Investigation of pedestrian crashes on two-way two-lane rural roads in Ethiopia

Tulu, Getu Segni; Washington, Simon; Haque, Md. Mazharul; King, Mark · 2015 · OpenAlex-citations

DOI: 10.1016/j.aap.2015.02.011

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the critical gap in pedestrian safety research within developing countries, specifically focusing on two-way, two-lane rural roads in Ethiopia. Pedestrian fatalities constitute 55% of all road traffic deaths in Ethiopia, yet existing prediction models are largely derived from Western contexts and often lack essential exposure data, such as pedestrian volumes. The research aims to develop accurate pedestrian crash prediction models by quantifying the relationships between crash frequency and variables including traffic flow, pedestrian exposure, road geometry, and site characteristics. The methodology employed a Random Parameter Negative Binomial (RPNB) model to analyze longitudinal data collected from 18 road segments within a 100 km radius of Addis Ababa. These segments were selected from six major rural routes and characterized by the absence of sidewalks, street lighting, and major intersections. Data collection spanned a three-year period (2010–2012), resulting in 162 observations divided into three four-month seasonal intervals to account for variations in traffic and pedestrian flows. Crucially, the study incorporated primary field data for pedestrian crossing volumes, counted during specific daytime hours, alongside secondary data on Annual Average Daily Traffic (AADT), road geometry, and crash records verified by police officers. The results identified seven significant explanatory variables. The log of the product of AADT and daily pedestrian crossing volume was positively associated with crashes, indicating a nonlinear increase in risk with higher exposure. Lane width showed a negative association, with wider lanes reducing crash frequency, likely because narrow lanes force pedestrians into closer proximity with traffic. Mountainous terrain increased crash risk due to steep gradients and sharp curves, while gravel shoulders decreased risk by discouraging drivers from using the shoulder for overtaking. Demographic factors also played a role: a higher proportion of pedestrians under 19 years old and those crossing in groups were associated with fewer crashes, potentially due to quicker reflexes or greater visibility. Conversely, the proportion of pedestrians crossing in pairs exhibited random parameter effects, suggesting its impact varied significantly across different sites. The significance of this study lies in its provision of the first robust pedestrian crash prediction models for rural roads in a developing country context. By integrating specific pedestrian exposure data, the research offers transport planners and engineers tools to identify high-risk locations and evaluate infrastructure improvements. The findings highlight unique local factors, such as the safety benefits of gravel shoulders and the varying risks associated with pedestrian demographics, providing a foundation for targeted safety interventions and further site-level studies in similar environments.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• incidence prevalence
• vru crash typology
• demographic disparities
• comparative international
• pedestrian behavior perception
• urban rural setting

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: crash risk outcomes