Recent Geometric Design Research for

Improved Safety and Operations

 Topic Panel
◦ Steven Hersey (CO DOT)
◦ Thomas Hicks (MD SHA)
◦ John Mason (Auburn University)
◦ R.J. Porter (University of Utah)
◦ Jason Simmers (Kittelson & Associates)
◦ Heidi Sykes (CA DOT)
◦ Brian Walsh (WA DOT)
◦ Clayton Chen (FHWA)
◦ Jeffrey Shaw (FHWA)
 TRB Staff
◦ Rich Cunard
◦ Jo Allen Gause
 NCHRP Synthesis 432
◦ Description on TRB website and link to order
printed copy:
http://www.trb.org/Main/Blurbs/166996.aspx

◦ Download PDF file:

http://tinyurl.com/NCHRPsyn432
 Chapters 1-4
 Brief Q&A Break
 Chapter 5-7
 Q&A
 Closing
 Background
 Study Objective

Image credit: Khang Nguyen, Kittelson & Associates

 Much geometric design-related research
completed since 2000
 Related also to safety and operations
 Variety of sources and publications
 Synthesis 432 brings together those
sources and updates Synthesis 299
 Topics addressed in the review:
◦ Design speed and additional design controls and
criteria
◦ Horizontal and vertical alignment
◦ Cross-section
◦ Intersections
◦ Interchanges
◦ Access management
◦ Pedestrian and bicycle considerations
 Summarize recent literature
 Based on Green Book chapters

Image credit: AASHTO

 Research summarized in the Synthesis:
◦ Recommended changes to Green Book and MUTCD
◦ Produced two FHWA Roundabout Informational
Guides
◦ Contributed to:
 Guides on access management, pedestrian and bicycle
accommodation, designing for older drivers
 Highway Capacity Manual
 Highway Safety Manual

 Cross-references to these and other sources

provided throughout the report
 Design Vehicles
 Design Speed
 Design Consistency
 Driver Characteristics
 Work Zone
Considerations

Image credit: James Robertson, Texas Transportation Institute

 Recommended changes to Green Book design vehicles
(Harwood et al. 2003b):
◦ Remove WB-15 [WB-50] design vehicle, no longer common in U.S.
◦ Increase kingpin-to-center-of-rear-tandem (KCRT) distance for
WB-19 [WB-62] design vehicle from 12.3 to 12.5 m [40.5 to 41.0
ft].
◦ Replace WB-20 [WB-65] design vehicle with WB-20 [WB-67]
◦ Add a three-axle truck (SU-8 [SU-25]) and a Rocky Mountain
Double (WB-28D [WB-92D])
 Included in 2011 edition

Three-Axle Single-Unit (SU-8 [SU-25]) Design Vehicle

Image Credit: Harwood et al. 2003b
 Recommended revisions to the design speed
selection process (Fitzpatrick and Carlson
2002) included:
◦ Consideration of anticipated posted or
operating speed
◦ Modifying values recommended for different
functional classes, rural vs. urban, etc.
◦ Explicit consideration of tangent length as a
design element
 Speed relationships (Fitzpatrick et al. 2003a)
◦ Strongest relationship: between operating speed
and posted speed limit (V85 = 7.675 + 0.98 PSL)
◦ Minimal impact of design speed on operating
speeds except for tight horizontal radius or low K-
value
o Other effects include access

density, pedestrian activity,

on-street parking, and
median presence

Image credit: Marcus Brewer, Texas Transportation Institute

 NCHRP Project 15-17 (Wooldridge, et al. 2003)
◦ Recommended definition for design consistency: “the
conformance of a highway’s geometric and operational
features with driver expectancy.”
◦ List of data needs for future evaluations of design
elements:
 Cross section
 Horizontal and vertical alignment
 Railroad grade crossings
 Narrow bridges
 Driveways
 Preview sight distance
 Climbing and passing lanes
 Frequency of decisions
 NCHRP’s Human Factors Guidelines for Road
Systems (HFG) (Campbell, Richard, and Graham
2008)
◦ Designers/traffic engineers must:
 Examine roadway environment for information conflicts
 Anticipate the information the road user requires
 Seek road environments that are self-explaining, quickly
understood, and easy for users to act upon
◦ HFG recommends:
 Integration of good human factor principles into design
 Designers and traffic engineers jointly examine the road
environment for likelihood that users will be able to perform
the required tasks
 NCHRP Report 581 (Mahoney et al, 2004)
procedure for establishing appropriate work
zone design speed
◦ Equal to or slightly greater than the target speed
(desirable free-flow operating speed)
◦ Applicable to radius of curvature and
superelevation
◦ Used to determine appropriate sight distance for
design speed < 40 mph
◦ May also be used for minimum length of sag curves
 Sight Distance
 Horizontal Alignment
 Vertical Alignment
 Work Zone Considerations

Image credit: James Robertson, Texas Transportation Institute

 Stopping Sight Distance (SSD) (Fambro, et al
(2000)
◦ AASHTO’s SSD and vertical curve values were
greater than other countries
◦ Developed a (SSD) model to update the then-
current 1994 Green Book
◦ 90th %ile PRT = 2.5 s
◦ 10th %ile deceleration rate = 3.4 m/s2 (11.2 ft/s2)
◦ Results included in 2011 Green Book
 Passing Sight Distance (Harwood, et al. 2008)
◦ Evaluated differences between Green Book and
MUTCD, recommended revisions to Green Book
Passing Sight Distance for Design of Two-Lane Highways. (Harwood, et al. 2008)
Metric US Customary
Assumed Assumed Assumed Assumed
Speed of Speed of Passing Speed of Speed of Passing
Design Passed Passing Sight Design Passed Passing Sight
Speed Vehicle Vehicle Distance Speed Vehicle Vehicle Distance
(km/h) (km/h) (km/h) (m) (mph) (mph) (mph) (ft)
30 11 30 120 20 8 20 400
40 21 40 140 25 13 25 450
50 31 50 160 30 18 30 500
60 41 60 180 35 23 35 550
70 51 70 210 40 28 40 600
80 61 80 245 45 33 45 700
90 71 90 280 50 38 50 800
100 81 100 320 55 43 55 900
110 91 110 355 60 48 60 1000
120 101 120 395 65 53 65 1100
130 111 130 440 70 58 70 1200
75 63 75 1300
80 68 80 1400
  Perceived interactions between
horizontal and vertical curvature
(Bidulka, Sayed, and Hassan 2002)
 Erroneous perceptions influenced
by vertical curves increased as
◦ the sight distance increased
◦ the horizontal curve radius increased
◦ the length of vertical curve per 1%
change in grade decreased
 Faster on horizontal curves with sag
Image credit: Hassan et al. 2002 and slower on curves with crest
 Recommended establishing profile
and predicted operating speed
based on a 3D model, rather than a
traditional 2D model.
 Drivers on curves with radii ≥350 m (1146 ft)
(Schurr, et al. 2002)
◦ As deflection angle increased, speed decreased
◦ As curve length increased, speed increased
◦ As approach grade increased, 85th %ile speed decreased
 Driver behavior on horizontal curves (Liles and
Taylor 2006)
◦ Routinely exceeded posted speed limit and advisory speed
◦ More errors at:
 Curves with limited or no visibility
 Horizontal curves adjacent to vertical curves, esp. crests
obscuring horizontal curves
 Curves combined with other elements (e.g., intersections)
 Distribution of truck weight/power ratios (Torbic,
et al. 2005)
◦ 102-108 kg/kW (170-180 lb/hp) for freeways in CA, CO
◦ 126 kg/kW (210 lb/hp) in PA
◦ 120 kg/kW (200 lb/hp) in the 2001 Green Book
 Headlamp angle
in sag curve
design equation
change from 1°
to 0.75° - 0.90°
(Hawkins and
Gogula 2008)

Image credit: Hawkins and Gogula 2008

 NCHRP Report 581 (Mahoney, et al. 2004)
◦ Extended sight distance is desirable
◦ Recommended SSD for design speeds less than 40 mph
(most common basis for agency design criteria)
◦ Maximum superelevation (emax) typically a matter of policy
◦ Superelevating curves requires transitions, which induce
alignment and drainage considerations
◦ Maximum superelevation and grade for permanent
roadways is typically appropriate for work zones
 Allocation of Traveled Way Width
 Lane Width
 Shoulders
 Medians
 Roadside

Image credit: Marcus Brewer, Texas Transportation Institute

 Potts and Harwood (2003)
◦ Review of European 2+1 roads validated recommended
use in the United States
 Gattis et al. (2006)
◦ Passing activity greatest at the beginning of passing lane
segment, greatest benefits within the first 0.9 mi
 Brewer et al. (2011)
◦ Additional length less useful than additional segments,
particularly at lower volumes
◦ Avoid major intersections, consider terrain and right-of-
way, avoid ending passing lanes on uphill grades, and
discourage lengths greater than 4 miles
 Potts, Harwood, and Richard (2007)
◦ Found no general indication that lanes less than 12
ft on urban/suburban arterials increased crashes.
◦ Suggested that design policies should provide
flexibility for use of lane widths less than 12 ft.
 Gross et al. (2009)
◦ 11-12 ft lanes provide optimal safety benefit
◦ 12-ft lanes optimal design for 26- to 32-ft total
paved widths
◦ 11-ft lanes perform equally well or better for 34- to
36-ft total paved widths
 Shoulder Width
◦ Wider lane and shoulder widths are associated with a reduction in
segment-related collisions on rural frontage road segments.
(Lord and Bonneson 2007)
 Edge Treatments
◦ 11-29% crash reductions for shoulder rumble strips on rural and
urban two-lane roads and freeways. (Torbic et al. 2009)
◦ “Safety Edge” treatment - small but positive results at 56 of 81
treated sites, estimated reduction in total crashes of 5.7%
(Hallmark et al. 2006, Graham et al. 2010)

Image credit: Hallmark et al. 2006

 NCHRP Report 633 (Stamatiadis et al. 2009): CMFs for
average median width for 4-lane roads, 12-ft lanes
Width 10 20 30 40 50 60 70 80
(ft)
CMF 1.00 0.91 0.83 0.75 0.68 0.62 0.57 0.51

 Impact of median designs on crash frequency (Tarko

et al. 2007)
◦ Results were significantly different for different crash types
◦ Reducing median width without adding barriers increases
severity
◦ Reducing median width and installing concrete barriers
eliminated opposite-direction crashes, doubled single-
vehicle crashes, increased crash severity, and reduced
same-direction crashes.
 NCHRP Project 16-04 (Dixon et al. 2008)
◦ Preferred lateral offset to rigid objects of at least
6 ft from face of curb (min. 4 ft)
◦ No rigid objects within 10 ft of taper point at lane
merge locations, should be consistent with the lane
width (typically 12 ft)
◦ No rigid objects in sidewalk buffer 3 ft or less
o No rigid objects in proximity

of driveways, on the
immediate far side of a
driveway, or in sight triangle

Image credit: Marcus Brewer, Texas Transportation Institute

 Intersection Configuration
 Alignment
 Effect of Skew
 Auxiliary Lane Design
 Intersection Sight Distance

Image credit: Kay Fitzpatrick, Texas Transportation Institute

 Modern Roundabouts
 Innovative Designs
 Pedestrian and Bicycle Facilities
 Transit Considerations
 Access Management

Image credit: Marcus Brewer, Texas Transportation Institute

 Intersection Diagnostic Review Module (IDRM)
(Kindler et al. 2004)
◦ Component of IHSDM to aid designers in assessing
safety consequences of geometric design decisions
◦ Identifies concerns by “using models of the criticality of
specific geometric design situations”.
◦ Uses 21 specific models to address 15 high-priority
issues related to the intersection as a whole and to
individual approach legs.
◦ Does not select a particular treatment as appropriate
◦ User may determine whether a treatment is appropriate
treatment based on available evidence and engineering
judgment
 FHWA’s Signalized Intersections:
Informational Guide (Rodegerdts et al. 2004)
◦ Avoid approach grades more than 6 percent (3
percent for 50 mph or greater)
◦ Avoid intersections along horizontal curves of the
intersecting road
◦ Desired intersection platform with cross slope less
than 2 percent
 Ray et al. (2008): Approach curvature can
force reduced speeds at high-speed
approaches, discouraged for downhill
 Highway Design Handbook for Older Drivers
and Pedestrians (Staplin et al. 2002)
◦ Maximum skew angle of 15 degrees to
accommodate age-related performance deficits
with restricted right-of-way
◦ Prohibit right-turn-on-red for more than 15
degrees
 Son, Kim, and Lee (2002): Maximum 20-
degree skew angle when design vehicle is a
large vehicle or semitrailer
 Harwood et al. (2002)
◦ Adding left-turn lanes reduced crashes 10-44% at
signalized and unsignalized intersections
◦ Right-turn lanes reduced total intersection crashes
4-14%
 Potts et al. (2007)
◦ Method to identify cost-effective installation of
right-turn lanes at unsignalized intersections and
major driveways
◦ Economic analysis uses site-specific values for
through-traffic volumes and right-turn volumes,
crash frequency, and construction cost
 Staplin et al. (2002):
◦ Recommend PRT value at least 2.5 s for older drivers
◦ Recommend gap models use a gap at least 8.0 s, plus 0.5 s
for each additional lane crossed by the turning driver
 Yan and Radwan (2005):
◦ Sight distance problems for cars on left-turn lanes with 14-
to 18-ft medians at high design speeds.
◦ Developed equations relating sight distance and offset for
parallel left-turn lanes and sight distance and taper angle
for taper lanes.
 Easa and Ali (2006):
◦ Developed ISD model for stop-controlled intersections on
three-dimensional alignments.
◦ Presented design aids for the minimum lateral clearances
for different radii of horizontal curve and design speeds.
 General Principles from NCHRP 672,
Roundabouts: An Informational Guide,
Second Edition (Rodegerdts et al. 2010) :
◦ Slow entry speeds and consistent circulating speeds
through deflection
◦ Provide the appropriate number of lanes and lane
assignment for capacity, lane volume balance, and
lane continuity
◦ Provide smooth channelization that is intuitive
◦ Provide adequate accommodation for design vehicle
◦ Design to meet needs of pedestrians and cyclists
◦ Provide appropriate sight distance and visibility
 Alignment guidance from first FHWA Guide
(Robinson et al. 2000):
◦ Optimally located when the centerlines of all approach
legs pass through the center of the inscribed circle
◦ A slight offset to the left (i.e., the centerline passes to
the left of the roundabout’s center point) is acceptable
◦ Almost never acceptable for an approach alignment to be
offset to the right of the roundabout’s center point

Image credit: Robinson et al. (2000)

 Lane Arrangement for Multilane
(Rodegerdts, et al. 2010)
◦ Ensure that design provides appropriate lanes
within the circulatory roadway and on each exit to
ensure lane continuity
◦ Primary caution is path overlap, when the natural
path of one traffic stream overlaps the path of
another; typically has a negative effect on safety
and/or capacity
◦ Good entry design aligns vehicles into the
appropriate lane within the circulatory roadway; exit
design also provides appropriate alignment for
drivers to intuitively maintain the appropriate lane
 Additional guidance from Robinson et al.
(2000) and Rodegerdts et al. (2010) on:
◦ Entry design speed (15-30 mph, depending on
number of lanes and approach speed)
◦ Inscribed circle diameter (40-300 ft, based on
number of lanes and design vehicle)
◦ Typical entry width (14 to 18 ft)
◦ Superelevation (2% away from central island)
◦ Intersection sight distance (5.0 s critical headway)
◦ Pedestrian considerations (refuge island
dimensions)
Displaced Left-Turn
 Relocates left-turn movement to
other side of opposing roadway
 Considerable savings in average
control delay and queue length
(Jagannathan and Bared 2005)
 Increased capacity with similar
footprint to large roundabout and
small increase in costs (Simmonite
and Chick 2005)

Image credit: NCHRP Synthesis 432

Median U-Turn (aka Michigan U-Turn)
Eliminates direct left turns from
Minor street


major and/or minor approaches, in
Major street
favor of U-turn plus right turn
 Recommended optimum directional
crossover spacing 660 ft (±100 ft)
(Hughes et al. 2010)
a) Major Street Movements

Needs wide medians (at least 36 ft,

Minor street


often 60-100 ft)
Major street

b) Minor Street Movements

Image credit: NCHRP Synthesis 432
Restricted Crossing U-Turn (aka Superstreet)
 Intended for arterials with
Side street
more dominant flow on the
Main street
major road
 Only right turns from minor
 Maybe left turns from major
 Crossover spacing and median
width needs similar to MUT
Side street

(Hughes et al. 2010)

Main street

Image credit: NCHRP Synthesis 432

Quadrant Roadway
 Removes left turns from congested intersection, rerouted to a
connector roadway in one quadrant (Hughes et al. 2010)
 Two-phase signal operation at the main intersection, with
three-phase signal at each terminus of the connector road
 Considerations include location and alignment of connector
(e.g., right-of-way, distance from main intersection)

Arterial
Arterial

Cross street

Cross street
Qu a d r a n t

Qu a d r a n t
ro ro
a dw a y a d wa y

a) Left Turn Pattern From The Arterial b) Left Turn Pattern From A Cross Street
Image credit: NCHRP Synthesis 432
 Additional Designs
◦ Double Crossover (Bared et al. 2005)
◦ Arterial Interchange (Eyler 2005)
◦ Rural Intersection Alternatives (Maze et al. 2010)
 J-Turn
 Offset T
 Left-Turn Median Acceleration Lanes
 Offset Right-Turn Lanes
 Key elements that affect a pedestrian facility
(Rodegerdts et al. 2004)
◦ Keep corners free of obstructions
◦ Maintain adequate lines of sight
◦ Ensure curb ramps, transit stops, pushbuttons are
easily accessible and meet ADAAG design standards
◦ Ensure crosswalks clearly indicate where crossings
should occur
◦ Limit exposure to conflicting traffic and provide
refuges where necessary
◦ Ensure the crossing is free of barriers
 General intersection design principles and
guidelines for transit considerations (Eccles et
al. 2007):
◦ Provide simple intersection designs
◦ Provide clear visual cues to make busway intersections
conspicuous
◦ Maximize driver and pedestrian expectancy
◦ Separate conflicting movements
◦ Minimize street crossings
◦ Incorporate design features that improve safety for
vulnerable users
◦ Coordinate geometric design features and traffic
control devices
 Optimal situation is avoid driveway conflicts
before they develop (Neuman et al. 2003b)
◦ Coordinate with planners and zoning boards to
establish policies and procedures
◦ Avoid high-volume driveways near congested or
otherwise critical intersections
◦ Understand the consequences of granting driveway
requests

Image credit: Dan Walker, Texas Transportation Institute

 Design of Ramps and Ramp Terminals
 Alternative Interchange Designs
 Work Zone Considerations

Image credit: Marcus Brewer, Texas Transportation Institute

 Interchange Ramp Design
◦ Chaudhary and Messer (2002)
 Desired distance between cross street and freeway merge
point is at least 400 m (1312 ft) for ramps with metering
◦ Fitzpatrick and Zimmerman (2007)
 Source of adjustment factors for speed-change lane
length in 2004 Green Book was 1954 AASHO Blue Book,
based on applying “principles of mechanics to rates of
speed change for level grades.”
 Proposed a new procedure with an alternative set of
adjustment factors reflective of modern vehicles and
roadways
 Ramp and Interchange Spacing (Ray et al. 2011)
◦ Separately defined “ramp spacing” & “interchange spacing”
◦ Recommended ramp spacing values be primary
consideration in freeway and interchange planning and
design
◦ Guidelines based on four areas of emphasis: geometric
design, traffic operations, signing, and safety
◦ Ramp spacing of less than 900 ft is likely not geometrically
feasible for exit-exit pairs, 1600 ft for entrance-exit pairs
 Diverging Diamond (Hughes et al. 2010)
◦ Traffic moves to left side of roadway between ramps
◦ Radius should accommodate 25-30 mph
◦ Lane width approximately 15 ft
◦ Design should accommodate WB-67
◦ Avoid nearby intersections with long cycles
◦ Ped signals may be needed at free-turning movements
o Extend median noses beyond
off-ramp to improve
channelization and prevent
wrong-way maneuvers

Image credit: Hughes et al. (2010)

 Displaced Left-Turn (Hughes et al. 2010)
◦ Functions similar to DLT at-grade intersection
◦ Typically cross opposing through traffic 400-500
ft upstream of signal-controlled ramp terminals
◦ Minimum median widths preferred

Image credit: NCHRP Synthesis 432

 Summary of Findings
 Barriers to Widespread Implementation
 Recommendations for Further Research

Image credit: Marcus Brewer, Texas Transportation Institute

 Many results incorporated a series of complex
equations and/or multiple assumptions to begin
analysis.
 Similarly, computer-based simulation and
modeling is increasingly common, but not all
have access to needed software or expertise.
 Innovative intersection treatments have led to a
wide variety of potential outcomes, but research
to support those outcomes is not yet mature.
 Going forward, the cost of any treatment will
likely be further scrutinized; low-cost/high-
benefit treatments will become more desirable.
 Turning treatments at unsignalized locations
 Toll plaza and managed lane designs
 Intermediate speed (40-50 mph) roadways
 Tradeoffs between operational benefits and
safety for new uses of existing infrastructure
and right-of-way
 Relationships between roadway design, motor
vehicle speed, and motorist yielding behavior
at pedestrian treatments
 Marcus Brewer
 Texas Transportation Institute
 m-brewer@tamu.edu

 NCHRP Synthesis 432

◦ http://www.trb.org/Main/Blurbs/166996.aspx

◦ http://tinyurl.com/NCHRPsyn432