CE254  Design of Transportation Facilities 
seal2.jpg (53572 bytes)
Highway Design -  Class Notes






   xus23.jpg (19696 bytes)

US Route 23, Tennessee

FHWA 1996 Award of Excellence


Highway Design


Norman Washington Garrick
Spring 1998
Office Hours: TWF 11-12
return to CE254 Mainpage
E-mail to Prof. Garrick

Lecture 1


Geometric Design of Highways

Design procedures applicable for new construction, rehabilitation and reconstruction


Highway design requires skills from multiple disciplines - engineering, architecture, landscape design


Designer must satisfy many (sometimes conflicting) objectives or constraints








Geometric Design of Highways

In general, the design used in the USA is based on the AASHTO policy or some modification of this policy


Highway design relies to a great extent on engineering judgement. Therefore, the emphasis in this class is on understanding the basis for the policy


Shortcomings of the AASHTO Design

No systematic or fundamental consideration of environmental, societal and aesthetic issues


In this class we will try to address some of these issues


Road Classification


Function Classification is used to determine Design Speed and Roadway Type


Under AASHTO roads are classified as Arterials, Collectors and Locals


Roads are also divided into Rural and Urban


Functional Classification is based on the Hierarchy of Movement


Hierarchy of Movements refers to the concept that a trip can be broken into six distinct components


Road Classification

Hierarchy of Movements - six components

Main movement







Each type of movement should be handled by a facility specifically designed for that function


For function classification roads are grouped according to the type of function they are intended to provide


Main difference is based on the degree to which the road is designed to provide Access or Mobility


Road Classification

The definition of the types of roads which make up the functional categories varies depending on whether the system is in a rural or urban area


Rural Areas

Principal Arterials - statewide or interstate travel, travel between large urban areas

Minor Arterials -travel between cities, large towns and other major traffic generators, trip length and travel density greater than on collectors and local streets

Collectors - intracounty travel

Local - short destination trips, land and property access


Urban Areas

Principal Arterials - high volumes, long trip length

Minor Arterials -moderate trip length

Collectors - trips serving neighborhoods by collecting traffic from local streets

Local - roads that are primarily for providing access to properties


Other Methods of Road Classification


Functional classification is used in the USA but other approaches are used in other countries

One criticism of the AASHTO approach is that attention is paid only to the vehicle carrying function of the road and little or no attention to the other functions that the road might be required to perform

Two alternative approaches are from Denmark (roads are classified as traffic and non-traffic) and Japan


Proposed Japanese Method of

Road Classification


Level 1

Roads Exclusively for Motor Vehicles

Fully access control

Roads for Car Mobility

Partial access control - vehicle mobility emphasized

Roads for Regional Travel

Multi-function roads - provides mobility and access

Level 2

Roads for District Travel

pedestrian traffic given priority over vehicular

Pedestrian/Cycle Paths

Closed to motor vehicle traffic

Level 3

Urban Malls

Roads designed to create urban space

Lecture 2


return to CE254 Mainpage

Vehicles and Drivers

Factors which control the AASHTO design

Vehicle Characteristics

for example, effect of vehicle size

Drivers Characteristics

for example, what is the reaction time for stopping

Traffic and Highway Capacity

How many lanes?


This lecture - vehicle and drivers


Vehicle Characteristics

What are the important characteristics?


Turning path


Vehicle emissions

Vehicle noise


What is an appropriate design vehicle?

Depends on the parameter in question


What are some of the design features affected by vehicle characteristics?


Design Vehicle for Dimension Parameters

AASHTO recognizes three general classes of vehicle

Cars -- Trucks -- Buses/Recreational Vehicles


Design parameters are provided for a total of 15 design vehicles overall


The Design Vehicle is defined as a vehicle of larger dimensions and larger minimum turning radius than almost all vehicles in a given class


The design vehicle is NOT the average vehicle in a given class


For a given design problem, the design engineer must select the controlling design vehicle


This controlling design vehicle should be the largest vehicle using the facility with some frequency


In some cases, the controlling design vehicle is a specialized vehicle type for which no data is available in AASHTO - in such cases, the appropriate dimensions and turning radii must be developed - special software are available for this purpose


AASHTO gives two sets of important parameters: dimensions and turning path


Vehicles Dimensions and Turning Path

Table II-1 - Dimensions

width, height, length, overhang, wheel-base

Table II-2

Minimum turning radius

Figure II-1 to II-15

Turning path for fifteen design vehicles


Information in Table II-2 and Figure II-1 to II-15 developed from scale models and computer plots


Very important to note that all turning information given is for a 15 Km/hr turn


Larger vehicles can be accommodated - slower speeds or with some encroachment


For arterials WB-15 is usually the controlling vehicle

The designation, 15, gives the approximate length of the wheelbase of the vehicle in meters


Vehicle Performance

Two important factors of highway design are

Tractive power

Braking Ability

What values of acceleration and deceleration rates can we use for design?


The values used for design are generally very conservative - for example, value for low-horsepower car is used to determine acceleration rate for design




List the various types of design features are affected by the acceleration and the deceleration rates?



Vehicle Pollution

Two important types of environmental pollution are affected to some extent by highway design

Air pollution


Air Pollution

Mostly a function of

vehicle characteristics - type, age

operating characteristics - speed, stop-and-go mode,

operation in cold mode

environmental conditions - air temperature


Highway design probably has a minor role in mitigating air pollution


The overall planning process is probably the most important factor for reducing air pollution


To a lesser extent, design of traffic operations plays a role


Vehicle Pollution

Noise Pollution

Highway designers have more options for affecting noise pollution


For example, the decision as to whether or not a freeway should be on a viaduct or below grade has a significant effect on the noise experienced in surrounding areas


Noise effects can also be mitigated by landscaping and by using barriers - use of barriers is very controversial


Sources of Noise

engine exhaust, gears, fans (especial for trucks)

tire-interaction - important source, surface type plays big role (bricks - rumble strips)

short-term sources: brakes, backfires, horns, alarms



Amount of Noise

cruising versus stop-and-go -- speed -- type of surface




Driver Performance

As important as any other factor for developing design standards - very complex issue because it extends into the realm of human behavior and psychology


Significant amount of transportation related research is conducted by psychologist (including research here at UCONN)


Driver characteristics of interest includes

reaction time

decision making ability

sight/hearing acuity

Intelligent Highways!!!


One significant problem is the range of ability of drivers - what level of ability should we design for?


Task of Driving

Research model the the task of driving as having three sub-task


Control - steering and speed control

Guidance - vehicle following, collision avoidance, weaving etc

Navigation - trip planning and route following



Which is most complex task?



A given driving situation can be described with reference to these subtask


Stressful Drive

examples of very stressful driving situation?


what factors contribute to making stressful drive?


simultaneous decisions (control, guidance, navigation)

high speeds

no individual control over speed


obviously dangerous - little room for error


Non-stimulating Drive



what factors contribute?


non-stimulating situation is dangerous because it may lead to driver being inattentive


Driving errors results from

Driver deficiencies

fatigue, intoxication, inexperience, poor judgement

Complexity of Driving Situation

too many decisions, too much information, too little time


deficient design, inconsistent design

Conditions Leading to Inattention


One goal of design is to try to reduce errors by providing clear, timely information


Roadway Information

It is important to recognize that the information system on the road includes both a formal and and informal system


Formal System
Traffic control devices

signs, marking, delineation, signals



Informal System
roadway and it environment

alignment, grade, width


Visibility is a key factor in ensuring a workable information system

Also important is that there be consistency between the formal and the informal system


Driver reaction Time and Expectation

The time it takes for the driver to react to a situation varies with

complexity of the information


expected or unexpected

individual driver


For example, braking time varies from 0.66 seconds to 2.70 seconds




In general, it helps if design is in accordance with driver expectancy in order to reduce reaction time

Lecture 3


return to CE254 Mainpage

Traffic and Highway Capacity

Last class we looked at the characteristics of the vehicles and drivers which affect design


This class we look at traffic -- the macroscopic view


Three main issues - three main parameters that must be determined by the designer





One important point for this class - how do we derive suitable values of these parameters


Traffic Demand

The highway designer must determine the ‘design volume’ for the proposed facility. This design volume is the volume of traffic that will use the facility in the design year


The design volume that is used typically is the hourly volume in the design direction


The volume information normally available is average annual daily traffic (AADT)


AADT is not of adequate details for design - in most cases, the the design hourly volume must be estimated from the AADT


Design Hourly Volume

The typical procedure for estimating volume for design is

Estimate design hourly volume

Determine Directional Split

Determine composition


What is the design hour?

Traffic volume shows tremendous temporal variation - it is not reasonable to base our design on the hour in the year which has the highest volume. The design hour is a compromise


Volume Variation Plot

Design Hourly Volume - Rural

Rural Roads typically

Design hourly volume (DHV) = 30 HV

The 30HV is the hour in the year the 30th highest hourly volume - the reason for choosing this value as the the DHV is self evident from looking at the plot on the previous slide


The 30HV is not much higher than the HV for a larger number of hours in the year - at the same time, when we use the 30HV we are under-designing for a small fraction of the hours (29 hours) in the year


30HV versus AADT

In order to determine the DHV we need to know what the 30HV is as a percentage of the AADT


This information is usually not available - percentage is estimated from typical curve for the roadway type


30HV range from 0.12 to 0.18 of AADT 70% of all roads

PeakHV range from 0.16 to 0.32 of AADT 70% of all roads


There are some situations where the use of 30HV as the DHV is not suitable


What type of situations?

Design Hourly Volume - Recreational Routes

For routes where traffic is highly seasonal (such as recreational routes) 30HV is not suitable as DHV

Because of the seasonal of the traffic distribution, peak hourly volume is usually significantly higher than the 30HV


Example, Route 6 on Cape Cod


Recommended solution in AASHTO is to use a design volume which is 50% of the peak - this solution balances congestion in the season to required capacity out of season



Design Hourly Volume - Urban

Urban Roads typically also

Design hourly volume (DHV) = 30 HV

For roads with a pronounced peak and alternative solution

Find day with highest afternoon peak (afternoons usually larger than morning) and average over 52 weeks


In practice the resulting value is usually not much different from 30HV


Directional Distribution

Traffic volume data also show significant spatial variation which affects the interpretation and use of this type of data


The directional distribution is important because it affects the size of facility that is needed for a given traffic volume


Example - Route I 384


Directional distribution can be estimated by observing the distribution on parallel facilities - this type of data is fairly consistent from year to year



Design of major intersections and interchanges require even more detailed information - DHV of all movements for morning and evening peaks


Highway Capacity

In order to determine how large the road should be needed to determine not only the DEMAND but also the CAPACITY


Demand is given by the Design Hourly Volume (DHV) (discussed earlier)

Capacity is given by the Design Hourly Service Volume


Characterizing highway capacity is not straightforward

Capacity affected by highway and traffic conditions


Three parameters are used to analyze highway capacity

Running speed (mph) - u

Flow (veh/hr) - q

Density (veh/mile) - k


Speed-Flow Relationship

Curve for ideal conditions

q = u * k

Illustration of the Different Levels of Service

Design Service Volume

The design service volume is determined for highway design by specifying a suitable level of service (level of service A to F)


Maximum capacity is achieved at level of service (LOS) E

However, we generally design for a higher level of service than LOS E


Once the level of service is determined - the design service volume is the maximum flow rate for that level of service


DSV is given in Table II-5 for different types of roads and for different LOS

Recommended LOS is given in Table II-6

Composition and Projected Demand


For highway design applications we need to determine (at a minimum) the fraction of PC versus trucks (defined as >9000lbs with dual tires on rear axle)


This information is used to

correct volume for truck traffic

determine need for climbing lanes and passing lanes

design turning radii and other turning facilities


Projected Demand

Design is based on projected not current demand

20 year design live is recommended for new projects

Growth is difficult to predict - usually a growth percentage is assumed based on past data


Design Speed

Under the AASHTO method, design speed is probably the most important design parameter


Selected based on topography, land use, rural/urban and classification


Design speed - the maximum safe speed over a specified section of roadway under favorable conditions

(favorable conditions refer to situations in which the only constraint to higher speed is the roadway design)


The idea of using a design speed as the basis for design is that it promotes uniform operating conditions in which drivers know what to expect


Design Speed

Design features directly determined by design speed?

SSD, PSD, DSD, horizontal and vertical curvature

degree of superelevation

clear zone, median width, need for barriers


Policy recommends the use of highest practical design speed

The idea of ‘traffic calming’ goes in the opposite direction and recognize that there is some situations where lower speeds should be actively promoted

Running Speed - average stop speed on a given section of highway (measured by average spot speed). RS is affected by both congestion and design conditions and is usually less than DS even at low traffic volumes

Lecture 4


Sight Distance

Sight distance is the length (along the road) over which the driver has a clear view of obstacles in the roadway


The term sight distance is used in two different context for highway design:

sight distance for drivers travelling along the roadway

sight distance for drivers at intersections


In today's class we are concerned with sight distance for vehicles travelling along the roadway


Sight Distance

The design must select suitable sight distance criteria for use in design - minimum sight distance requirements directly affect vertical curve length


Horizontal curvature design is not directly affected by sight distance requirements - once a preliminary alignment has been selected, the designer must check to see that sight distance requirements are met at all points on the roadway



Basis for determining minimum sight distance

Procedure for measuring sight distance



Stopping Sight Distance

AASHTO define stopping sight distance (SSD), decision sight distance (DSD) and passing sight distance (PSD)


The most fundamental is SSD - the SSD value selected for a roadway is the minimum sight distance that must be provided at all locations on a roadway


DSD and PSD are only considered under specific conditions


SSD is a direct function of the design speed - the higher the design speed the longer the sight distance needed


Stopping Sight Distance

SSD is defined as the distance needed for a driver to execute a hurried stop so as to avoid striking an object in the roadway


SSD = brake reaction time + breaking distance


Brake Reaction Distance is calculated from the brake reaction time


2.5 seconds is taken to be representative of the reaction time for the 90th percentile drive to react to a simple unexpected signal


The distance is calculated assuming uniform speed during the reaction time


The speed used is directly related to the design speed but is not the design speed


Braking Distance

Braking Distance is derived directly from dynamics assuming uniform acceleration (a = v dv/dx)


Braking Distance, bd = v squared/ 254 f

d - distance (m), v - initial velocity (kph)

f - coefficient of friction between tires and road


Coefficient of friction (f) varies with road, tire condition and initial speed

Important point is that AASHTO assume a value (of f) that is applicable to wet pavements

The assumed value range from 0.40 at low speeds to 0.28 at high speeds


Design values - Table III-1

ex: DS = 110 kph Assumed Speed 91 to 110

f = 0.28 SSD 179.5 to 246.4 meters

Adjustment for grade warranted for separated roads


Decision Sight Distance

Decision Sight Distance (DSD) is the distance required for driver to perceive an unexpected or difficult to perceive situation and to select and execute an appropriate safety maneuver


The DSD is design to allow for an evasive maneuver in situations where evasion would be better than a hurried stop


DSD is considered at critical locations where complex decisions might be required or in areas of visual ‘noise’


Example of locations for DSD

Intersections, interchanges, locations of changes in width (toll plazas, lane drop)


DSD Maneuvers

DSD are given for five different types of maneuvers - the choice of the most suitable is left up to the engineer


DSD Maneuvers

A - Stop on rural road

B - Stop on urban road

C - Speed/path/Direction change on rural

D - Speed/path/Direction change on suburban

E - Speed/path/Direction change on urban


Table III-3 gives values - determined empirically


Example: DS = 120 kph

SSD = 203 to 286 meters

DSD - A = 305 meters

DSD - B = 505 meters

DSD - C = 375 meters

DSD - D = 415 meters

DSD - F = 470 meters

Passing Sight Distance

Passing sight distance is only applicable on roads where it is possible for the driver to pass by crossing into the opposing lane of traffic

two-lanes/two-way roads


The passing sight distance is the distance needed to ensure that the driver will be able to safely complete a normal passing maneuver (the minimum PSD is not designed to allow for multiple passing maneuvers)


PSD is needed for a) safety and b) efficiency

Provision of frequent passing sight distance significantly increase the service volume (capacity) of the highway


Elements of the PSD

Two important assumption for developing minimum PSD are

The vehicle being passed is operating at uniform speed

The passing vehicle is faster by 15 kph during the act of passing


The components of PSD are

d1 - distance traveled during perception and reaction and initial acceleration to point of encroachment

d2 - distance traveled while passing vehicle occupies opposing lane

d3 - distance between passing vehicle and opposing vehicle at end of maneuver

d4 - distance traveled by opposing vehicle for 2/3 of time passing vehicle is in opposing lane


Ex: DS=120 SSD=286 DSD=470 PSD=792

Table III-5


Measuring Sight Distance

The appropriate sight distance value is used to design the vertical alignment of the highway - the horizontal design is not determined primarily by sight distance concerns


Once a preliminary alignment is developed the engineer can check to determine the amount of sight distance available at each point at the road


This is typically done by checking separately the vertical and horizontal alignments - two common types of sight line obstruction

a) Road surface

b) Obstacles outside traveled way


Eye and Object Heights

In order to measure sight distance on the height of the observers eye and the object eye must be know


Assumed values (in mm)

SSD eye - 1070 object - 150

DSD eye - 1070 object - 150

PSD eye - 1070 object - 1300



Lecture 5


Horizontal Alignment

The horizontal alignment is a combination of tangent and curved components. The curves are most commonly circular curves but spiral transitions are sometimes used.


Today’s class

Properties of the horizontal curve

Determining maximum degree of curvature


Components of Circular Curves

PC - point of curvature

PI - point of intersection

PT - point of tangency

T - length of tangent (PC - PI or PI - PC)

M - Middle ordinate

L - length of curvature

D - external angle (in degrees)


Circular Curves

Traditionally, the steepness of the curvature is defined by either the ‘radius’ (R) or the ‘degree of curvature’ (D)


The degree of curvature is not used in the metric version of the policy because D is defined in terms of feet


The degree of curvature is defined as the angle subtended by an arc of length 100 ft

(this is referred to as the arc definition of D. There is also a chord definition but the arc definition is the one used for highway design applications)


Derive the relationship between D and R!


R = 5730/D


Length of Curve

Derive the relationship between L, external angle and R!


L = R D / 57.3


Note that for a given external angle the length of curve is directly related to the radius


In other words, the longer the curve, the larger the radius of curvature


How do we determine (for the following parameters) appropriate design values? external angle, curve length, curve radius


The starting point is to determine the minimum radius for design

Once Rmin is determined, the designer can use any value of R > Rmin


Minimum Radius

Equation for Minimum Radius


Derived by considering the forces acting on the occupants of a vehicle negotiating a curve and the resulting comfort level of the occupants


The minimum radius is a function of the velocity, the allowable side friction and the degree of superelevation


Derive the equation relating V, f, e, R!


f+e = V2/gR

dimensionally consistent


f+e/100 = V2 /127R

where V is kph, R is meters, g=9.809 m/s2



Maximum Superelevation

The minimum radius is a function of speed, maximum superelevation and maximum allowable side friction


What are the factors affecting the maximum superelevation and maximum allowable side friction?


Maximum Superelevation

What are the practical factors limiting side friction?


Low Speed

High CG/Loose suspension of some cars


The emax is selected based on the climate, terrain, and the likelihood of slow moving traffic


Recommended Values for emax

Absolute maximum value recommended (expect on gravel roads) is 12%


Maximum value in areas with snow and ice is 8%


Maximum value in areas where slow traffic is likely (urban areas) is from 4% to 6%


No superelevation is recommended in urban areas where congestion is expected



Side Friction Factor

A friction force is developed between the tires and the road surface to counteract forces developed during cornering


The amount of side friction that will develop is given by


f = V2 /127R + e/100


The allowable side friction (fallow) for design is much less than the impending friction that would be developed to counteract sliding


As with braking friction, maximum side friction is a function of speed, road type and condition, and tire condition


The fallow is based on the level that is comfortable and safe for car occupants

Typical values: 0.16 at 100 kph, 0.10 at 110

Lecture 6

Horizontal Alignment Design

Last class we looked at the procedure for determining minimum radius of curvature for our design


This class we look at the overall procedure for the design of each curve in the alignment

Also we will look at the procedure for determining superelevation runoff


Elements in Horizontal Alignment Design

Determine Rmin

Determine R for each curve (R>Rmin)

Determine emax

Determine e for each curve

Determine method of superelevation runoff

Determine the length of superelevation runoff



Determination of ‘e’ for each Curve

The degree of superelevation for each curve depends on the radius - ‘e’ decreases as R increases (this is given as the relationship between e and 1/R)


The AASHTO guide a parabolic (as oppose to a simple straight line relationship) between e and 1/R


Figures III-10 to 14 give design ‘e’ as a function of R and V -- each figure is for a different emax



Design Values for Horizontal Curve

Design values are given in design tables


Tables III-7 to 11 (each table contain data for different emax)


Information in Tables

Rmin vs Design Speed

L - min length of superelevation runoff

Smallest R for normal crown (NC)

Smallest R adverse crown removed

Required ‘e’ for each value of R


Lecture 7

Superelevation Runoff

On curves where superelevation is needed the roadway cross-section must be changed from a normal crown to a superelevated section - this change must be effected gradually over a relatively long distance


This distance consists of two stages: tangent runout and superelevation runoff



NC -----------RC----------------Full Superelevation


Transition both at beginning and end of curve


For design we need

To determine our method of superelevation

The length of the superelevation runoff


Length of Superelevation Runoff

The length of superelevation runoff is determined based on appearance - we want to try to avoid sharp breaks in the appearance of the profile at the pavement edge and to avoid sharp differences in the profile of the center-line and of the pavement edge


General Guideline: difference in centerline profile and edge profile should be < 0.5%


In Class Assignment

Calculate the length of run-off if the differences in slope between edge and centerline is to be <0.5%


Assume: Two-way, two-lane road, 3.6m wide lanes, nc = 0.015, e = 10%


Table III-14 gives minimum recommended L for different design speed, e, lane width

Table based partly on profile guidelines in Table III-13

Methods for Attaining Superelevation

Different approaches may be used in changing the cross-section from normal crown to superelevation


Three common methods

Revolve about centerline

Revolve about inside edge of curve

Revolve about outside edge of curve


Method 1 (about centerline) is most commonly used - the main advantage is that total change needed for each edge is not as large as with other methods

Method 3 (about outside) this has some advantages from an appearance point of view since the changes (and resulting distortions occurs on the lower side and is not as noticeable to drivers


The choice depends to some extent on need to accommodate drainage


Location of Runoff

Ideally the runoff should take place on a transitions section (such as we have with a spiral curve). If no spiral is available then we compromise: some of the runoff on tangent and some on curve


Typical Design (no spiral)

Tangent runout on tangent

2/3 of superelevation runoff on tangent

1/3 superelevation runoff on curve


Divided Highways

For divided highways, there is the additional question of whether to treat the whole cross-section as one unit for superelevation


Three approaches


Case 1: For narrow medians - whole cross-section (including median) treated as a whole unit


Case 2: Median width < 30ft - median held plane and pavement in each direction rotated separately


Case 3: Wide median - pavement in each direction treated separately (with different elevations at the median edges)


In some cases, lane widening is used on curves


Sight Distance on Horizontal Curves

Object on the inside of a curve limits the available sight distance - this formula is used to determine the relationship between sight distance and object offset


Derive the formula for the middle ordinate


M = R (1-cos (SD/200))


Spiral Curves

A transition curve is sometimes used in horizontal alignment design

It is used to provide a gradual transition between tangent sections and circular curve sections. Different types of transition curve may be used but the most common is the Euler Spiral


Advantages of Spiral Transition

Provides a smooth transition

Provides place for superelevation runoff


Properties of Euler Spiral

(reference: Route Surveying and Design, 5thEd by Meyer and Gibson]





Euler Spiral

Characteristics of Euler Spiral

Radius of spiral at any point is proportional to its length at that point

The spiral is defined by ‘k’ the rate of increase in degree of curvature per station (100 ft)




Central Angle

As with circular curve also important for spiral

Definition is a little different





compare for circular curves


Example for Euler Spiral

Note: the total length of curve (circular plus spirals) is longer than the original circular curve by one spiral leg


Example: The central angle for a curve is 24 degrees - the radius of the circular curve selected for the location is 1000 ft.


a) Determine the length of the curve (with no spiral)


b) If a spiral with central angel of 4 degrees is selected for use, determine the i) k for the spiral, ii) length of each spiral leg, iii) total length of curve


c) Verify that the condition given in the note above is correct

Lecture 8




Horizontal Alignment Design: Esthetic Factors

Functional considerations in the AASHTO method of design afford the designer a great deal of discretion in selecting the design elements - the main restriction is that the curve radius should be greater than the minimum radius.


In this case, form does not necessarily follow function. The designer must be aware of the factors affecting the form (or appearance) of their design and must work to incorporate form into the design


Why is good form important?

Highways are prominent and are a relatively permanent part of the man-made environment


Good form may enhance the function of highways - particularly from a safety stand point


Continuity of Alignment

Some of the factors that the designer must consider are i) appropriate scale for the various elements, ii) appropriate sequence, iii) appropriate transition from one element to the other


‘Man-made America’ provide some guidelines for incorporating form into the highway design process


The fundamental consideration is the need to ensure continuity of the alignment

Continuity is considered to be desirable because

Continuous alignment matches the path of vehicles (promote safety)

Continuous alignment is a better match to the natural landscape (promote esthetics and perhaps economy)


Continuity of Alignment

Continuity refers to the overall 3-D form of the roadway


Road might be continuous in horizontal alignment but overall might be disjointed due to a lack of coordination between the horizontal and vertical alignment


To achieve continuity we must consider



Coordination of horizontal and vertical


Continuity of Horizontal Alignment

The horizontal alignment may consist of the following elements: tangent, spiral curves and circular curve


Continuity is determined by

Continuity of Form - how the various elements fit together

Continuity of Scale - the relative scale of the various elements


A continuous alignment should

appear smooth, free flowing

have no kinks or breaks obvious to the eye

have elements that appear to be part of a whole not individual pieces


Continuity of Form

Form - combination of elements to provide continuity


Two major concerns

The discontinuity at the point where a tangent is connected to a circular curve

The sequence of elements


Tangent-Curve Discontinuity

Point of contact appears as a sudden break in the path (pg 166 - Merritt]





Use of spiral curve removes this discontinuity

[pg 180]


Length of Spiral

How long should the spiral be?


One criteria - spiral should be as long as needed to accommodate superelevation runoff

This is considered to be insufficient from the point of view of appearance


Rule of thumb for appearance

S:C:S should be 1:2:1

(this ratio would be 1:7:1 if the superelevation criteria is used]


In addition, it is recommended that length of spiral should NOT be greater than the length of circular curve (to avoid the appearance of a sharp bend in the middle of the curve)


Spirals are also recommended for use with compound curves (C-S-C) is R2 > 1.5 R1 (or vice versa)


Sequence of Horizontal Elements

The alignment should be continuous in terms of the relationship of successive elements


For example, a tangent should not be put in between two curves that are in the same direction (broken back curve)


Also a single curve should not be significantly sharper than others on a given section of roadway

(For Example: if significantly sharper curves are to be used in a city then there should be a gradual change from very flat curves to steeper curves as the city center is approached)


Continuity of Scale

Scale - relative length of the different elements


Two main issues

Length of curve

Ratio of tangent to curve


Desirable Length of Curve

Short curves results in a discontinuous alignment - they look like kinks in the alignment (they may also be unsafe if they are so short that drivers over look them)


At freeway speeds - the eye focus is at 1000 - 2000 ft. Curves should be at least that long to be visually significant


Recommended desirable range for freeways

L = 1500 to 5500


At a design speed of 40 mph a range of 1000 to 2000 ft is more appropriate


Ratio of Tangent to Curve

Most common design - long tangent to short curve


For example on Merritt - alignment is 80% on tangent

This results in a discontinuous alignment where each element appear to be separate


One alternative - short tangent to long curve (Spline Alignment)

Example: GS Parkway where only about 20% of alignment is on tangent

This results in a visually continuous alignment in which the tangent appears to be part of a continuous compound curve


One difficult with this design is that it is difficult to incorporate spiral curves when the tangent is so short


Curvilinear Alignment

Second alternative - Curvilinear Alignment

Long, flat circular curves (simple or compound) connected by long spiral

Two-thirds of length on circular curve


Requires differ approach to setting out alignment

Road set out on the basis of arcs (circular) which are as long and flat as practical and are then joined by spiral transition


Tangents used in cities or in areas of flat terrain where they would be a better fit than flat curves


1/R Diagrams

1/R Diagram is a visual and numerical tool for evaluating the continuity of the horizontal alignment


On 1/R plot

Tangents - plot as zero

Curves - plot as straight horizontal lines

Spirals - plot as sloping straight lines



Lecture 9


Vertical Alignment Design

The vertical alignment is a combination of tangent and curved components. The curve sections in this case are PARABOLIC CURVES


The main design elements for vertical alignment

Design of Vertical Curve

Determination of maximum grade

Determination of suitable length of grade (and design of climbing lanes where appropriate)


This class

Properties of parabolic curve

Design of vertical curves

Parabolic Curve


Grade changes gradually from G1 at VPC to -G2 at VPT


Total change in grade is A (the algebraic difference in grade in %)


A = |G1 - G2|


Curve characterized by K, the rate of change of curvature (given as the length of curve for a 1% change in grade)


K = L/A


Which is the gentler curve - small K or large K?



Formulas for Parabolic Curves

The Offset from the Tangent (y)

The offset is proportional to square of the horizontal distance from the VPC


y = Ax2/200L


The Distance to the turning point (xt)

What is the grade at the turning point?

What is the change in grade from VPC to tp?

K is length for 1% change in grade. What is formula for xt?




Elevation of turning point

Elevation at turning point = (Tangent elevation at xt) - yt

yt is the offset at the turning point. What is formula?

yt = Ax2t/200L

Example Calculations for Parabolic Curve


G1 = -1% G2 = +2.2%

Elevation of VPC = 125.230 m

Station of VPC = 2+500 (== 2500 m)

Station of VPI = 2+600


What is length of curve? K-value?


Find a) Station and Elevation of Low Point, b) Elevation at station 2+550 and 2+650.



Design of Vertical Curves

The main design issue is the determination of the minimum length (or minimum K) for a given design speed


The criterion that determine the minimum differs depending on the curve type (crest or sag)


Four basic curve types

Type I and II (crest curves)

Type III and IV (Sag Curves)


Design of Crest Vertical Curves


What practical factors are affected by length of curve?


Most stringent factor is the need for sufficient sight distance


Crest Vertical Curves

If the sight distance requirements are met then safety, comfort and appearance will not be a problem


Based on sight distance, formulas for minimum curve length


if S<L,



if S>L,



Try the first equation, if the result is inconsistent with S<L then try the second


for SSD, h1 = 1070 (1.07 m) , h2 = 150 (0.15 m)


if S<L,


if S>L,


Design Charts for Crest Vertical Curve

Design values are given in FIII-39 and 40

(III-39 is based on upper limit for SSD and III-40 on the lower limit)


Figure is a plot of A vs Lmin for different design speeds


Minimum K value is also given at each design speed - this K can be used as the design control (expect at low A values)


At low A, a limiting value of L is specified (this limiting value of L is approximately 0.6 V)


The curve length should always be greater than the value in FIII-39 or 40

Drainage Criteria

For proper drainage on type I crest curve (with curbs) the AASHTO recommend that the minimum G (slope) 15m from the low point should be at least 0.3%


What is the maximum K value that correspond to this criteria?


K = 51 (m per %)


This value of K is NOT considered an absolute maximum for design - the guide states that if this value is exceeded attention should be paid to ensure that the site will drain adequately




Sag Vertical Curves

Obviously SSD is not a problem for sag curves

What are the physical issues precluding the use of very short curves?


Four criteria considered

Headlight sight distance

Rider comfort




The AASHTO design is based on headlight sight distance

Minimum Length for Headlight Sight Distance






Sag Vertical Curves

What is the HLSD? HLSD == SSD


Drainage criteria is exactly the same as for crest vertical curves - if section is curbed then we need special attention to drainage if K>51


Design Charts

Figure III-41 and 42

Table III-37


Lecture 10


Vertical Alignment Design - Maximum Grade

The maximum grade is determined based on vehicle operating characteristics


In general, passenger cars have

Little or no loss of speed on 4-5% upgrades

Somewhat higher speeds for downgrades

No noticeable change on grades < 3%


For trucks,

Operating speeds on the level is approximately the same as for passenger cars

Significant loss of speed on up-grades

Increase in speed of up to 7% on downgrades


Maximum Grade and Length of Grade

Maximum Grade

AASHTO recommendation for maximum grade

5% at DS of 120 kph

7-12% at DS of 50 kph

Maximum should only be used were unavoidable

<Glenwood Canyon>


Minimum Grade

No curb - 0%

With curb - 0.5% (0.3% on high type pavements)


Length of Grade

How long should the slope be for a given grade?

The main issue is that the longer the slope the more the reduction in truck speed

If the reduction in truck speed is too great then we have a safety problem and a reduction in highway capacity

Critical Length of Grade

Accident rates have been show to increase dramatically as the speed differential between PC and trucks become greater than 15kph


Speed of truck on upgrade function of


Length of grade

Weight/horsepower ratio

Entering speed


The AASHTO guide is based on wgt/hp ratio of 180 kg/KW but charts are also given for recreational vehicle


The designer need to determine if 180 kg/KW is appropriate for a given situation


Critical Length of Grade

AASHTO guide provides charts for determining the speed reduction - Figure III - 25A, III - 25B

Using the charts

a) Truck enter 6% upgrade at 50 kph. What is distance for 15kph reduction in speed? (III-25A)


b) Truck enter 3% downgrade at 30 kph. What is the speed after 500 m?


Measuring Length of Slope

The critical length determined from above is the length on a tangent section - length on a vertical curve must be converted to an equivalent tangent grade length


For type I and III curves - 1/4 of the curve length is taken as part of the grade under consideration

For type II and IV curves - the length is measured from the VPI


Measuring Equivalent Length of Grade


What is the equivalent tangent length for the 5% grade



L = 1/4 of type III curve + distance from start of tangent to the PVI of type IV curve

L = 1/4*200 + 1000 = 1050 m

Climbing Lanes

If the equivalent tangent length of curve exceeds the critical length then CLIMBING lane may be warranted

Aside from the grade length, factors that are considered in evaluating the need for climbing lane include i) traffic volume and ii) traffic composition


Climbing lane should

Begin where truck speed falls to below 15 kph of the overall running speed

End where truck speed is again within 15 kph of the overall running speed (a practical compromise design is to end the lane where the truck can return to the normal lane without undue hazard, i.e., where there is sufficient SD for safe passing)




Lecture 11


Continuity in Vertical Alignment

From an esthetic point of view, one of the goal in alignment design is to achieve a continuous alignment. Continuity is desirable so that the road fit the terrain, does not have a jarring aspect and present a path which is easy to follow


To achieve continuity two different elements are considered

Continuity of Form

Continuity of Scale


Continuity of Form

For vertical curves continuity of form is not a problem. There is no sudden break in grade at the PC or PT since the parabolic curve provides a gradual transition in grade


Continuity of Scale

Continuity of scale relates to the relative lengths of the tangent and curves


If the curve is very short, it is seen as a sudden break in the alignment


Common flaws from very short curves

Board Effect, Hump, Break


Length that is adequate from the point of view of function is too short for appearance


Function Design: DS = 110 kph

Crest - kmin = 80

Sag - kmin = 43


Minimum desirable for appearance is 300 m

This minimum is particularly important for Sag and crest with small ‘A’

For crest with large ‘A’ minimum length for function is adequate for appearance


The recommended design in Man-made America is a curvilinear profile


Curvilinear profile is defined as one with greater than 50% of the alignment on curve (traditionally only 25% of alignment is on curve)


This curvilinear alignment can be achieved by

reducing number of curves

increasing length of remaining curves

using compound curves


Coordination of Vertical and Horizontal Alignment

Continuity in plan and profile does not guarantee overall continuity of alignment in 3-D


Vertical and horizontal alignment must be designed to be in sync with each other


Two main issues for coordination

Relative scale of vertical and horizontal elements

Relative location of horizontal and vertical curves


General Rules for Proper Coordination

Curves in vertical and horizontal alignment should be about the same length


Horizontal and vertical curves should not start and end at the same point. Horizontal curve should lead and generally remain longer


Horizontal and vertical curves should generally coincide in location (location of vertices). A shift of up to a quarter phase is OK. A shift of half a phase is not desirable (appear as break in alignment)


Plots for Evaluating Coordination

1/R Diagram

Visual and numerical tool for evaluating the continuity of the horizontal alignment


On 1/R plot

Tangents - plot as zero

Curves - plot as straight horizontal lines

Spirals - plot as sloping straight lines


Area under plot proportional to degree of curve


1/100K Diagram

Visual tool for evaluating vertical alignment

Vertical curve is roughly equivalent to a circular curve with radius of 100K


Area under plot is proportional to ‘A’


These two plots can be used together to assess coordination of vertical and horizontal alignment



Cross-sectional Elements

Cross-section MAY include (depending on road type)

Traveled way (shoulder not included)




Roadside barriers

Side slopes

Drainage channels

Main design issues

Selection of elements




Traveled Way

The traveled way is the portion of the cross-section that is designed for vehicle travel

Lane width of Traveled Way

Lane width affect

Safety - vehicle users and pedestrian

Comfort - vehicle users and pedestrian



Width chosen should reflect the road way type

In USA, usually 2.7 to 3.6 meters

AASHTO tend to recommend wider lanes but narrower lanes are preferable in some cases

<<Traffic Calming>>



Cross-slope of Traveled Way

On road sections with no superelevation, the traveled way must have some cross-slope for drainage

The type of cross-slope that may be used varies

Crowned (plane and curved)

Unidirectional slope

Undivided Pavements

Crowned cross-section used (curved or plane)

Curved crown (usually parabolic Curve)

Advantages - facilitate drainage because of increase in slope towards edges

Disadvantages - a) more difficult to construct, b) poses more difficult design problem at intersections, c) very large slope at outer edge for multilane roads


Cross-slope of Divide Highways

Two options for cross-slope on divided highways

Crowned separately

Unidirectional slope

Crowned Separately

Advantage - rapid drainage and smaller difference between high and low point

Disadvantage - More inlets and underground drainage, more difficult design at intersection

This design recommended for area with freeze/thaw and high rain areas

Unidirectional Slope

Main advantage is more comfortable for lane change

Drainage may be to or from median

To median - outer lanes (higher volumes) free of water but drainage is over high speed lanes

From median - savings in drainage cost


Rate of Cross-slope

Trade-off - should be steep enough for drainage but not too steep to be objectionable for comfort and appearance

Typical values - 1.5% to 2% on high-type, high speed roads (2% is about 1/4 in to 1 ft)

Greater than 2% not desirable due to roll-over effect in lane change for plane crowns

However, 2.5% maybe necessary in high rain areas

(Rounded crowns alleviate the roll-over problem)

Higher rates are recommended for lower type pavements because of drainage

intermediate - 1.5 to 3.0%

low - 2.0 to 6.0%

Generally the higher values in the range should be used with curbed sections to prevent sheeting of water



Shoulder are recommended for rural highways with high volumes, freeways and some urban highways

Function of the Shoulder

Stopped vehicles, escape

Ease of driving, increase capacity, higher speed

Snow removal/storage

Lateral clearance for signs/guard rail


Structural support for pavement

Pedestrian, bicycles, bus stops

Shoulder Width

AASHTO recommend shoulder width of 0.6 to 3 meters

Three (3) meters allow a stopped vehicle to clear pavement edge by about 2 feet

Three meter width used on high type roads


Shoulder Width

Walls and barriers should be off-set by at least 0.6 meters from edge of shoulder

Shoulders operate best if they are continuous.


Partial shoulders (reduced width) are sometimes used where it is too costly to provided full shoulder (for example, on bridges)

Intermittent (non-continuous) shoulders are also an option where continuous shoulders are not feasible (for example where the road is in a mountainous area - Big Sur)

Shoulder Cross-section

Shoulders form an important element of lateral drainage

For drainage and for safety it is important that the shoulder be flush with the pavement

Recommended slope depend on shoulder material

AC or PC - 2 to 6%

Gravel or crushed rock - 4 to 6%

Grass - 8%

With curb minimum recommended is 4%

Shoulder Cross-section

Direction of Slope of Inside Shoulder

With depressed median - slope to median

With raised median - slope away from median (except in region of high snowfall)


Shoulders on Superelevated Sections

Avoid large roll-over between the shoulder and the pavement






Maximum roll-over should be 8%

Methods for reducing roll-over

Flatten shoulder near pavement

Rounded shoulder (problem: some drainage on pavement, difficult for construction)

Plane shoulder with multiple breaks


Shoulders Contrast

AASHTO recommends different color and texture for shoulder to clearly delineate shoulders from through-lanes

Differentiation by texture is fairly common

Helps to prevent use of shoulder as a through-lane

The ConnDOT recently added rumble strips at pavement edge - main reason is probably as a warning but also help in delineating shoulder



Used for

Drainage control

Delineation/separation of pavement edge

Delineation/separation of pedestrian walkway

Used for urban facilities, rarely on rural facilities

Two types





Barrier Curbs

Designed to discourage vehicles leaving the road

used to separate vehicles and pedestrians

used along walls or tunnels to prevent scraping

Six to eight inches high

Not suitable for high speed arterials (DS>60kph) - may cause vehicles to vault barriers

Also not used with roadside barriers

Mountable Curbs

Designed to allow vehicle to cross barrier (if needed)

Commonly used to outline channelized island and at outer edge of shoulders

Should not be used

next to high speed lanes

with concrete barriers

in front of flexible barriers



Provided as a separate way for pedestrian - used on urban roads, less frequently on rural

Provision of walkways significantly improve safety and considered critical in developed sections of rural towns - unfortunately, frequently not provided around developments such as malls

AASHTO recommends that walkways be provided along ANY road which does not have a shoulder

Sideways in urban areas generally raised above the level of the road and separated by a curb

Sideways in rural areas may be at the same level as the road (since speeds are generally higher on rural roads - separation between road and walkway should be greater



The area between the roadway (including any shoulder) and private property is called the BORDER - the walkway is a part of the border

Border provide space


street lights

fire hydrants

street hardware


Width of border varies widely but the minimum is 2.5 meters

Residential area sidewalks should be 1.25 to 2.5 meters - width depends on pedestrian traffic volume

Walkways placed directly next to curb should be wider than walkways that are separated from the curb by a planted strip - allow space for street hardware, opening of doors and proximity of traffic


Overview of Drainage System

Purpose: Remove all surface water from roadway and intercept and remove water from adjacent areas. Drainage system consists of both surface and sub-surface elements

Drainage Elements




Longitudinal ditches

Culverts and Bridges < Cross Drainage

Curb, gutters < Urban

Inlets and storm drains < Urban/Divided highways



Drainage Channels and Sideslope

Main design criterion



economy of construction and maintenance

Good results are achieved by

flat side slopes

broad drainage channels

liberal warping and rounding of corners


Drainage Channels

Used to collect surface water from ROW


Roadside channel in cut

Toe-of-slope channels in fill

Interceptors at top of cut slopes

Chutes - down cut or fill slopes



Flat slopes are desired for safety and stability but ROW availability may force steeper slope

Three separate areas are important for safety

Top of slope


Toe of slope

Top of slope

must be rounded to reduce hazard




Slopes steeper than 3:1 need barrier (and may also be unstable)

Slopes gentler than 6:1 should be provided were feasible (errant vehicle has a high probability of recovering on such gentle slope)



Overall Design of Cross-section

Two important issues in the overall design of the cross-section are safety and appearance

The overall cross-section design has a disproportionate effect on the appearance of the road (specifically on the public’s perception of the appearance)

In generally, factors that enhance appearance also enhance safety

The most important factor is that the cross-section should be CONTINUOUS

The cross-section should be visually continuous internally and with the adjacent natural landscape


Continuity of Cross-section

Discontinuous cross-section

Steep slopes on low fill

Constant grade

Visually separate elements

Continuous cross-section

Gently slopes on low fill

Variable grade

Each element flow together

The road should appear to be part of the landscape not a foreign object


Cross-section Slope

Man-made America recommends for appearance

0 to 12 feet FILL - 6:1 slope

10 to 20 feet FILL - 4:1 slope

0 to 10 feet CUT - 4:1 slope

10 to 25 feet CUT - 3:1 slope

From a safety point-of-view the FHWA recommends the following for the foreslope

Steeper than 3:1 need barrier (and may also be unstable)

Gentler than 6:1 should be provided were feasible since errant vehicle has a high probability of recovering

Barrier are unavoidable in some situations but they result in a visually discontinuous alignment

Both Man-made America and FHWA recommends rounding and warping corners


Variable Cross-slope

Design using constant cross-slope is pervasive problem which works against continuity of cross-section

Man-made America suggests the use of a grading plan instead of cross-sections as the tool for design

Grading Plan

Much more precise

Easier and more precise for earthwork calculations

More accurate drainage design

More efficient for design work and design presentation

Major drawback is that contractors are not familiar with reading grading plan - a solution might be to design using grading plan and then produce cross-section from the grading plan for the contractor


Grading Plan

Grading Plan - Class Project

Project Information

Topo map of road with constant cross-slopes

Topo map of road with variable cross-slopes


Draw cross-sections at the two locations shown for each of the two grading plan

Calculate the earth volume for each plan (only for the length of road covered by the cross-sections developed above)


Traffic Barriers

Two general classes - longitudinal barriers and crash cushions

Longitudinal barriers are designed to redirect or contain an out-of-control vehicle and so ensure that the vehicle does not leave the roadway. Used at locations that are particularly hazardous for vehicle to attempt to traverse

Crash cushions are used to stop or redirect an out-of -control vehicle by absorbing the energy of impact and so to prevent vehicle striking a fixed object near the roadway

Traffic barriers run along the side of the road - for example, along narrow medians

Crash cushions are used at spot locations to shield a specific object - for example, they are used in the gore area of freeways (area between an exit ramp and the mainline)

The road design should be done to reduce to an absolute minimum the need for barriers and cushions


Longitudinal Barriers

Three types




Classification based on barrier deflection after impact


Designed to contain rather than redirect

Needs significant lateral clearance behind barrier


Redirect the vehicle through bending and tensile strength of the barrier

Need much less lateral clearance


Longitudinal Barriers


No deflection of barrier

Redirect vehicle

Energy is dissipated by deformation of the sheet metal of the vehicle

No damage to barrier

Only used with very SHALLOW angle of impact

Rigid barriers used in narrow medians and shoulders and at other locations where replacement is difficult

Factors Affecting Type of Longitudinal Barrier

Major factor is that deflection characteristics of the barrier should be compatible with space available

Other factors

Safe transitions and end treatments

Initial and maintenance cost



Where are longitudinal barriers needed?

Guardrail cause about 1200 fatalities each year!

This is 10% of total fatalities caused by fixed objects!

Guardrail are designed to save live but they can also be injurious - engineers must use guardrails only where the risk from the roadside is greater than the risk from the guardrail

Comprehensive reference

"Roadside Design Guide" - AASHTO

The Clear-zone Concept helps to determine whether or not barriers are needed

The idea of the clear-zone is to provide a traversable and unobstructed roadside area beyond the DRIVING lane



The clear-zone width depends on the

Design speed


Slope (in the clear zone)

Horizontal Alignment

Example (on tangent)

(ref: Figure 3.1)

Design speed: 40 mph

Clear zone slope (fill): 10:1

Clear zone width - 7 to 12 ft (depending on vol)

Design speed: 60 mph

Clear zone slope (fill): 10:1

Clear zone width - 16 to 30 ft (depending on volume)



As can be seen above, the Clear-zone maybe partly a traversable slope

A slope of less than 4:1 is considered to be traversable and recoverable (driver can stop and return to road)

Slope between 4:1 and 3:1 considered traversable but non-recoverable (driver not able to stop - will reach bottom of slope). For fill slope, need a run-out area at base of slope

Slope greater than 3:1 considered non-traversable (vehicle is likely to overturn)

A barrier is almost always needed for such a slope (Steep and smooth cut rock slope may not need barrier)

Fixed objects in the clear-zone should generally be shielded


Features which may Warrant Barrier

Barriers maybe needed for the following situations

Bridges and Drainage

Bridge piers, Abutments and Railing ends

Shielding almost always needed

Culverts, pipes, headwalls

based on size, shape, location

Ditches (parallel)

depends on design

Ditches (traverse)

shielding required if head-on collision likely

Culverts, pipes, headwalls

based on size, shape, location


Features which may Warrant Barrier

Slopes and Retaining Wall

Cut Slopes (smooth)

not generally required

Cut Slopes (rough)

depend on likelihood of impact


Depends on fill height and slope

Embankment (cross slopes)

Potential head-on collision, slope should be 6:1 or less (10:1 preferred)

Retaining Wall

Depends on smoothness and angle of impact


Features which may Warrant Barrier

Other Obstacles

Sign/light Support

if non-breakaway

Traffic signal support

isolated signals in clear zone


Depends location and size

Water Body

Depends on location and water depth


Barrier Design

Road Cross-section

Optimally the area in the vicinity of the barrier should be clear and flat (especially for flexible and semi-rigid barriers)

Curbs should be avoided near barrier - if used should be less than 9" in front (preferably, in line or behind)

Longitudinal barriers may be

Roadside barriers

Median barriers


Roadside Barriers

Used to

Shield vehicle from roadside hazard

Shield pedestrians, cyclists and property

Design Features of Roadside Barriers

Two design elements are particularly important for safety

End treatment


End Treatment - untreated or square approach posed an extreme hazard

Flaring is probably the best end-treatment for flexible or semi-rigid systems

Break-away cables also used

Rigid - end protected by crash-cushion


Roadside Barriers


The Transition from barrier to bridge railing also important - must design to reduce the risk of snagging or redirecting the vehicle into the bridge railing

Important design elements for transition

Rail and barrier should be aligned

Barrier should be secured to bridge

Barrier should have transition in stiffness



Median Barriers

Used on divided highways

For freeways

Generally used if median is less than 20 to 30 ft (depending on volume)

Never need above 50 ft

(See: Figure 6.1 in reference)

Aside from median width and volume another factor considered in determining the need for barrier is relative elevation of the two roadways

Most important safety issue is the END TREATMENT design

Also, barrier type should have deflection less than half the median width

Types of barriers: Steel on post, box-bean on post, concrete and earth berm



return to CE254 Mainpage
E-mail to Prof. Garrick