Estimation of the Influence of Artificial Roadway Lighting on Road Collision Frequency

• Main Author: Matout, Nagham
• Format: Others
• Published: 2013
• Online Access: http://spectrum.library.concordia.ca/978051/1/Matout_MASc_S2014.pdf; Matout, Nagham <http://spectrum.library.concordia.ca/view/creators/Matout=3ANagham=3A=3A.html> (2013) Estimation of the Influence of Artificial Roadway Lighting on Road Collision Frequency. Masters thesis, Concordia University.

Road accidents in Quebec registered a total of 104,070 collisions with 39,541 victims of injuries and 436 fatalities in 2012 alone. Driving in dark environments increases the risk of accident likelihood for which artificial roadway lighting is typically seen as a countermeasure. However, it is unknown if non-standard levels of lighting help in reducing collision frequency, representing the case for many inconsistently illuminated roads under municipal jurisdiction. This research collected illuminance measurements for the Arthabasca region in Quebec. The collected data was combined with available operational and geometrical characteristics as well as collision frequency, to investigate what variables explain nighttime road crashes and how different levels of artificial lighting correlate with them. It was found that the presence of an intersection and having a slippery road surface produced more collisions. Roads with a complex geometry as well as traffic volume explain higher collision rates. Either standard or non-standard illuminated roads resulted in an increase of road collision frequency as compared to dark sites. Definition of standard illumination seems not to correspond to the statistical evidence herein found. Increasing the minimum level of illuminance for standard lighting helps in reducing collision frequency at standard lit sites as illuminance levels were raised. Quebec warrant grid system seems to give preference to illuminate roads at either urban locations or in the proximity to an intersection. A good correlation between all illuminated sites and a variable containing urban and suburban land uses was found. Empirical evidence also suggests that dark locations correspond mostly to rural sites (possibly with lower volumes of cars) which observe lower frequency of road collisions.