Midwest Research Institute Overview

1
Midwest Research Institute
Solutions through science and technology
NCHRP Project 15-26 Passing Sight Distance
Criteria April 5, 2005
2
Project Objectives

• Evaluate the design and operational criteria for
  determining minimum PSD
• Modify the existing criteria or develop new
  criteria

3
Project Scope

• PSD Criteria for Geometric Design
• AASHTO Green Book
• PSD Criteria for Marking of Passing and
  No-Passing Zones
• MUTCD

4
Project Tasks

• Task 1 Review current PSD criteria and models
• COMPLETE
• Task 2 Review literature and research in
  progress
• COMPLETE
• Task 3 Identify factors that potentially
  contribute to PSD requirements
• COMPLETE

5
Project Tasks

• Task 4 Critique PSD criteria and develop work
  plan
• COMPLETE
• Task 5 Prepare interim report
• COMPLETE AFTER PANEL DISCUSSIONS

6
Project Tasks

• Task 6 Execute approved work plan
• WILL BEGIN WHEN AUTHORIZED
• Task 7 Prepare new or modified PSD criteria
• WILL FOLLOW TASK 6

7
Project Tasks

• Task 8 Prepare final report
• WILL FOLLOW TASKS 6 7
• Task 9 Prepare and deliver final presentations
• WLL FOLLOW TASK 8

8
Project Schedule

• Planned Start Date for Phase II
• 2/1/05
• Actual Start Date for Phase II
• 4/6/05 ?
• Approximately two months behind schedule

9
Project Schedule

• Submit Draft Final Report
• 2/28/06
• Submit Revised Final Report
• 5/31/06

10
Outline of Presentation

• Review of Current PSD Criteria
• Section 2 of interim report
• Assessment of Current PSD Criteria and
  Alternative Models
• Sections 3 and 4 of the interim report

11
Outline of Presentation

• Potential Work Plans for Execution in Phase II
• Section 5 of interim report
• Priorities and Budget Allocations for Phase II
• Section 6 of the interim report

12
PSD Design Criteria

• AASHTO Green Book
• PSD d1 d2 d3 d4
• d1 P-R time plus initial acceleration
• d2 distance traveled in left lane
• d3 clearance distance
• d4 distance traveled by opposing vehicle
  2/3 d2

13
PSD Design Criteria

• ASSUMPTIONS
• Passed vehicle travels at uniform speed
• Passing vehicle reduces speed and trails the
  passed vehicle as it enters the passing section
  (delayed pass)

14
PSD Design Criteria

• Passing driver requires a short period to
  perceive the passing section and begin to
  accelerate
• d1 maneuver time 3.6 to 4.5 sec (older field
  data)
• d1 acceleration rate 1.38 to 1.51 mph/sec
• (2 to 2.2 ft/sec2)

15
PSD Design Criteria

• Passing is accomplished under
• delayed start
• hurried return in the face of opposing traffic

16
PSD Design Criteria

• Left-lane distance (d2)
• average speed of passing vehicle during left-lane
  occupancy exceeds speed of passed vehicle by 10
  mph
• at low speeds, passing vehicle above design
  speed, passed vehicle below design speed
• at speeds of 55 mph or more, both vehicles below
  design speed
• left-lane occupancy time 9.3 to 7.3 sec,
  depending on speed (older field data)

17
PSD Design Criteria

• Clearance distance (d3)
• 100 to 300 ft, depending upon speed
• Opposing vehicle distance (d4)
• d4 d2 would allow for completion of passing
  maneuver without need to abort
• d4 2/3 d2 will require abort in some cases

18
PSD Design Criteria

• CRITIQUE
• AASHTO model is extremely conservative
• If model were d1 d2 d3 d2, passing driver
  would know that no abort is required even before
  beginning to pass
• If model d1 d2 d3 2/3 d2, an abort is
  potentially required only very early in the
  maneuver

19
PSD Design Criteria

• Inclusion of P-R and initial acceleration
  distance (d1) is very conservative
• passing maneuver could be aborted easily and
  safely if opposing vehicle appears during d1
• an opposing vehicle that appears during d1 is a
  long way from the passing vehicle

20
PSD Design Criteria

• Ending section of adequate PSD when PSD falls
  below d1 d2 d3 2/3 d2 is extremely
  conservative
• at the end of the section of adequate PSD, a
  passing driver has almost enough sight distance
  to start a pass and complete it without the need
  to abort
• passes can be started safety beyond this point,
  if the passing driver can abort the maneuver, if
  an opposing vehicle appears

21
PSD Design Criteria

• Assumption of passed vehicle speed substantially
  less than design speed, particularly for design
  speeds of 55 mph or more, is not conservative
• Assumption of constant speed differential
  between passing and passed vehicles, independent
  of design speed is questionable

22
PSD Marking Criteria

• MUTCD
• presents warrants for no-passing zones
• passing zones merely happen where no-passing
  zones are not warranted
• where the distance between successive no-passing
  zones is less than 400 ft, no-passing markings
  should connect the zones

23
PSD Marking Criteria

• There is no quantitative model for the MUTCD
  criteria
• MUTCD criteria were first developed in a 1940
  AASHO policy

24
PSD Marking Criteria

• MUTCD/1940 AASHO ASSUMPTIONS
• no P-R time needed for pass initiation because
  passing driver can abort the maneuver if an
  opposing vehicle appears
• speed of passing vehicle is equal to 85th
  percentile speed or posted or statutory speed
  limit

25
PSD Marking Criteria

• speed differential between passing and passed
  vehicle ranges from 10 to 25 mph, with higher
  speed differentials at higher speeds
• speed of opposing is 5 to 15 mph less than speed
  of passing vehicle, with higher speed
  differentials at higher speeds
• compromise between PSD values for delayed and
  flying passes

26
PSD Marking Criteria

• CRITIQUE
• Speed of passing vehicle equal to 85th percentile
  speed or speed limit is conservative
• many passing vehicles may travel faster

27
PSD Marking Criteria

• Speed differential between passing and passed
  vehicles that increases with increasing speed is
  not conservative
• it seems more likely that speed differential
  would decrease as the speed of the passed vehicle
  increases

28
PSD Marking Criteria

• Speed of opposing vehicle less than 85th
  percentile speed or speed limit seems unrealistic
• Consideration of flying passes is not
  conservative
• Minimum 400-ft passing zones are not compatible
  with delayed passes on high-speed highways

29
Comparison of PSD Criteria
30
Comparison of PSD Criteria
31
Comparison of PSD Criteria

• AASHTO Criteria
• Driver eye height 3.50 ft
• Target height 3.50 ft (reduced from 4.25 ft in
  2001)
• MUTCD Criteria
• Driver eye height 3.50 ft
• Target height 3.50 ft

32
International PSD Criteria

• PSD Design Criteria _at_ 100 km/h
• Australia (more than US at beginning of PSD, less
  than US at end of PSD)
• Austria (about the same as the US)
• Britain (less than US at beginning of PSD, much
  less at end of PSD)
• Canada (about the same as the US)
• Germany (slightly less than the US)
• Greece (slightly less than the US)
• South Africa (about the same as the US)

33
International PSD Criteria

• PSD Marking Criteria _at_ 100 km/h
• Australia (slightly less than the US)
• Britain (less than the US)
• Canada (more than the US)
• South Africa (about the same as the US)

34
Safety Performance for Passing Maneuvers

• HSIS study (1994) found that
• passing-related accidents constitute 2 of total
  non-intersection accidents on rural two-lane
  highways
• passing-related accidents are more severe than
  non-passing related accidents

35
Safety Performance for Passing Maneuvers

• Fatal and serious injury accidents
• 13.9 of passing-related accidents
• 9.4 of total accidents
• Passing-related accidents are estimated to
  constitute 3 of total fatal and serious injury
  accidents on rural two-lane highways

36
Safety Performance for Passing Maneuvers

• FARS 2003 data
• 13,000 fatal accidents/year at non-intersection
  locations on rural two-lane highways
• if 3 are passing-related accidents, there are
  390 fatal passing-related accident per year
• Not all passing-related accidents involve limited
  PSD

37
Safety Performance for Passing Maneuvers

• COLLISION TYPES FOR PASSING-RELATED ACCIDENTS
• Single-vehicle ROR 30
• Sideswipe, same direction 25
• Sideswipe, opp direction 7
• Rear-end 17
• Head-on 7
• Other/unknown 15

38
Safety Performance for Passing Maneuvers

• Safety record of passing-related accidents is
  generally good
• It is unlikely that safety performance of rural
  two-lane highways can be modified significantly
  by changing PSD criteria
• Potential cost-effectiveness of changes to PSD
  criteria need be investigated

39
PSD Criteria in Relation to the Good Safety
Record of Passing Maneuvers

• PSD design criteria are conservative
• PSD values used in marking criteria are more
  appropriate than they seem
• Short 400-ft passing zones may not be often used
  for high-speed passes
• Most drivers may be conservative in making
  passing judgments
• A buffer area is present downstream of every
  passing zone

40
Conceptual Presentation of the Changes in Sight
Distance Needed to Complete or Abort a Passing
Maneuver as the Passing Maneuver Progresses
AASHTO
d1 d2 d3 2/3 d2
MUTCD
41
Critical Position in the Passing Maneuver

• Models based on the critical position assume that
  drivers may abort the passing maneuver until the
  critical position is reached

42
Critical Position in the Passing Maneuver

• Critical position concept
• any passing driver who has not yet reached the
  critical position must have sufficient PSD to
  abort the maneuver
• any passing driver who has passed the critical
  position must have sufficient PSD to complete the
  maneuver
• any passing driver at the critical position must
  have sufficient PSD to complete or abort the
  maneuver

43
Critical Position in the Passing Maneuver

• Two models based on the critical position provide
  PSD values similar to the MUTCD
• Glennon (1988)
• Hassan et al. (1996)

44
Critical Position in the Passing Maneuver

• First recognized by VanValkenburg and Michael
  (1971)
• they called it the point of no return
• they visually identified the critical position as
  occurring when the vehicles are approximately
  abreast (rear bumper of passed vehicle opposite
  middle of passing vehicle)
• field measurements of distance traveled by the
  passing vehicle from the abreast position to the
  completion of the pass

45
Critical Position in the Passing Maneuver

• Weaver and Glennon (1972)
• defined the critical position as the point at
  which the time required to complete the maneuver
  is equal to the time required to abort the
  maneuver
• proposed a PSD model that did not incorporate
  this definition
• stated that critical position occurs when
  vehicles are approximately abreast

46
Critical Position in the Passing Maneuver

• Harwood and Glennon (1976)
• defined the critical position as the point at
  which the sight distance required to complete the
  maneuver is equal to the sight distance required
  to abort the maneuver
• proposed a PSD model that did not incorporate
  this definition

47
Critical Position in the Passing Maneuver

• Lieberman (1982)
• used critical position concept in modeling PSD
• defined needed PSD as d7 PSDc
• d7 is distance to from the start of the pass to
  the critical position
• formulated a model to determine the relative
  position of the passing and passed vehicles at
  the critical position (?c)
• model for ?c appears incomplete

48
Critical Position in the Passing Maneuver

• Saito (1984)
• postulated that there are two possible locations
  of the critical position in a passing maneuver
• head-to-tail position (head of passing vehicle
  opposite tail of passed vehicle)
• abreast position (passing vehicle alongside
  passed vehicle)
• model inputs are not fully stated

49
Critical Position in the Passing Maneuver

• Glennon (1988)
• formulated PSD model based on equivalent sight
  distance to complete and abort a passing maneuver
  from the critical position
• included an explicit model to calculate the
  relative positions of the passing and passed
  vehicles at the critical position (?c)

50
Critical Position in the Passing Maneuver

• Glennon (1988)
• included terms for the lengths of the passing and
  passed vehicles
• incorporated speed differential between passing
  and passed vehicles that decreases as speed
  increases
• deceleration rate in abort maneuver 8
  ft/sec2 (as opposed to 11 ft/sec2 in SSD)

51
Critical Position in the Passing Maneuver
52
Critical Position in the Passing Maneuver

• Harwood and Glennon (1989)
• used Glennon model with some changes in input
  data from Glennon (1988)
• passenger car length, 19 ft rather than 16 ft
• truck length 75 ft
• smaller speed differential for truck as the
  passing vehicle
• lower deceleration rate in aborting a pass, 5
  ft/sec2 for a truck, as opposed to 8 ft/sec2 for
  a passenger car

53
Critical Position in the Passing Maneuver
54
Critical Position in the Passing Maneuver

• Rilett at al. (1989)
• recommended a minimum speed for passing abort
  maneuver Vd 2m
• inclusion of a minimum speed substantially
  lengthens PSD values
• Good et al. (1991) stated that it is unreasonable
  to expect that, in the face of an opposing
  vehicle, the passing driver would decelerated to
  Vd 2m and then continue at constant speed

55
Critical Position in the Passing Maneuver

• Rilett at al. (1989)
• recommended headway after pass abort greater than
  the 1 sec used by Glennon
• 1 sec headway appears appropriate for passenger
  cars
• headway greater than 1 sec appears needed for
  trucks

56
Critical Position in the Passing Maneuver

• Hassan et al. (1996)
• recommended two modifications to the Glennon
  model
• recommended incorporation of P-R time for pass
  abort decision (but this may already be part of
  the Glennon model)
• recommended providing PSD for pass completion
  where critical position occurs with passing
  vehicle ahead of passed vehicle (positive value
  of ?c)

57
Critical Position in the Passing Maneuver
58
Critical Position in the Passing Maneuver
59
Critical Position in the Passing Maneuver
60
Critical Position in the Passing Maneuver
61
Buffer Area Downstream of Passing Zones
62
Buffer Area Downstream of Passing Zones

• Buffer area results from short zone marking
  concept
• passes must legally be completed prior to end of
  zone
• drivers can complete passes safely even if in
  critical position at end of zone
• used in 46 of the 50 states

63
PSD Marking Criteria

• HSIS study suggests no major safety problems
  associated with PSD marking criteria
• Glennon and Hassan et al. models suggest that
  MUTCD criteria may be about right, but for the
  wrong reason
• Differences between Glennon and Hassan et al.
  models need to be resolved

64
PSD Marking Criteria

• Cost of changing PSD criteria would be
  substantial remeasuring PSD for all two-lane
  roads with centerlines
• Cost-effectiveness of potential changes in
  marking criteria needs to be resolved

65
Potential Approach to PSD Marking Criteria

• IF SUPPORTED BY PHASE II RESULTS
• Retain MUTCD criteria
• Offer a better engineering rationale for the
  MUTCD criteria based on the Glennon model, the
  Hassan et al. model, or some variation of these
  models

66
Potential Approach to PSD Marking Criteria

• Consider need to change 400-ft minimum passing
  zone length

67
Short Passing Zones

• Cant accommodate delayed passes on high-speed
  highways
• May accommodate some flying passes
• May accommodate passing very slow-moving vehicles
  tractor on the road
• May not contribute much to LOS
• Do drivers use short zones legally? safely?

68
Short Passing Zones

• Jones (1970)
• Texas roadways with 70-mph speed limits
• field studies at three passing zones with lengths
  of 400, 640, and 880 ft
• comparative data for passing zones with lengths
  of 1,640 and 2,600 ft

69
Short Passing Zones

• Jones (1970)
• defined passing opportunity as
• trailing vehicle within 4 car lengths (80 ft)
• appeared to be awaiting a change to pass
• average of 125 passing opportunities observed per
  zone

70
Short Passing Zones
71
PSD Design Criteria

• What is the rationale for use of longer PSD
  values in design than in marking?
• Does the good safety record of passing maneuvers
  indicate that current marking criteria are
  sufficient for safety?
• Does the use of longer PSD values in design than
  in marking enhance safety?
• Should the design process explicitly consider the
  passing and no-passing zones that will be marked
  on the completed highway?

72
Potential Alternative Approaches to PSD Marking
Criteria

• 1 Retain current AASHTO Green Book criteria
• 2 Use the same PSD criteria for design as for
  marking
• 3 Use PSD criteria for design with a defined
  relationship to PSD criteria for marking
• PSD X PSDc

73
Potential Alternative Approaches to PSD Marking
Criteria

• 4 - Use the same PSD criteria for design as for
  marking, but count only passing sections with
  specified minimum length
• 5 Use longer value of PSD to define beginning
  of passing section and shorter PSD to define end
  of passing section

74
Other Issues

• Trucks
• Older drivers

75
PSD for Trucks
76
PSD for Trucks
77
PSD for Trucks

• Glennon model may need longer headway after abort
  for truck as passing vehicle
• Truck can pass PC on any vertical curve where PC
  can pass a truck reevaluate in light of changed
  object heights

78
PSD for Trucks

• PC passing truck or truck passing PC may not be a
  logical design or marking scenario
• would eliminate some current passing zones and
  shorten others
• would prohibit some passes of PCs that are safe
  and are currently legal
• would reduce LOS
• might encourage illegal passes
• no indication of safety benefits

79
Older Drivers

• Reduced P-R time
• Reduced visual acuity
• Reduced ability to judge distances and speeds
• Less likely to travel at high speeds
• Less likely to pass
• Less aggressive
• More likely to drive passed vehicle than passing
  vehicle

80
PSD for Older Drivers

• FHWA Highway Design Handbook for Older Drivers
  recommended using Green Book PSDs instead of
  MUTCD PSDs
• NCHRP Project 20-7(118) recommended caution in
  implementing this Handbook recommendation

81
PSD for Older Drivers

• Use of longer PSD in marking would eliminate some
  current passing zones and shorten others
• Older drivers would still probably be more
  reluctant than younger drivers to pass
• Handbook recommendation did not consider the
  safety cushion provided by the buffer area

82
Key Considerations in Changing PSD Criteria

• Safety considerations (effect on accident
  frequency and severity)
• Traffic operational considerations (effect on
  level of service)
• Economic considerations (benefits to compensate
  for increased costs) on both existing and new
  highways
• Rationality and consistency of PSD criteria
  (understanding and acceptance by engineers)

83
Potential Phase II Work Plans

• A Benefit-cost analysis
• B Identify and analyze of passing- related
  accidents
• C Review accident data for field sites
• D Safety of passing maneuvers completed beyond
  the end of a passing zone

84
Potential Phase II Work Plans

• E Safety and operations of short passing zones
• F Field data to quantify parameters of revised
  PSD models
• G Application of PSD criteria to actual
  terrain
• H Traffic operational effects of alternative
  PSD criteria

85
Potential Phase II Work Plans

• I Comparison of PSD and SSD criteria
• J Guidance on determination of percentage of
  roadway length with PSD

86
A Benefit-Cost Analysis for Changing PSD
Marking Criteria

• Objective
• determine whether safety benefits could possibly
  justify cost of remeasuring PSD
• need to quantify
• cost per mile of remeasuring PSD
• number of miles of two-lane roads with marked
  centerlines
• number of accidents equivalent to cost of
  remeasuring
• number of accidents available for reduction

87
BIdentification and Analysis of Passing-Related
Accidents

• Objective Conduct further evaluation of
  passing-related accidents
• Identify passing-related accidents in manner
  similar to HSIS study
• Obtain more complete severity data
• Estimate available PSD at accident sites
• Determine portion of passing zone (or no-passing
  zone) where accidents occur

88
BIdentification and Analysis of Passing-Related
Accidents

• Investigate involvement of trucks
• Investigate involvement of older drivers in
  passing, passed, and opposing vehicles
• Determine weather and pavement conditions under
  which passing-related accidents occur

89
CReview of Accident Data for Field Sites

• Objective
• Assure that field sites used in Work Plans D and
  F do not have adverse accident experience
• Determine accident experience for field sites in
  Work Plan E

90
DSafety of Maneuvers Completed Beyond the End of
a Passing Zone

• Objective Determine the extent and consequences
  of pass completions beyond the end of a passing
  zone
• Sites to be used
• passing zone length 1,000 to 2,500 ft
• frequent passing activity
• no-passing zone for at least 2,000 ft downstream
  of passing zone

91
DSafety of Maneuvers Completed Beyond the End of
a Passing Zone

• Data to be collected
• percent of maneuvers completed in marked passing
  and no-passing zones
• relative positions of passing and passed vehicles
  at the end of the passing zone
• location of passing vehicles return to normal
  lane
• sight distance at return to normal lane
• traffic conflicts or severity of return maneuver
• Video recording and manual observation

92
ESafety and Operations of Short Passing Zones

• Objectives Determine whether short passing
  zones
• create safety problems
• contribute substantially to LOS
• Sites to be used
• passing zone length 400 to 800 ft
• Data to be collected
• similar to Work Plan D

93
FQuantify Parameters of Revised PSD Models

• Objective Quantify parameters of alternative
  models
• speed differential between passing and passed
  vehicles
• distance traveled by the passing vehicle from the
  beginning of passing zone to the critical
  position
• deceleration rate use in aborting a passing
  maneuver

94
FQuantify Parameters of Revised PSD Models

• Data collection approach
• Video recording
• Manual observation
• Traffic classifiers or laser guns for speed
  measurement

95
GApplication of Revised PSD Criteria to Actual
Terrain

• Objective Investigate effects of changing PSD
  design and marking criteria
• Obtain plan and profile data for actual terrain
• as-built plans
• Washington HSIS data

96
GApplication of Revised PSD Criteria to Actual
Terrain

• How will revised PSD criteria change
• percentage of roadway length that meets design
  PSD criteria
• percentage of roadway length in marked passing
  zones
• lengths of passing zones
• PSD available within passing zones
• Relative contributions of PSD criteria and
  terrain

97
GApplication of Revised PSD Criteria to Actual
Terrain

• Interviews with experienced designers
• choice of target percentage of roadway length
  that meets design PSD criteria
• application of design criteria

98
HTraffic Operational Effects of Alternative PSD
Criteria

• Objective How will revised PSD design and
  marking criteria affect the level of service on
  two-lane highways?
• Tool TWOPAS traffic operational computer
  simulation model

99
I Comparison of PSD and SSD Criteria

• Compare PSD values used in design and marking to
  SSD values used in design
• Compare resulting vertical curve lengths in light
  of
• sight distance criteria
• assumed driver eye height
• assumed object height

100
J Guidance on Percentage of Roadway Length with
Design PSD

• Provide guidance or recommendations on average
  frequency of passing opportunities that should be
  provided on two-lane highways
• dependent on traffic volumes
• dependent on vehicle mix
• needs to implement desired LOS for functional
  class and terrain

101
J Guidance on Percentage of Roadway Length with
Design PSD

• Added passing lanes along the road may reduce the
  need for PSD to maintain LOS between passing lanes

102
Project Budget

• Phase I 64,066
• Phase II 235,934
• TOTAL 300,000

103
Phase I Budget

• Budgeted expenditures 64,066
• Actual expenditures (est.) 61,000
• Unexpended funds (est.) 3,066

104
Allocation of Phase II Budget
105
Cost Estimates for Task 7 Work Plans

• A Benefit-cost analysis of PSD 10,000
• marking revisions
• B Identification and analysis of 100,000
• passing-related accidents
• C Review of accident data for 5,000
• field study sites
• D Safety of passing maneuvers 30,000
• completed beyond the end
• of the passing zone

106
Cost Estimates for Task 7 Work Plans

• E Safety and operations of 30,000
• short passing zones
• F Field data collection to quantify 50,000
• PSD models
• G Application of revised PSD 10,000
• to actual terrain
• H Traffic operational effects 15,000
• of alternative PSD criteria

107
Cost Estimates for Task 7 Work Plans

• I Comparison of PSD and SSD 5,000
  criteria
• J Guidance on determining the 10,000
• percentage of roadway length
• with adequate PSD
• TOTAL 265,000

108
Cost Estimates for Task 7 Work Plans

• Estimated cost for all 10 work plans
  265,000
• Available funds for Task 7
• 181,000
• Difference
• 84,000

109
Potential Modifications

• Eliminate Work Plan B entirely
• saves 100,000
• frees up 16,000 for field work
• Scale back Work Plan B to 16,000
• existing data bases only
• no videolog review
• no hard-copy accident report review