TY - JOUR
T1 - A framework for estimating the safety effects of roadway lighting at intersections
AU - Donnell, Eric T.
AU - Porter, Richard J.
AU - Shankar, Venkataraman N.
PY - 2010/12
Y1 - 2010/12
N2 - National- and state-level guidance documents conclusively state that fixed lighting improves intersection safety. The sentiment is consistent with other design and safety manuals and is supported by a series of consistent safety findings; however, most published lighting-safety research is focused on rural, stop-controlled intersections and is limited by several methodological issues. The relationship between safety and intersection lighting at rural, signalized and urban locations is not as well documented. Methodological advancements in highway safety analysis justify new estimations of the safety effects of intersection lighting. This paper describes a proposed framework to estimate the safety effects of fixed lighting at a variety of intersection types and locations. Several key issues are explored including availability of relevant crash, lighting, and roadway inventory data; relevant data element structures; proposed analysis taxonomies to assess lighting-safety effects within and across different intersection classifications; specification and estimation of models to estimate expected crash frequencies during day and night; techniques to interpret model parameters, including variable elasticity; and tests of model transferability across states. A sample framework execution using Minnesota intersection data is provided. Results indicate a much lower overall safety benefit from lighting than published studies, but are consistent with estimates included in Highway Safety Manual research.
AB - National- and state-level guidance documents conclusively state that fixed lighting improves intersection safety. The sentiment is consistent with other design and safety manuals and is supported by a series of consistent safety findings; however, most published lighting-safety research is focused on rural, stop-controlled intersections and is limited by several methodological issues. The relationship between safety and intersection lighting at rural, signalized and urban locations is not as well documented. Methodological advancements in highway safety analysis justify new estimations of the safety effects of intersection lighting. This paper describes a proposed framework to estimate the safety effects of fixed lighting at a variety of intersection types and locations. Several key issues are explored including availability of relevant crash, lighting, and roadway inventory data; relevant data element structures; proposed analysis taxonomies to assess lighting-safety effects within and across different intersection classifications; specification and estimation of models to estimate expected crash frequencies during day and night; techniques to interpret model parameters, including variable elasticity; and tests of model transferability across states. A sample framework execution using Minnesota intersection data is provided. Results indicate a much lower overall safety benefit from lighting than published studies, but are consistent with estimates included in Highway Safety Manual research.
KW - Intersections
KW - Lighting
KW - Night-to-day crash ratios
KW - Statistical analysis
KW - Traffic safety
UR - http://www.scopus.com/inward/record.url?scp=77956227854&partnerID=8YFLogxK
U2 - 10.1016/j.ssci.2010.06.008
DO - 10.1016/j.ssci.2010.06.008
M3 - Article
AN - SCOPUS:77956227854
VL - 48
SP - 1436
EP - 1444
JO - Safety Science
JF - Safety Science
SN - 0925-7535
IS - 10
ER -