landscaping and roadway elements and their influence on speeding as well as identifying cost-effective ways to modify designs.

IDOT aims to use the findings to develop preliminary guidance for self-explaining roadways, where appropriate.

“Safety is IDOT’s highest priority,” Seck-Birhame said. “The general public will benefit from road designs that nudge them in obeying traffic laws, notably speed limits. IDOT will benefit from another design tool in our effort against speeding.”