Estimating the safety performance of urban intersections in Lisbon, Portugal

被引:29
作者
Gomes, Sandra Vieira [1 ]
Geedipally, Srinivas Reddy [2 ]
Lord, Dominique [3 ]
机构
[1] Natl Lab Civil Engn, Dept Transportat, P-1700066 Lisbon, Portugal
[2] Texas Transportat Inst, College Stn, TX 77843 USA
[3] Texas A&M Univ, Zachry Dept Civil Engn, College Stn, TX 77843 USA
关键词
Crash prediction models; Generalized linear modeling; Road accidents; Urban intersections; Lisbon; MOTOR-VEHICLE CRASHES; POISSON-GAMMA; MODELS; ACCIDENTS;
D O I
10.1016/j.ssci.2012.03.022
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
According to official statistics, a large percentage of crashes in Portugal are reported on urban roads. For instance, from 2004 to 2007, about 70% of all injury accidents and 43% of the fatalities occurred inside urban agglomerations. This important safety problem has also been observed on the urban network of Lisbon. Understanding this significant problem, the Government of the Portuguese Republic via its research grant agency - The Foundation for Science and Technology - funded a project whose primary objective consists of developing tools that would help estimating the safety performance of various components of the urban highway system in Lisbon. This paper documents one component of the safety tools that were developed and describes the steps that were taken to develop predictive models for estimating the safety performance of signalized and unsignalized intersections of Lisbon. Several crash predictive models were developed using the Poisson-gamma modeling framework. Two types of models were estimated: flow-only and models with covariates. They were estimated using crash and other related data collected at 44 three-legged and 50 four-legged intersections for the years 2004-2007, inclusively. It was found that some highway geometric design characteristics were associated with the crashes occurring at urban three- and four-legged intersections in Lisbon. (C) 2012 Elsevier Ltd. All rights reserved.
引用
收藏
页码:1732 / 1739
页数:8
相关论文
共 24 条
[1]   NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION [J].
AKAIKE, H .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :716-723
[2]  
[Anonymous], 1998, ECONOMETRIC SOC MONO
[3]  
ANSR, 2007, AN 2007 SIN ROD
[4]  
BAUER KM, 2000, FHWARD99094 US DEP T
[5]   MODELS FOR PREDICTING ACCIDENTS AT JUNCTIONS WHERE PEDESTRIANS AND CYCLISTS ARE INVOLVED - HOW WELL DO THEY FIT [J].
BRUDE, U ;
LARSSON, J .
ACCIDENT ANALYSIS AND PREVENTION, 1993, 25 (05) :499-509
[6]  
DGV, 2005, DIR GER VIAC SIN ROD
[7]  
DGV, 2004, DIR GER VIAC SIN ROD
[8]  
DGV, 2006, DIR GER VIAC SIN ROD
[9]   Accident prediction models for urban roads [J].
Greibe, P .
ACCIDENT ANALYSIS AND PREVENTION, 2003, 35 (02) :273-285
[10]  
Hall R. D, 1986, 65 TRANSP ROAD RES L