Open AccessAnalysis of safety impact of paved shoulder width on Czech secondary roads
Author(s) -
Jiří Ambros,
Zuzana Křivánková,
Robert Zůvala,
Kateřina Bucsuházy,
Jindřich Frič
Publication year - 2021
Publication title -
archives of transport
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.309
H-Index - 14
eISSN - 2300-8830
pISSN - 0866-9546
DOI - 10.5604/01.3001.0015.6148
Subject(s) - shoulders , crash , multinomial logistic regression , czech , sample (material) , motor vehicle crash , computer science , environmental science , transport engineering , poison control , statistics , engineering , mathematics , human factors and ergonomics , medicine , environmental health , physics , linguistics , philosophy , surgery , thermodynamics , programming language
Traffic safety is influenced, among other factors, by characteristics of the roads, which include the width of the shoulder. Shoulder width was noted to have a large effect on crash frequency, as well as on traffic speed. In this paper, we focused on paved shoulders. Previous studies confirmed that increasing the width of the paved shoulder is associated with a decrease in crash frequency. However, wider shoulders may encourage higher driving speed, which is related to an increase of impact speed and crash severity – this issue was hypothesized, but not statistically investigated. Thus, conclusions based on crashes and speeds contradict each other, and there is no simple answer to the question of the safety impact of wide shoulders. To address this gap, we analyzed a sample of two most typical categories of Czech secondary roads, which differ only in the paved shoulder width (S9.5 roads with 0.75m-wide shoulder, and S11.5 roads with 1.75m-wide shoulder) and thus present a suitable example for studying the safety impact of paved shoulder width. We used generalized linear models of crash frequency, and multinomial logistic models of crash severity (separately for single-vehicle and multi-vehicle crashes), as well as a statistical test of differences in speed for the two road categories.The results showed that: Firstly, there were fewer crashes on S11.5 roads compared to S9.5 roads; this was true for both single-vehicle and multi-vehicle crashes. Secondly, single-vehicle crashes on S11.5 roads were more severe compared to S9.5 roads; the change of severity in multi-vehicle crashes was not statistically significant. Thirdly, driving speeds on S11.5 roads were approx. by 7 km/h higher compared to S9.5 roads. These findings support the hypothesis of an association between wider shoulders, higher speeds, and increased crash severity, especially in the case of single-vehicle crashes. As a practical solution, various speed management measures, including widening to a 2+1 road, may be recommended.