Please use this identifier to cite or link to this item:

TitlePortuguese two-lane highways: modelling crash frequencies for different temporal and spatial aggregation of crash data
Author(s)Costa, Jocilene Otília
Jacques, M. A. P.
Pereira, Paulo A. A.
Freitas, Elisabete F.
Soares, Francisco Emanuel Cunha
KeywordsCrash contributory factors
Crash prediction models
Generalized estimating equations
Longitudinal data
Two-lane highways
Issue date2018
PublisherTaylor & Francis
CitationCosta J. O., Jacques M. A. P., Pereira P. A. A., Freitas E. F., Francisco S. PORTUGUESE TWO-LANE HIGHWAYS: MODELING CRASH FREQUENCIES FOR DIFFERENT TEMPORAL AND SPATIAL AGGREGATION OF CRASH DATA, Transport, doi:10.3846/16484142.2015.1073619, 2018.
Abstract(s)The identification of contributory factors to crash frequencies observed in different highway facilities can aid transportation and traffic management agencies to improve road traffic safety. In spite of the strategic importance of the national Portuguese road network, there are no recent studies concerned with either the identification of contributory factors to road crashes or crash prediction models (CPMs) for this type of roadway. This study presents an initial contribution to this problem by focusing on the national roads NR-14, NR-101 and NR-206, which are located in Portugalâ s northern region. They are two-lane single carriageway rural roads. This study analyzed the crash frequencies, average annual daily traffic (AADT) and geometric characteristics of 88 two-lane road segments. The selected segments were 200-m-long and did not cross through urbanized areas. The fixed length of 200 meters corresponds to the road length used in Portugal to define a critical point. Data regarding the annual crash frequency and the AADT were available from 1999 to 2010. Due to the high number of zero-crash records in the initial database, the data were explored to identify the best statistical modeling approach to be adopted. The GEE procedure was applied to 10 distinctive databases formed by grouping the original data in time and space. The results show that the different observations within each road segment present an exchangeable correlation structure type. This paper also analyses the impact of the sample size on the modelâ s capability of identifying the contributing factors to crash frequencies. The major contributing factors identified for the two-lane highways studied were the traffic volume (expressed in AADT), lane width, vertical sinuosity, and density of access points. Acceptable CPM was identified for the highways considered, which estimated the total number of crashes for 400-m-long segments for a cumulative period of two years.
AccessRestricted access (UMinho)
Appears in Collections:C-TAC - Artigos em Revistas Internacionais

Files in This Item:
File Description SizeFormat 
2076-Manuscript No names_Revisionx.pdf
  Restricted access
302,64 kBAdobe PDFView/Open

Partilhe no FacebookPartilhe no TwitterPartilhe no DeliciousPartilhe no LinkedInPartilhe no DiggAdicionar ao Google BookmarksPartilhe no MySpacePartilhe no Orkut
Exporte no formato BibTex mendeley Exporte no formato Endnote Adicione ao seu ORCID