Bootstrapping DEA Scores for Road Safety Strategic Analysis in Brazil

• Main Authors: Jorge Tiago Bastos, Yongjun Shen, Elke Hermans, Tom Brijs, Geert Wets, Antonio Clóvis Pinto Ferraz
• Format: Article
• Language: English
• Published: Atlantis Press 2015-12-01
• Series: International Journal of Computational Intelligence Systems
• Subjects: Road safety; Bootstrapping; Data Envelopment Analysis; Clustering; Brazilian states
• Online Access: https://www.atlantis-press.com/article/25868662.pdf
• ISSN: 1875-6883

In this paper, three risk indicators on road safety are combined into a composite indicator in order to assess the overall fatality risk for the 27 Brazilian states using the so-called multiple layer data envelopment analysis model. The states are first clustered and next, a range of bootstrapped scores is generated to manifest the estimated variability in the road safety performance. Bootstrapping the original DEA scores showed to be a useful strategy to assess the robustness of the states’ ranking.