TY  - JOUR
PY  - 2022//
TI  - Traffic accident prediction model for rural highways in Pernambuco
JO  - Case studies on transport policy
A1  - Macedo, Márcia R. O. B. C.
A1  - Maia, Maria L. A.
A1  - Kohlman Rabbani, Emília R.
A1  - Lima Neto, Oswaldo C. C.
A1  - Andrade, Maurício
SP  - 278
EP  - 286
VL  - 10
IS  - 1
N2  - Due to the need to update the current guidelines for highway design to focus on safety, this study sought to build an accident prediction model using a Geographic Information System (GIS) for single-lane rural highways, with a minimum of statistically significant variables, adequate to the Brazilian reality, and improve accident prediction for places with similar characteristics. This analysis was conducted on 215 km of single-lane road segments of highway BR-232 in the State of Pernambuco. The development of a database made it possible to associate accident records for the period 2007 to 2016 from Federal Highway Police (PRF) data with the geometric parameters of the highway, obtained through geometric reconstruction of the vector data available at the National Department of Transportation Infrastructure (DNIT) and the semi-automatic extraction of highways from satellite imagery. The homogeneous segments were analyzed and classified by the Spatial method (Kernel-KDE density). A Generalized Estimating Equation (GEE) model was estimated to model the frequency and severity of accidents. The results indicate that increase in the slope and the radius impact the increase in the frequency of accidents and the reduction of the severity of accidents in curves.<p />  <p>Language: en</p>
LA  - en
SN  - 2213-624X
UR  - http://dx.doi.org/10.1016/j.cstp.2021.12.009
ID  - ref1
ER  -