Geometric Design 09-05-21
Geometric Design 09-05-21
Course Notes
1. Introduction
Highway geometric design comprises the processes necessary for choosing appropriate geometric
features for a highway. These are the visible elements of the road. The geometric designer is the
architect of the road since geometric design includes a fair measure of art amongst the science.
The proper geometric design of a highway ensures that drivers use the facility with safety and
comfort. The process achieves this by selecting appropriate cross section, vertical and horizontal
curvature along with physical features of the road such as sight distances and superelevation. The
ultimate aim of the procedure is a highway that is both justifiable in economic terms and
appropriate to the local environment. Design speed is the determining factor in the selection of
the important geometric characteristics of a highway.
2. Design Speed
Design speed is the primary factor in selection of other design element of new highway
including horizontal curve radii, superelevatio/n rates, and vertical curve elements. These
are various definitions of design speed:
AASHTO (1994): maximum safe speed that can be maintained over a specified section
of highway when conditions are so favourable that the design features of the highway
govern
AASHTO (2004): A selected speed used to determine the various geometric design
features of a roadway
DMRB-UK (1993): A selected speed consistent with the anticipated vehicle speeds on
the road
Road Design Manual – Kenya – Maximum safe speed that can be maintained over a
specified section of a road when conditions are so favourable that the features of the
road govern
2.1 Selection of Design Speed
The design speed of a highway serves as a guide in the selection of the physical features
referred to above. Selection of the correct design speed ensures that issues of both safety
and economy in the design process are addressed. The chosen design speed must be consistent
with the anticipated vehicle speeds on the highway under consideration.
The standard design speeds are 50km/hr, 60km/hr, 70km/hr, 85km/hr, 100 km/hr and
120km/hr. These bands are based on the premise that it is considered acceptable if 85% of
drivers travel at or below the designated design speed for a given highway, generally inducing
a situation where approximately 99% of the drivers travel at or below one speed category
above the design speed (i.e. if the speed limit is set at 85km/hr, it can be assumed that 85%
of the drivers will travel at or below this value while 99% will travel at or below 120km/hr).
The guidance of the roads design manual (Kenya) in selection of design speed is shown below.
For a redesign or highway upgrade it is common to choose the 85th percentile as the design
speed on the basis that it constitutes the most appropriate and judicious choice.
3. Sight Distance
Sight distance is the length of the roadway a driver can see ahead at any particular time.
The sight distance available at each point on the highway must be such that, when a driver is
traveling at the highway’s design speed, adequate time is available after an object is observed
in the vehicle’s path to make the necessary evasive maneuvers without colliding with the
object.
The two types of sight distance are stopping sight distance (SDD) and passing sight distance
in %
Exercise 1
You have been tasked with designing a highway. The design speed has been selected to be 80
km/hr. Assuming a perception-reaction time of 2.5s, determine the safe stopping sight
distance you would provide at a level section of the proposed highway
𝑉1 = 80km/hr = 22.22 m/s
G= 0%
a= 11.2 ft/s2 = 3.41 m/s2
g= 9.81m/s2
𝑉1 2
𝑑𝑠 =d + 𝑑𝑟 = + 𝑉1 × 𝑡𝑃𝐼𝐸𝑉
2𝑎
22.22 2
𝑑𝑠 = + 22.22 × 2.5
2∗3.41
= 127.95m
(Compare this with the recommendations of the Kenya’s road design manual)
𝑽𝟏 𝟐
𝑑𝑠 =d + 𝑑𝑟 = 𝒂 + 𝑉1 × 𝑡𝑃𝐼𝐸𝑉
𝟐𝒈( ±𝑮)
𝒈
33.332
𝑑𝑠 = 3.41 3 + 33.33 × 2.5
2×9.81( + )
9.81 100
= 233.28m
𝑽𝟏 𝟐
𝑑𝑠 = 𝒂 + 𝑉1 × 𝑡𝑃𝐼𝐸𝑉
𝟐𝒈( ±𝑮)
𝒈
3.2 Passing Sight Distance
Passing sight distance which also referred to as full overtaking sight distance refers to the
length of visibility required by driver of a vehicle to complete an overtaking manouvre without
colliding with an opposing vehicle and without cutting off the passed vehicle.
The passing sight distance will also allow the driver to successfully abort the passing maneuver
(that is, return to the left lane behind the vehicle being passed) if he or she so desires.
The following assumptions regarding the movement of the overtaking vehicle during a passing
maneuver:
1. The vehicle being passed (impeder) is traveling at a uniform speed.
2. The speed of the passing vehicle is reduced and is behind the impeder as the passing section
is entered
3. On arrival at a passing section, some time elapses during which the driver decides whether
to undertake the passing maneuver.
4. If the decision is made to pass, the passing vehicle is accelerated during the passing
maneuver, and the average passing speed is about 16km/hr more than the speed of the
impeder vehicle.
5. A suitable clearance exists between the passing vehicle and any opposing vehicle when the
passing vehicle reenters the left lane.
Since it requires a greater distance to overtake than to stop, the FOSD values are greater than
the SSD values.
It is usually necessary to consider whether to design for overtaking only at crest curves on
single carriageways, since overtaking should not be a problem on dual carriageways and
visibility is usually more than adequate for overtaking at sag curves on single carriageways.
The passing sight distance is the sum:
𝑑 = 𝑑1 + 𝑑2 + 𝑑3 + 𝑑4
Where:
𝒅𝟏 is the initial maneuver distance, which is the sum of distances traveled during perception
and reaction time plus the initial period of acceleration until the vehicle encroaches the
passing lane. The corresponding time for this initial maneuver is 𝑡1 : d1 is given by the
expression:
v is the average speed of the passing vehicle
m is the relative speed of the impeder and passing vehicles
a is the average acceleration
𝒅𝟐 is the distance traveled while the vehicle is occupying the passing lane. The corresponding
time is 𝑡2
𝒅𝟑 is the clearance length, for margin of safety between the passing and opposing vehicles.
AASHTO recommends that 30 ≤ 𝒅𝟑 ≤90 m for 56 ≤ v ≤90 100 km/h
𝒅𝟒 is the distance travelled by the opposing vehicle during the passing maneuver during time
𝑡1 + 𝑡2 . AASHTO recommends that the opposing vehicle may be assumed to travel with speed
v; and 𝑑4 may be taken as
2
𝑑4 = 𝑑
3 2
Roadways and railway tracks are never straight from end to finish. They changing directions.
Straight sections are usually connected by curves.
The purpose of the curves is to enable a vehicle travelling along one of the straights to safely
and comfortably change direction and continue its journey along the other straight. The
change in direction from one straight to the next is known as the deflection angle (Ø)
A road design is made up of a combination of straights (tangents) and the curved sections.
Together they form what is known as the HORIZONTAL ALIGNMENT.
The curves joining straight sections could be of constant or varying radius. Circular curves,
which are horizontal curves of constant radius while Transition curves are curves of varying
radius. Transition curves are preferred in order to avoid a sudden/abrupt change from a
tangent with infinite radius to a curve of finite radius
A simple circular curve consists of one arc of constant radius R as Figure 4 (a). A simple circular
curve is described either by its radius, for example, 200m-radius curve, or by the degree of
the curve (D). D is the angle subtended at the centre of a 100 m arc. This is known as a degree
curve
Compound circular curves: these consist of two or more consecutive simple circular curves
of different radii without any intervening straight section. A typical two-curve compound
curve is shown in Figure 4 (b). The purpose of such curves is to avoid certain points, the
crossing of which would involve great expense and which cannot be avoided by a simple
circular curve.
Reverse circular curves: These curves consist of two consecutive curves of the same or
different radii without any intervening straight section and with their centres of curvature
falling on opposite sides of their common tangent point. They are much more common than
compound circular curves and, like them, can be used to avoid obstacles. More usually,
however, they are used to connect two straights which are nearly parallel and that would
otherwise require a very long simple circular curve
Arc VW = 100 m and subtends an angle of D° at the centre of curvature O. The curve TU is,
therefore, a D° degree curve. The relationship between radius curves and degree curves is
given by the formula:
= 423.03m
P.C= P.I – TL = (400 + 644.75) – 229.51 = 400+ 415.24
PT = P.C + LC = (400+ 415.24 ) + 423.03 = 400+ 838.27
Exercise 4
The two tangents 1A and AB are to be joined by a simple horizontal 6 ° curve. Determine
i) Radius of the curve
ii) Length of the curve
iii) Tangent length AB
Solution
= 506.75 m
4.2 Super-elevation
When a vehicle is moving around a circular curve at constant speed v, it experiences a radial
𝑣2 𝑚𝑣 2
acceleration a = and a radial force acting on the vehicle F = usually referred to as the
𝑅 𝑅
centrifugal force. There is also an outward radial force acting toward the centre of curvature
as a result of the centripetal acceleration. This radial/lateral force is non-existent on a straight
highway section.
The radial forces on a vehicle moving on a horizontal curve is sustained by the friction
between the tires and the pavement. The friction force (F) acts along the cross slope of the
roadway in a perpendicular direction from the normal force.
To further counter-balancing the effect of the centripetal force, the road is inclined toward
the centre of the curve. This inclination of the roadway toward the centre of the curve is
known as super-elevation.
All forces acting on the vehicle must be in equilibrium for the vehicle to resist the tendency
to slide up or down the pavement while travelling through the curve.
Figure 9: Forces acting on a vehicle moving around a horizontal curve with super-elevation
The friction force is equal to the side friction factor, µ, multiplied by the normal force (N).
Resolving forces and equating:
𝑚𝑣 2
𝑚𝑔𝑆𝑖𝑛(𝜃) + 𝐹 = 𝐶𝑜𝑠 (𝜃)
𝑅
𝐹 = 𝜇𝑁
𝑚𝑣 2
But 𝑁 = 𝑚𝑔𝐶𝑜𝑠 (𝜃) + 𝑆𝑖𝑛(𝜃)
𝑅
𝑚𝑣 2 𝑚𝑣 2
𝑚𝑔𝑆𝑖𝑛(𝜃) + µ{ 𝑚𝑔𝐶𝑜𝑠 (𝜃) + 𝑆𝑖𝑛(𝜃)} = 𝐶𝑜𝑠 (𝜃)
𝑅 𝑅
𝜇𝑣 2
The term tan(𝜃) can be ignored because it’s extremely small to get the expression
𝑅𝑔
𝑣2
tan(𝜃) + 𝜇 = 𝑅𝑔
If in addition we express velocity in kilometres per hour rather than metres per second, and
given that g equals 9.81m/s2, the following generally used equation is obtained
𝑣2
e + 𝜇 = 127𝑅
In the UK is assumed that, at the design speed, 55% of the centrifugal force is balanced by
friction, with the remaining 45% being counteracted by the cross-fall (e)
The KENYA ROAD DESIGN MANUAL Part 1 also provide guidance for minimum radii for wholly
circular curves based on a maximum super elevation of 6%) as shown below:
Figure 11: KENYA ROAD DESIGN MANUAL Part 1 guidance on minimum radii for Horizontal
curves (based on a super elevation of 6%)
4.3 Transition Curves
A vehicle travelling with a constant speed v along a curve of radius r is subjected to a radial
force F such that F= ma = mv2/R, where m is the mass of the vehicle. This force is, in effect,
trying to push the vehicle back on to a straight course
Circular curves have a limitation. A vehicle moving from a straight section to a circular curve
experiences the full radial force (F) instantaneously. This is a problem for the safety of the
passengers in the vehicle. If R is small and the vehicle is travelling too fast, the practical effect
of this is for the vehicle to skid sideways, away from the centre of curvature, as the full radial
force is applied. In severe cases the vehicle could overturn.
The purpose of transition curves is to permit the gradual introduction of centrifugal forces.
Such forces are required in order to cause a vehicle to move round a circular arc rather than
continue in a straight line.
If the transition curve is to introduce the radial force in a gradual and uniform manner, the
product of the radius of curvature at any point on the curve and the length of the curve up to
that point is a constant value i.e rl = K. This is the definition of a spiral, and because of this,
transition curves are also known as transition spirals or clothoids.
The rate at which the radial acceleration changes is a very important parameter and it can be
used to calculate the value of 𝑳𝑻 . It is known as the rate of change of radial acceleration (c)
and should be considered as a safety or comfort factor, the value of which has an upper limit
beyond which discomfort is too great
𝐯𝟐
Rate of change in radial acc. on circular curve = 𝐑⁄
(𝐭𝐢𝐦𝐞 𝐫𝐞𝐪𝐮𝐢𝐫𝐞𝐝 𝐭𝐨 𝐭𝐫𝐚𝐯𝐞𝐫𝐬𝐞 𝐭𝐫𝐚𝐧𝐬𝐢𝐭𝐢𝐨𝐧 𝐜𝐮𝐫𝐯𝐞)
𝑳𝑻
Time (t) to traverse Transition curve =
𝒗
𝒗𝟐 𝒗𝟑
∴ Rate of change of radial acc. = c = = ……………………………………..2
𝑹𝒕 𝑹𝑳𝑻
A maximum value for c is of 0.3 m/s3, although it may be necessary to increase this value up
to 0.6 m/s3 in difficult cases. In practice, wherever possible, transition curves with c values at
0.3 m/s3 should be used to ensure that they achieve the comfort requirement without being
excessively long
From equation 2 and assuming an acceptable value of c = 0.3 m/s3
It has been proved that B is the mid-point of the transition (see Bannister and Raymond,
1984 for details)
𝐿𝑇
𝐵𝑇 =
2
𝑳𝑻 𝟐⁄
𝑺= 𝟐𝟒𝑹
Exercise 5
A transition curve is required for a single carriageway road with a design speed of 100 km/hr.
The bearings of the two straights in question are 17° and 59°. Assuming a value of 0.3 m/s3
for C, calculate the following:
I. The transition length, L
II. The shift, S
III. The length along the tangent required from the intersection point to the start of
the transition, IT
Figure 14: UK guidance on choice of radii of horizontal curves and other design standards
Exercise 6
A sound wall is to be constructed at the edge of shoulder, along the inside of a horizontal
curve of an urban freeway. The inside lane is 3.8 m wide, with a shoulder of 1.20 m. The radius
of the curve measured up to the outer edge of the shoulder is 45 m.
i) Determine the sight distance of this section of the curve with the sound wall
ii) If the minimum sight stopping distance required is 50m, discuss the
options available to the engineer
Exercise 7
A 2-lane 7.3m wide single carriageway road is being designed to have a curve radius of
600m. The minimum sight stopping distance required is 160m.
i) Calculate the required distance to be kept clear of obstructions in metres
5. Vertical alignment
5.1 Introduction
In the same way that horizontal curves are used to connect intersecting straights in the
horizontal plane, vertical curves are used to connect intersecting straight-line gradients in
the vertical plane. The combination of these gradients and vertical curves is known as the
vertical alignment.
These vertical parabolic curves must therefore be provided at all changes in gradient to avoid
grounding, provide safe sight distances and for comfort as shown in the figure below
Figure 17: Reasons why vertical curves are necessary
Gradients are usually expressed as percentages, for instance, 1 in 50 = 2%, 1 in 25 =4%. The
road design manual provides guidance on the maximum gradients which are dependent on
the highway design speed and terrain. See the figure below:
In design calculations, the algebraic difference A between the gradients is used. Gradients
rising in the direction of increasing chainage are considered to be positive and those falling
are considered to be negative. This leads to six different combinations of gradient, which are
shown in the figure below:
In the above figure, chainage has is assumed to increase from left to right and various entry
gradients (m% values) are shown intersecting with various exit gradients (n% values). A is
obtained from:
A = (entry gradient %) – (exit gradient %)
There are two types of vertical curve, namely, crest curve and sag curve and are defined as
follows:
A crest curve, which can also be referred to as a summit or hogging curve, is one for which
the A is positive e.g (a), (b) and (f)
A sag curve, also referred to as a valley or sagging curve, is one for which A is negative, e.g
(c), (d) and (e)
The design of the vertical alignment involves the selection of suitable grades for the straight-
line sections and the appropriate length of vertical curves
5.2 Geometry of Vertical Curves
5.2.1 Circular Vertical Curves
θ=α+β
= m/100 + (-n)/100
= (m-n)/100
θ = A/100, Where A is algebraic difference in gradients
θ = Lv/R (radians) or Lv = Rθ
𝑚𝐿𝑣
Elevation at PVC = Elevation at PVI –
200
3∗ 500
= 1000 –
200
= 992.5 m
Point of vertical intersection (PVI): The location where the entrance grade, m%,
and the exit grade, n%, intersect.
Point of vertical curvature (PVC): The point where the vertical curve begins and
the highway leaves the tangent.
Point of vertical tangency (PVT): The point where the vertical curve ends and
the highway returns to the tangent.
Entrance grade (m%): The grade of the tangent leading into the vertical curve.
The grade is expressed as a percent.
Exit grade (n%): The tangent leading out of the vertical curve. The grade is
expressed as a percent
Tangent offset, y : the change in elevation from the tangent to the curve
If Y is taken as the elevation of the curve at a point x along the parabola, then:
…………………….1
Integrating Equation 1:
𝒅𝒀
= 𝒌𝒙 + 𝑪 ……………………………… 2
𝒅𝒙
i. When x = 0
𝑑𝑌
= m (m being the slope at the start of the vertical curve or the gradient of the first
𝑑𝑥
straight line)
∴ 𝑪 = m ……………………………………...3
ii. When x = L
𝑑𝑌
= 𝑛 (n being the slope of the second straight line gradient)
𝑑𝑥
n = kL + C = kL + m……………………….. 4
Rearranging Equation 3
(𝐧− 𝐦)
k = (n- m)/ L = …………………..5
𝑳
Substituting Equations 3 and 4 into Equation 2
𝒅𝒀 (𝐧−𝐦 )
= 𝒙+𝒎 ………………….6
𝒅𝒙 𝑳
Integrating Equation 6
(𝐧−𝐦) 𝟐
𝒀= 𝑳
𝒙 + 𝒎𝒙 …………………...7
From Figure 19,
(𝒚 + 𝒀)
𝒎= ………………….……..8
𝒙
Substituting Equation 8 into Equation 7
(𝐦− 𝐧) 𝟐
𝒀= 𝑳
𝒙 + (𝒚 + 𝒀)……………..9
(𝐧−𝐦) 𝟐
𝒚=− 𝟐𝑳
𝒙 = (𝐦−𝐧)
𝟐𝑳
𝒙𝟐 ………10
where x is the distance along the curve measured from the start of the vertical curve and y is
the change in elevation from the tangent to the curve
High/Low Point
Identifying and locating the high or low point of a vertical curve is important for drainage
considerations and for determining where the appropriate clearance is provided underneath an
overhead structure or on top of underground utilities
The high or low point, also known as the turning point, of a vertical curve occurs when the
slope of the highway is equal to zero.
Therefore, using basic calculus principles, the derivative of the parabolic equation can be used
to find the location of the high/low point
𝒅𝒀 (𝐧−𝐦 )
= 𝒙+𝒎 =0
𝒅𝒙 𝑳
𝑳𝒎
𝒙= ……………….. 12
𝒎−𝒏
𝑳𝒎𝟐
∴ 𝒚= …………...13
𝟐(𝒎−𝒏)
Quiz 9
A vertical alignment for a single carriageway road consists of a parabolic crest curve
connecting a straight-line uphill gradient of +4% with a straight-line downhill gradient of -3%.
Assuming an absolute minimum curve length of 210m for a crest curve, calculate:
i. Calculate the vertical offset at the point of intersection of the two tangents at PI
ii. Calculate the vertical and horizontal offsets for the highest point on the curve
Solution
Using Equation 11
(m−n) (0.04−(−0.03))×210
e= L= = 1. 8375 m
8 8
𝑳𝒎 𝟐𝟏𝟎(𝟎.𝟎𝟒)
𝒙= = = 120m
𝒎−𝒏 𝟎.𝟎𝟒+𝟎.𝟎𝟑
𝑳𝒎𝟐 𝟐𝟏𝟎×𝟎.𝟎𝟒𝟐
𝒚= = = 2.4m
𝟐(𝒎−𝒏) 𝟐(𝟎.𝟎𝟒+𝟎.𝟎𝟑)
(𝒚 + 𝒀)
m=
𝒙
Y = mx – y
=0.04(120) - 2.4
= 2.4m
Quiz 10
A vertical curve needs to be designed to connect the gradients -4.5% and - 1.0% of a Rural 2
lane single carriageway road. The design speed is 85km/h.
iv. Calculate the vertical offset at the point of intersection of the two tangents at PI
v. Calculate the elevations at the start and end of the vertical curve if the elevation at
the crest is 1650m
vi. Calculate the vertical and horizontal offsets for the highest point on the curve
vii. Compute the elevations of points at 10m intervals on the vertical curve
Quiz 11
Compare crash rates and severity on horizontal curves and straight sections of highways
Quiz 12
Define the term “design standards”.
Are design standards necessary? Why?
Do design standards change with time? Why?
Why do design standards vary between counties?