Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
SAFETY EVALUATION OF PERMANENT
RAISED PAVEMENT MARKERS
SUMMARY S.1 BACKGROUND AND OBJECTIVE
Permanent raised pavement markers (PRPMs) are delineation devices that are often
used to improve preview distances and guidance for drivers in inclement weather and
low-light conditions. Recent studies in New York, Texas, and Pennsylvania have raised
concerns about the safety effects of PRPMs after potential negative side effects were
reported. These studies pertained to single jurisdictions only, and their results were
questioned because of some identified data and methodological difficulties. NCHRP
Project 5-17 responds to the need to use state-of-the-art analytical methods and exten-
sive data to comprehensively assess the safety effects of PRPMs and to identify criti-
cal design parameters. The primary objectives of NCHRP Project 5-17, as presented in
the project statement, are "to assess the safety effects of permanent raised pavement
markers (PRPMs) and to develop guidelines for their use."
An empirical Bayesian before-and-after safety evaluation methodology was selected
to address methodological issues and challenges associated with previous efforts under-
taken to evaluate PRPMs. The methodological issues and challenges (e.g., regression
to the mean and traffic volume changes) were identified during a comprehensive liter-
ature review of previous PRPM evaluation studies. In total, 29 states were surveyed by
iTRANS and assessed for possible inclusion in the study. Six states were selected on
the basis of their ability to provide the necessary crash, traffic volume, roadway attribute,
and PRPM installation data required to perform the evaluation study. This study gath-
ered data to evaluate the effects of snowplowable PRPMs on nonintersection crashes
on a representative sample of two-lane roadways, four-lane expressways, and four-lane
freeways in the states of Illinois, Missouri, Pennsylvania, New York, Wisconsin, and
New Jersey. Because of data-intrinsic constraints that were proved impossible to over-
come despite all efforts by the research team, it was not viable to perform a sound safety
assessment of the effect of PRPMs on four-lane expressways.
S.2 STUDY METHODOLOGY
Safety performance functions (or crash prediction models) were developed for var-
ious crash types: total, fatal and injury, nighttime, nighttime fatal and injury, daytime,
OCR for page 2
2
daytime fatal and injury, wet weather, dry weather, and guidance-related. These safety
performance functions (SPFs) served as a statistical tool to determine the overall effec-
tiveness of PRPMs for particular crash types at the PRPM treatment locations.
Further disaggregate analysis, using regression techniques, investigated the rela-
tionship between the effect of PRPMs on nighttime crashes and various roadway, traf-
fic, and PRPM design factors. The purpose of this disaggregate analysis was to deter-
mine some of the specific conditions under which PRPMs are effective in reducing
crashes or vice versa.
S.3 DISCUSSION OF RESULTS
The analysis showed that the nonselective implementation of PRPMs on two-lane
roadways, overall, does not significantly reduce total or nighttime crashes, nor does it
significantly increase these crash types. On the other hand, for those locations where
PRPMs were implemented on the basis of selective policies (i.e., poor crash history,
among other criteria), the analyses produced mixed results. Positive effects were found
in New York for total and nighttime crashes where PRPMs were installed at locations
selected on the basis of the wet weather nighttime crash history. Similar safety effects
were not found in Pennsylvania, where PRPMs were implemented at locations selected
on the basis of total nighttime crash history. The analysis results have also revealed that
selective implementation of PRPMs requires a careful consideration of traffic volumes
and roadway geometry (degree of curvature). At low volumes (where the annual average
daily traffic [AADT] is less than 5,000 vehicles per day [veh/day]), PRPMs can in fact
be associated with a negative effect, which is magnified by the presence of sharp curva-
ture. For example, for PRPMs installed on roadways with AADTs ranging between 5,000
and 15,000 veh/day and with a degree of curvature greater than 3.5, an increase of
nighttime crashes of 26 percent can be estimated from the model.
Overall, the installation of PRPMs at noninterchange locations on four-lane freeways
showed neither a positive nor a negative overall safety effect on total and nighttime
crashes. However, some significant reductions were recorded for wet weather crashes
at locations on four-lane freeways, and there are indications that PRPMs are only effec-
tive in reducing nighttime crashes where the AADT exceeds 20,000 veh/day.
The results from the disaggregate analyses were used to develop guidelines for the
use of snowplowable PRPMs for two-lane roadways and four-lane freeways. The
guidelines are based on a two-step procedure. First, the expected safety benefit after the
installation of PRPMs is determined in relation to the expected reduction in future
nighttime crashes. Second, a positive expected safety effect is followed by an analyti-
cal engineering procedure relying on safety performance functions or crash prediction
models for roads with and without PRPMs to determine the potential cost-effectiveness
of implementing PRPMs at a location. The guidelines were discussed in the context of
the present "Manual on Uniform Traffic Control Devices" (MUTCD), and modifica-
tions are proposed for future editions.
