CONTACT US: Contact info
|
Citation |
Goyani J, Arkatkar S, Joshi G, Easa S. Transp. Res. F Traffic Psychol. Behav. 2024; 102: 33-53. |
|
Copyright |
(Copyright © 2024, Elsevier Publishing) |
|
DOI |
|
PMID |
unavailable |
|
Abstract |
Driver risk perception ability depends on how they perceive and interpret information from different sources to safely and comfortably operate their vehicles based on the highway features ahead. The present study is divided into two parts to review these aspects: (a) identifying the factors that influence driver risk-taking behavior, and (b) evaluating the soundness of the identified factors by comparing them with quantitative risk perception. For that, 1075 user perception data were collected using a developed questionnaire form through face-to-face interviews. A questionnaire form consists of a total of 33 questions/variables. After applying the factor reduction techniques, namely principal component analysis, the most significant variables (19 questions/variables) influencing the user's perception are identified. Then, considering these significant variables, five separate risk perception models were developed using the Structural Equation Modeling technique for five varying curve geometries. The results revealed that the user's risk perception changes for the same user for varying curve geometry. It is also identified that male drivers have more risky behavior (lower risk perception) than female drivers. Besides, Garrett's ranking technique is also adopted to rank the user risk perception for the varying curve geometry to identify the dangerous curves by assigning the rank. Lastly, to check the feasibility of the results, the risk perception score is also compared with the quantitative risk score. Based on the study results, it is concluded that the study findings will be helpful for highway planners and authorities to introduce some safety measures to alert users of the potential hazards on the roadside and increase their capacity for taking risks on complex curves; hence, the likelihood of crashes can be minimized. Language: en |
|
Keywords |
Garrett ranking; Quantitative risk; Road crash data; Structural equation modeling; User’s perception; Varying curve geometry |