Screening Needs for Roadway Lighting

1
Screening Needs for Roadway Lighting

• Prepared for Presentation
• to the Project Steering Committee
• March 26, 2003

2
Project Team
Steering Committee Carl Andersen, FHWA Travis
Bridewell, VDOT Richmond Dist. Pamela Brookes,
VDOT-TED Butch Cubbage, VDOT Richmond Dist. Mark
Hodges, VDOT-TED Karen Rusak, VDOT-TED Jon
Sayyar, VDOT-TED Al Smith, VDOT-TED Benjamin
Cottrell, VTRC Wayne Ferguson, VTRC Michael
Perfater, VTRC Jim Havard, IES

3
Project Team (cont.)
Technical Assistance Robert Rasmussen,
VDOT-TED Ralph Jones, VDOT-TED Robert Feldman,
VDOT Richmond Dist. Llewellyn Slayton, VDOT
Richmond Dist. Timothy Rawls, VDOT Hampton Roads
Dist. Ray Khoury, VDOT Hampton Roads
Dist. Jaroslaw Jastrzebski, VDOT Nothern Virginia
Dist. Jim Gillespie, VTRC

4
Project Team (cont.)
Center for Risk Management of Engineering
Systems Department of Systems and Information
Engineering, University of Virginia James
Lambert, Research Assistant Professor of Systems
Engineering and Center Associate
Director (lambert_at_virginia.edu, [phone],
fax [phone]) Yacov Haimes, Professor of Systems
Engineering and Civil Engineering and Center
Director Thomas Turley, M.S. Student Raynelle
Deans, B.S. Student Andrew Miller, B.S.
Student Jenny Murrill, B.S. Student James
Sanders, B.S. Student
5
Outline

• Background
• Methodology development
• Overview of the screening method
• Exposure assessment
• Site parameters assessment
• Data analysis to support the screening method
• Example
• Conclusions and recommendations

6
Background
7
Background

• Causes of higher nighttime crash rate
• Low light level affecting visual capabilities
• Increased alcohol usage
• Fatigue
• Over-representation of young drivers
• Nighttime fatality rate three times day fatality
  rate
• About 15,000 lives could have been saved if
  nighttime crash rate could have been reduced to
  daytime crash rate in 1993 only
• Source CIE 1992 report-Road lighting as an
  accident countermeasure

8
Background (cont.)

• Potential positive impacts of fixed roadway
  lighting
• Reduction of nighttime accident rate
• Increased safety for pedestrians
• Facilitation of traffic flow
• Inspiration of community growth
• Aid to police protection
• Promotion of business
• Source AASHTO 1984 - An informal guide to
  roadway lighting

9
Background (cont.)

• Potential negative impacts of fixed roadway
  lighting
• Light pollution
• Light reflected form roads causing sky to glow
• Aesthetic effect caused by clutter
• Energy waste
• Glare
• Source Environmental Effects of Roadway
  Lighting By Carl Shaflik,
• August 1997

10
Background (cont.)

• A warrant is a factual evidence that there is
  reason for a proposed project
• Meeting a warrant does not obligate the agency to
  undertake the project
• Proposed project should be considered further in
  view of what resources are available, the
  traffic, the severity of hazards, and other
  considerations
• Source USDOT/FHWA (1978)

11
Background (cont.)

• NCHRP warrants
• Weighted score method
• Burdensome level of detail
• Diminished relevance since 1970s
• Source NCHRP (1974)

12
Background (cont.)

• AASHTO warrant for complete interchange lighting
• CIL-1 Traffic entering and leaving the freeway
• ADT gt 10,000 urban conditions
• ADT gt 8,000 suburban conditions
• ADT gt 5,000 rural conditions
• CIL-2 Traffic on crossroad
• ADT gt 10,000 urban conditions
• ADT gt 8,000 suburban conditions
• ADT gt 5,000 rural conditions
• CIL-3 When existing substantial commercial or
  industrial development, which are lighted at
  night, is located in immediate vicinity of the
  interchange
• Where the crossroad approach legs are lighted
  for ½ mile or more on each side of the
  interchange
• CIL-4 Ratio of night to day accident rate is 1.5
  or higher than the statewide average for
  unlighted similar section and studies show a
  significant reduction in nighttime accident when
  lighting is introduced
• 
  Source AASHTO 1984 - An Informational
  Guide for Roadway Lighting

13
Background (cont.)

• General understanding of the issues of fixed
  roadway lighting are studied in Wilken et al.
  (2001), Kramer (1999, 2001), ANSI (2000),
  Cottrell (2000), Edwards (2000), IES (2000),
  Walton (2000), Watson (2000), Gransberg (1998),
  Sandhu (1992), APWA (1986), Janoff (1984, 1986).
• Safety benefits of lighting are investigated in
  Dewar and Olson (2002), Griffith (1994), Box
  (1989, 1972), Trivedi (1988), Janoff (1984,
  1986), and Marshall (1970) ).

14
Background (cont.)

• Understanding of benefit-to-cost methods is
  provided by IADOT (2001), NYMTC (2001), McFarland
  and Walton (2000), Janoff and McCunney (1979).
• Understanding of fixed roadway lighting design
  and engineering is found in Staplin et al.
  (2001), Khan et al. (2000), Garber (2000), Couret
  (1999), Crawford (1999), Shaflik (1997),
  Jefferson (1994), FHWA (1993), and Janoff and
  Zlotnick (1985).

15
Background (cont.)

• The International Commission on Illumination (CIE
  1990) summarizes more than sixty accident studies
  from fifteen countries focused on the benefits of
  roadway lighting.
• reduction of nighttime crashes for before-after
  study of lighting

16
Background (cont.)
17
Goal

• Risk assessment and management methodology for
  the screening of needs for fixed roadway lighting

18
Methodology Development
Overview of screening methodExposure
assessmentSite parameters assessment
19
Overview of Screening Method
20
Exposure Assessment
21
Exposure Assessment (cont.)
22
Exposure Assessment (cont.)
23
Exposure Assessment (cont.)
Accepted
Marginal
Rejected
24
Site Parameters Assessment

• Evolution of NCHRP (1974) screening method
• Develop new factors
• Combine redundant factors
• Discard unnecessary factors
• Revise factor scales from 1, 2, 3, 4 or 5 to
  Low, Moderate or High

25
Site Parameters Assessment (cont.)
26
Site Parameters Assessment (cont.)
27
Site Parameters Assessment (cont.)

• Visibility-loss scenarios

28
(No Transcript)
29
Site Parameters Assessment (cont.)
30
Site Parameters Assessment (cont.)
31
Methodology Development
Site parameters
Exposure assessment

32
Data Analysis to Support the Screening Method
Unlighted nodes studyUnlighted two-mile sections
studyUnlighted half-mile sections study
33
Unlighted Nodes Study

• Study of unlighted nodes in Richmond District
• Input data
• All crashes from 1/1/1997 through 12/31/2001
• 122,126 crashes (total)
• 83,467 daylight
• 38,659 nighttime
• 12,163 nighttime, unlighted
• Source data provided by Bob Rasmussen and Ralph
  Jones, VDOT

34
Unlighted Nodes Study (cont.)
35
Unlighted Nodes Study (cont.)
36
Unlighted Nodes Study (cont.)

• Analysis of the 37 nodes with the highest number
  of crashes

37
Unlighted Nodes Study (cont.)

• Selection of 37 unlighted nodes with higher
  number of crashes

38
Unlighted Nodes Study (cont.)
39
Unlighted Nodes Study (cont.)

• Selection of 37 unlighted nodes with highest
  number of crashes

40
Unlighted Nodes Study (cont.)

• Selection of 48 nodes with highest night-to-day
  crash rate ratios

41
Unlighted Nodes Study (cont.)

• Highest night-to-day crash rate ratios occur
  primarily with AADT from 15,000 to 23,000
• 75 of night-to-day crash rate ratios are between
  1.0 - 2.0

42
Unlighted Two-Mile Sections Study

• Study of a selection of unlighted two-mile
  sections in Richmond, Northern Virginia and
  Hampton Roads districts
• Night and day crashes on the 50 two-mile sections
  considered for a six years period from 01/01/96
  to 12/31/01

43
Unlighted Two-Mile Sections Study (cont.)
Indirect night-to-day crash rate ratio estimation
for three districts
44
Unlighted Two-Mile Sections Study (cont.)

• Richmond district unlighted two-mile sections

45
Unlighted Two-Mile Sections Study (cont.)

• Northern Virginia district unlighted two-mile
  sections

46
Unlighted Two-Mile Sections Study (cont.)

• Hampton Roads district unlighted two-mile
  sections

47
Unlighted Half-mile Sections Study

• Scaling impact on calculation of crash rate
• Division of a sample of two-mile sections into
  half-mile sections

Minimal length of section to be lighted
Crash
Dangerous spots of the section
48
Unlighted Half-mile Sections Study (cont.)
49
Unlighted Half-mile Sections Study (cont.)

• Small range of half-mile night-to-day crash rates
  
• No evidence that any particular half-mile section
  had an unusually high night-to-day crash rate

50
Data Analysis (cont.)

• Summary of all unlighted nodes and unlighted
  two-mile sections

51
Example
52
Example (cont.)

• Intersection of Route 460 and Interstate 85 in
  Richmond District
• Source Bridewell (2001)

53
Example (cont.)

• Intersection of Rt. 460 and Interstate 85
• Exposure assessment
• Night-to-day crash rate ratio 1.0
• Average daily traffic 45,000

54
Example (cont.)

• Exposure parameters for Rt. 460 and I-85

55
Example (cont.)

• Exposure parameters for Rt. 460 and I-85

56
Example (cont.)

• Exposure parameters for Rt. 460 and I-85

• Outcome of the screening method is Marginal
• Exposure assessment Marginal
• Site parameters Marginal
• 6 Low and 2 Moderate
• NCHRP original score was 45.5 over a warranting
  threshold of 75.0
• Project was rejected by the NCHRP screening method

57
Summary of Conclusionsand Recommendations
58
Summary of Conclusions

• The development of a limited set of factors
  avoids the dilution from which current methods
  suffer.
• The replacement of the four NCHRP forms
  (non-controlled access or controlled access
  facilities, interchange and intersections) by one
  single worksheet simplifies the method.
• A straightforward and defensible process
  supersedes the current method, which is based on
  an obscure weight-and-score approach.

59
Summary of Conclusions (cont.)

• A benefit-to-cost analysis is directly introduced
  into the screening method via the exposure
  assessment.
• An extensive use of regional crash histories
  provides strong support to the method.
• Several examples were developed to test the
  method on real needs for visibility improvement.

60
Summary of Recommendations

• VDOT should train appropriate staff of the
  district engineers in the use of the screening
  method.
• The method presented in this report should
  supersede the NCHRP and AASHTO methods developed
  in the 1970s.
• The results of this study should be explored at
  the national level with AASHTO, FHWA, LITES and
  NCHRP.
• Regional data collection and screening of
  night-to-day crash rates should be made regular
  using our method and harmonized with the
  generation of critical rate listing.

61
Summary of Recommendations (cont.)

• Our method should be revisited as technology
  evolves and use pattern change, in particular in
  distinguishing (i) evidence that any visibility
  improvement is beneficial to safety from (ii)
  evidence that lighting or any other available
  technology is uniquely beneficial.
• A trial period should be planned for the purpose
  of evaluation and refinement of the developed
  screening method.
• Our method should be incorporated in the
  development of holistic master planning that may
  reflect the specific needs of regions and
  localities.

62
Further Discussion