Details

Autor(en) / Beteiligte

• Yang, Hong
• Wang, Zhenyu
• Xie, Kun
• Ozbay, Kaan
• Imprialou, Marianna

Titel

Methodological evolution and frontiers of identifying, modeling and preventing secondary crashes on highways

Ist Teil von

• Accident analysis and prevention, 2018-08, Vol.117, p.40-54

Ort / Verlag

England: Elsevier Ltd

Erscheinungsjahr

2018

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• •Investigated state-of-the-art approaches for secondary crash (SC) identification.•Thoroughly examined the practices for modeling and predicting SC risk.•Explored current research efforts being conducted to prevent secondary crashes.•Synthesized the issues, needs and challenges associated with SC studies.•Discussed potential opportunities and perspectives to foster future SC research. Secondary crashes (SCs) or crashes that occur within the boundaries of the impact area of prior, primary crashes are one of the incident types that frequently affect highway traffic operations and safety. Existing studies have made great efforts to explore the underlying mechanisms of SCs and relevant methodologies have been evolving over the last two decades concerning the identification, modeling, and prevention of these crashes. So far there is a lack of a detailed examination on the progress, lessons, and potential opportunities regarding existing achievements in SC-related studies. This paper provides a comprehensive investigation of the state-of-the-art approaches; examines their strengths and weaknesses; and provides guidance in exploiting new directions in SC-related research. It aims to support researchers and practitioners in understanding well-established approaches so as to further explore the frontiers. Published studies focused on SCs since 1997 have been identified, reviewed, and summarized. Key issues concentrated on the following aspects are discussed: (i) static/dynamic approaches to identify SCs; (ii) parametric/non-parametric models to analyze SC risk, and (iii) deployable countermeasures to prevent SCs. Based on the examined issues, needs, and challenges, this paper further provides insights into potential opportunities such as: (a) fusing data from multiple sources for SC identification, (b) using advanced learning algorithms for real-time SC analysis, and (c) deploying connected vehicles for SC prevention in future research. This paper contributes to the research community by providing a one-stop reference for research on secondary crashes.