Enhancing Nighttime

1
Enhancing Nighttime
Washington County Highway Marking Test
Pavement Marking Visibilityfor Older Drivers

• Dr. Thomas Schnell
• Phillip J. Ohme
• Operator Performance Lab, Human Factors
• Department of Industrial Engineering
• The University of Iowa

2
Contact Information

• Principal Investigator
• Thomas Schnell, Ph.D.
• thomas-schnell_at_uiowa.edu
• http//arrow.win.ecn.uiowa.edu
• Researcher
• Phillip J. Ohme, M.S.
• phil-ohme_at_uiowa.edu
• 4231 Seamans Center for the Engineering Arts and
  SciencesIowa City, IA 52242
• Tel [phone]

3
Introduction Factors addressed

• Retroreflectance consistency as it pertains to
  the pavement marking application process
• Subjective evaluations of the test site
  treatments adequacy and safety
• Expert evaluations
• Driver participant evaluations
• Preview (detection) distance of various
  delineation treatments
• Amount of wear (time from application)
• Immediately after initial application
• One year after application (worn)
• Width of edge lines
• 100 mm (4-inch), conventional markings
• 150 mm (6-inch), wide markings
• 200 mm (8-inch), extra-wide markings
• Type of optics used in edge lines
• Paintbeads
• Ceramic element
• Wet-weather tape
• Presence/absence of supplemental retroreflectors
• Post delineators on right shoulder of road
• RPMs on centerline

4
Purpose of Pavement Markings(figure reproduced
from Zwahlen and Schnell, 1999)

• Unique traffic control device
• Pavement markings continuously convey information
• Drivers do not need to shift attention away from
  the road to gain guidance information
• Pavement marking limitations
• Wet-weather visibility
• Deterioration

5
Retroreflection
6
Diffuse Reflection
7
Specular Reflection
8
Fatal Crashes in the US (USDOT/NHTSA, 1996)

• More miles are driven during the day, but death
  rates based on miles driven are about 4 X higher
  at night than during the day

9
Evaluation of Pavement Marking Application Process
10
Dry and Wet Retroreflectance Measurements

• Recovery (bucket) method used for wet
  measurements
• Test standard adopted by CEN (1991)
• 10 liters of water poured evenly on the sample
• Height of 0.5 meters
• Water drains (recovers) for 60 seconds
• Total of 343 km (213 miles) of roadway sampled

11
Washington County Roads21819222
12
Jackson and Clinton County Roads52626467428

13
Dubuque County Road52
14
Initial Measurements White Edge Lines

• 89.7 decrease in median retroreflectance from
  dry to wet for white paintbeads

15
Initial Measurements Yellow Centerline

• 86.5 decrease in retroreflectance from dry to
  wet for yellow paintbeads
• 38.5 decrease from white to yellow under dry
  roadway conditions. (19.4 decrease under wet
  conditions)

16
New Pavement Markings with/without Supplemental
Retroreflectors (1999 Data)

• Dependent variable Detection Distance
• Between-subject variables
• 14 Participants
• Age at 2 levels 8 Young (23.2 years average), 6
  Older (67.8 years average)
• Order at 2 levels StartA, StartB (opposite
  directions)
• Weather (12 participants in Dry, 2 in Wet (both
  older))
• Within-subject variables
• Pavement marking treatment at 5 levels
• 100 mm (4-inch) edge lines, paintbeads
• 150 mm (6-inch) edge lines, paintbeads
• 200 mm (8-inch) edge lines, paintbeads
• Ceramic Element (100 mm width)
• Wet-Weather Tape (100 mm width)
• Raised Pavement Markers (RPMs), with/without
• Post Delineators, with/without
• 2 Replications

17
Washington Co. Test Site
18
Layout of Test Site
19
Raised Pavement Markers (RPMs)

• Normally spaced at 80 feet
• Research has shown that 120 feet spacing can be
  used on straight roads with no decrement in
  visibility
• Cost Savings

20
RPMs along Centerline
21
Post Delineators along Right Shoulder
22
RPMs and Post Delineators
23
Detection Distance Task
24
An Approaching Gap
25
Results (1999) Pavement Marking Treatments

• Weather and age effects for young and older
  participants under dry and wet roadway conditions
  (N1 16 observations, 8 participants for
  dry-young N2 8 observations, 4 participants
  for dry-older, N3 4 observations, 2
  participants for wet-older all at 2
  replications, error bars show standard deviations)

26
Results (1999) All Supplemental Retroreflectors

• Comparing detection distances with and without
  the use of retroreflectors, and with the use of
  different pavement marking material (N1 48
  observations, 8 participants for dry-young N2
  24 observations, 4 participants for dry-older, N3
  12 observations, 2 participants for wet-older
  all at 2 replications in 3 sections, error bars
  show standard deviations)

27
Statistically Comparing Width of Edge Lines (1999)

• Width has no statistically significant effect at
  alpha 0.05

28
Expert Evaluations

• Eleven experts (traffic engineers, law
  enforcement officials, and Iowa DOT employees)
  evaluated the test site and were asked various
  questions
• Against Post Delineator usage
• Supportive of RPMs

Expert evaluators' response to the question
"Should RPMs be considered for use in Iowa?"
29
Worn Pavement Markings under Dry and Wet
Conditions (2000)

• Subset of original Washington County test site
• Measured detection distance for various pavement
  marking treatments
• Gives more realistic picture of nighttime roadway
  delineation, since pavement markings were worn
  for one year

30
Experimental Design

• Dependent variable Detection Distance
• Between-subject variables
• 14 Participants
• Age at 2 levels 7 Young (22.9 years average), 7
  Older (71.3 years average)
• Order at 2 levels StartA, StartB (opposite ends
  of test site)
• Within-subject variables
• Weather at 2 levels Dry, Wet
• Pavement marking treatment at 5 levels
• 100 mm (4-inch) edge lines, paintbeads
• 150 mm (6-inch) edge lines, paintbeads
• 200 mm (8-inch) edge lines, paintbeads
• Ceramic Element (100 mm width)
• Wet-Weather Tape (100 mm width)
• Retroreflectance of edge lines
• 4 Replications

31
Test Site
32
Procedure for Wet Measurements
33
Results (2000) Percentage Increases
34
Results (2000) Age, Weather, Pavement Marking
Treatment Effect on Detection Distance

• Worn detection distances for young and older
  participants under dry and wet roadway
  conditions. (N 28 measurements, 7 participants
  error bars show standard deviations)

35
Statistical Analysis

• Dry model showed no significant variables or
  interactions
• Wet model showed showed significant differences
  in ceramic and wet-weather tape comparisons
  (shown at right)
• Edge line width has no statistically significant
  effect

36
Examples of Preview Times Needed (Worn
PaintBeads)
37
Examples of Preview Times Needed (Worn
Wet-Weather Tape)
38
Comparing New and Worn Pavement Markings

• 32.8 decrease from initial to worn for the dry
  condition on average.

39
PaintBeads Section, Wet
40
Ceramic Element Section, Wet
41
Wet-Weather Tape Section, Wet
42
Regression Model (All New Worn Data)

• Dependant variable Detection Distance
• Predictors
• Retroreflectance
• Continuous in mcd/m2/lx, range of 6 942
  mcd/m2/lx
• Width (of edge lines)
• 4, 6, 8 in inches
• Optics
• 1, 2, or 3 1paintbeads, 2ceramic element,
  3tape
• Weather
• 0 or 1 0dry roadway, 1wet roadway
• Time
• 0 or 1 in years0initial, 1worn by one year
• Gender
• 0 or 1 0male, 1female
• Age
• 0 or 1 0young, 1older

43
Final Regression Model

• Ln(Detection Distance)
• 1.52 0.457 Ln(retroreflectance)
• 0.0209 width 0.113 optics
• - 0.545 weather
• Adjusted R2 0.668 or 66.8
• Model is a simple tool that can predict detection
  distances
• Can be used to find minimum retroreflectances
  needed for adequate preview

44
Minimum Retroreflectance Values mcd/m2/lx
Recommended for 100 mm (4-inch) PaintBead Edge
Lines
45
Recommendations

• Increase the Retroreflectance
• Intensify pavement marking application schedule
• Paint roadways 2 TIMES often
• Many inherent problems will remain
• Improve paintbead pavement marking application
  process
• Further quality control research needed
• Possible to get 450 mcd/m2/lx, just need
  consistency
• Apply advanced pavement marking materials in
  high-risk areas (e.g. work zones, extreme curves)
• Ceramic element or wet-weather tape
• Implement RPM use on all state highways
• Done in Ohio
• Pavement marking retroreflectance is no longer a
  primary concern

46
Ongoing Research at the OPL

• Driver Roadway Looking Behavior and Workload as a
  Function of Pavement Marking Visibility
• Limited preview conditions
• Nationwide Survey of Pavement Marking Delineation
  Practices
• See if other states have had good/bad experiences
  with wider markings, RPMs
• Analysis of crash data for Washington County and
  Jackson County
• Standard vs. wider-than-standard pavement
  markings