An integrated GIS–MCDM and statistical analysis approach for assessing heavy vehicle accident outcomes
International Journal of Crashworthiness, vol.31, no.4, pp.376-411, 2026 (SCI-Expanded, Scopus)
- Publication Type: Article / Article
- Volume: 31 Issue: 4
- Publication Date: 2026
- Doi Number: 10.1080/13588265.2026.2679750
- Journal Name: International Journal of Crashworthiness
- Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Applied Science & Technology Source, Compendex, INSPEC, Natural Science Collection (ProQuest), Biological Science Database (ProQuest), Engineering Source (EBSCO), Materials Science & Engineering Collection (ProQuest), Technology Collection (ProQuest)
- Page Numbers: pp.376-411
- Keywords: Spatial analysis, heavy vehicle accidents, binary logistic regression, risk assessment, traffic safety
- Ataturk University Affiliated: Yes
Abstract
Analysis of heavy vehicle accidents is critical for traffic safety, as such crashes often result in serious injuries, fatalities, material damage, traffic disruptions, and economic losses. This study proposes an integrated approach combining geographic information systems (GIS), multi-criteria decision-making (MCDM), and statistical analysis to examine heavy vehicle accidents in Türkiye between 2015 and 2019. The methodology includes: (i) examining accident records and identifying 21 independent variables and 93 sub-variables, (ii) calculating their priority values using the fuzzy Logarithm Methodology of Additive Weights, (iii) conducting GIS-based spatial analyses, and (iv) applying binary logistic regression to evaluate accident outcomes as injury or fatality. The results indicate that road, environmental, accident-related, and driver-related factors significantly affect crash severity. Fuzzy LMAW results show that driver violations have the greatest negative impact. Spatial analyses reveal that severe crashes are mainly concentrated in western, north-western, south-western, and central Türkiye.