"Guide to Traffic Engineering Practice Part 1- Traffic Flow"

The guide discusses the interaction of the three main components of the road traffic system,
namely the driver, the vehicle and the environment.

"Guide to Traffic Engineering Practice Part 2 - Roadway Capacity"

This publication considers the issues that affect roadway design and provides a means of
determining the traffic carrying performance of a road or any element of it under the
prevailing roadway, traffic and control conditions.

"Guide to Traffic Engineering Practice Part 3· Traffic Studies"

Traffic studies show the traffic engineer how to perform traffic studies and use the collected
data in traffic planning, traffic management, economic studies, traffic and environmental
controls, road safety, land use/transport interaction studies, and in monitoring trends (eg, in
vehicle design and use).

Emphasis is placed on road/environment aspects of road crashes, with passing reference made
to the vehicle, driver and the socio-economic factors which are linked with most road crashes.

"Guide to Traffic Engineering Practice Part 5 . Intersections at Grade"

This guide explains the considerations and gives details of the procedures that should be
applied in designing intersections at grade for traffic and pedestrians in rural and urban areas.

Detailed guidelines are provided on the methods used to evaluate the traffic capacity and
delays at roundabouts, including analytical equations for calculating queuing and geometric
delays, degree of saturation and queue lengths.

"Guide to Traffic Engineering Practice Part 7 . Traffic Signals"

The publication provides detailed information on preliminary design, geometric

requirements, signal system components, signal displays and their location, signal phasing,
vehicle detection and coordination of traffic signals as well as the maintenance of traffic
signal systems.

"Guide to Traffic Engineering Practice Part 8· Traffic Control Devices"

Traffic Control Devices gives guidelines on the proper use of signs, traffic signals, pavement
markings, traffic islands and other devices placed or erected with the approval of a traffic
authority, to achieve safe regulation and guidance of traffic.
The guide shows how to get the most out of the existing road system by implementing
appropriate traffic management strategies.

The guide outlines the basic methodologies available for undertaking traffic management
studies and implementing recommendations, and also evaluating the effectiveness of traffic
management strategies.

"Guide to Traffic Engineering Practice Part 11 - Parking"

Basic parking surveys and methods of designating on street and off street parking are
described, as are parking issues relating to rural roads.

The guide shows how roadway lighting in its design, installation and maintenance can be
integrated into road design and traffic management practices, thus giving an overall product
that will meet traffic and community objectives.

"Guide to Traffic Engineering Practice Part 13 - Pedestrians"

The publications provides guidance on appropriate standards for walkways and footpaths, the
provision of pedestrian facilities for crossing roads, signing and other guidance methods.

"Guide to Traffic Engineering Practice Part 14 - Bicycles"

This publication stresses the importance of developing bicycle strategies at the national, state
and local government levels.

Lay, M.G., "Source Book for Australian Roads", Australian Road Research Board,
2nd ed., 1984.

Lay, M.G., "Handbook of Road Technology", Gordon and Breach Science Publishers,
two volumes, 1986.
Meyer, C.F., "Route Surveying and Design" 4th ed, International Text Book Company,
1969.

Underwood, R.T., "The Geometric Design of Roads", Macmillan Education Australia,

1991.
 Chapter 1
Traffic - Stream Characteristics
The prime purpose of roads is to provide for the safe, convenient, effective and
efficient movement of persons and goods. Thus, it is fundamental that the design of
roads be based on the behaviour of all road users, including pedestrians and bicyclists,
and on their expectations, and on the performance and geometric characteristics of
vehicles. At the same time, consideration must also be given to the whole range of
economic, social, environmental and other factors that may be involved.

In this section, various traffic considerations, including traffic volume, traffic flow
characteristics, flow analysis and queuing theory .

Traffic volume is one of the most basic considerations in the design of roads. Traffic
volume (or traffic flow) is defined as the number of vehicles, bicyclists or pedestrians
passing a given point during a specified period of time. It is usually expressed in
vehicles (or bicycles or pedestrians if appropriate) per hour, although for particular
purposes the time period may be per second or per day.

On any given section of road, the traffic volume varies from hour to hour, from day to
day and from season to season, the extent of the variation depending on factors such as
the function of the road, the nature of the traffic on it, whether it is in a rural or an
urban area and so on. Typical hourly, daily and seasonal traffic volume patterns are
shown in Figures 1, 2 and 3.

3000

.c
"-
.c
<l> 2000
>
----,
3:
0
.•...
u
.•...
.•...
0 1000
•...
'-
~ >.
'-
::l 600
0
I 400
200 "-

0 6 12 18 24
Time of Day

Figure 1 - Typical hour-by-hour variations in traffic flow on arterial

roads (source: Traffic Studies, 1988, NAASRA)
 ~ 20000
>-
a
"0

"~ 15000
>

3:
o //
••..10000 /"
,/
u
.•..
.•..a _ ./
/"--
Rural
•.. .......------"
->-
·0
5000
Q

MON TUE WED THU FRI SAT SUN

Day of ~eek

Figure 2 - Typical variations in traffic flow over days of the week

on arterial roads (source: Traffic Studies, 1988, NAASRA)

16
r-....... '\
15 I " 1\
3; 14 I \'vl \ /'
_o 13 I \ / \ Rural
I \...-- \
.~
••.... 12 I \
••....
~ 11
I \
.•.... \
>.10 \
\
-E 9
c
o 8
~
J FMAMJJASOND

Figure 3 - Typical variations in traffic flow on arterial roads over months

of the year (source: Traffic Studies, 1988, NAASRA)

Annual Average Daily Traffic

The annual average daily traffic volume (AADT) is the total traffic throughout the year
divided by 365. It is thus an average 24 hour daily traffic volume. The AADT volume
is suitable as a basis for planning and investigation to do with economic studies, and
for determining pavement and shoulder widths on undivided rural roads, but for most
design purposes a design volume based on an hourly volume rate is more appropriate.

As already indicated, hourly traffic volumes vary throughout the day, and in addition
they may vary considerably from day to day throughout the year. If the design of a
road was based on the highest hourly traffic volume in the year, the road would be
under-utili sed for all other hours of the year, and usually could not be justified.
Accordingly, one of two design hour volumes is often used for design purposes,
namely:-

1. the 30th highest hour volume;

2. the week day peak hour volume.
If the highest hourly volumes throughout the year are plotted, the result is usually along
the lines shown in Figure 4. The 30th highest hour is often selected for design
purposes, particularly for the higher volume rural arterial roads and for rural and urban
freeways.

% age
AADT 30

Partially recreational route

Secondary rural route
Main rural route
Suburban route
Urban arterial
200

Figure 4 - Typical relationship between hourly volumes and

AADT (source: Traffic Studies, 1988, NAASRA)

In urban areas, and particularly for urban street and intersection design purposes, an
alternative to the 30th highest hour volume is to use the normal week day peak hour
volume.

The ratio of the design hour volume expressed as a percentage of the AADT is known
as the design hour volume factor, and is commonly designated K.

The composition of traffic, i.e. the proportions of various types of vehicles in the traffic
stream, is an important design consideration. For most purposes, the traffic
composition is expressed as a percentage of trucks (T) (or heavy vehicles) in the traffic
stream, although in some cases a more detailed classification of different types of
vehicles may be appropriate. The value of T can vary considerably. Typical values are
of the order of 10 percent on rural arterial roads, but they may range from close to zero
on local residential streets to up to 30 percent or more on heavily trafficked industrial
roads.
On one- and two- lane undivided rural roads, the design volume is normally expressed
in terms of the two-way volume. However, on multi-lane roads, at intersections and in
some other cases such as on two-lane rural roads when climbing lanes are being
considered, the peak hour one-way volume may be more appropriate for design
purposes. This one-way volume, expressed as a percentage of the two-way volume, is
known as the directional split, and is commonly designated D.

The design of roads is usually based on the traffic expected to use them over their
design life. This requires the selection of a design year, and the estimation of the traffic
likely to use the facility in that year. It is extremely difficult to estimate realistically
future traffic volumes too far ahead and for many design purposes, a design year 20
years hence for pavements and 30 years hence for bridges is chosen.

Fundamental characteristics of traffic flow that are relevant to the design of roads
include ;-

1. Speed, which is the distance travelled per unit time, usually expressed in m/s or
km/h.

2. Density (or concentration), which is the number of vehicles per unit length of
roadway at a given instant, usually expressed in veh/km, although for particular
purposes, veh/m may be used.

3. Headway (or gap), which is the time interval between two consecutive vehicles
passing a given point, usually expressed in seconds.

4. Spacing, which is the distance between the front bumpers (or the front wheels)
of two consecutive vehicles at a given instant, usually expressed in metres.

In a stream of free flowing traffic on an ideal road all drivers can travel at their own
desired speed. Under these conditions, traffic speeds (or free speeds) are close to
being normally distributed, and can be represented by a cumulative normal distribution
curve as in Figure 5. As the volume of traffic, (V, veh/h) on a road increases, drivers
become increasingly affected by other vehicles.
 100

90
w 80
CJ
~ 70
Z
w
()
a: 60
w
0..
W 50
>
~ 40
-I
:)
:2 30
:)
()
20

10

80 90 100 110 120

SPEED Km/h

Figure 5 - Cummulative speed distribution for a two-lane

rural road (source: Traffic Studies, 1988, NAASRA)

1. By measuring the speed of each vehicle passing a given point on a road during a
selected time period. The arithmetic mean of the spot speeds so obtained is
called the time mean speed.

2. By measuring the speed of all vehicles on a given length of road at a given

instant. The arithmetic mean of the speeds so obtained is called the space mean
speed.

It turns out that the time mean speed is always slightly greater than the space mean
speed, unless all vehicles are travelling at the same speed, in which case the two means
are the same. The time and space mean speeds are related approximately as follows:-

- 2
Vt = V- s + -_-
<Js
( J. J )
Vs

where Vt = the time mean speed;

v s = the space mean speed; and
<Js = the standard deviation of the space speeds.
It is usually easier to measure spot speeds at a point on a roadway than to measure the speeds of
vehicles on a section of roadway at a given instant. For a set of spot speeds, the time mean speed
is given by :-
n

L(vI +v2 +...vn)

I

It is impossible to calculate the space mean speed from these same spot speeds by converting them
to travel times per unit distance, calculating the arithmetic mean of these travel times, and then
converting this mean travel time to a mean speed. Thus,

_ n
Vs= nil 1
L-+-+···-
I VI V2 Vn

The distinction between time and space mean speeds is important because the following
relationships apply:-

1. volume (veh/hr) = space mean speed (kmIh) x density (veh/km)

2. volume (veh/hr) = 3600 / mean headway (seconds)
3. volume (veh/hr) = 1000 x space mean speed (km/h) / mean spacing (m)
4. density (veh/km) = 1000/ mean spacing (m)

A simple identity provides the basic relationship among traffic flow, speed (space-mean speed),
and density:

q=uk

with typical units of flow (q), speed (u), and density (k) being vehicles per hour (veh/hr),
kilometres per hour (km/h), and vehicles per kilometre (veh/km), respectively.

The most intuitive starting point for developing a consistent, generalized traffic model is to focus on
the relationship between speed and density. To begin, consider a section of highway with only a
single vehicle on it. Under these conditions, the density will be very low (veh/km) and the driver
will be able to travel freely at a speed close to the design speed of the highway. This speed is
referred to as the free-flow speed (denoted here as uf) because vehicle speed is not inhibited by the
presence of other vehicles. As more and more vehicles begin to use a section of highway, the
traffic density will increase and the average operating speed of vehicles will decline from the free-
flow value as drivers slow to allow for the maneuvers of other vehicles. Eventually, the highway
section will become so congested (ie. will have such a high density) that the traffic will come to a
stop (u = 0) and the density will be determined by the length of the vehicles and the spaces that
drivers leave between them. This high-density condition is referred to as the jam density, kj"
One possible representation of the process just described is the linear relationship shown in Fig. 6.
Mathematically, such a relationship can be expressed as

However, it is important to note that field studies have shown that the speed-density relationship
tends to be nonlinear at low densities and high densities (ie. those that approach the jam density).

"
-.(

? UJ.
.Ii
v
~
~
J.
~ Kj
o ...,.
:!:>""'-s i11 (" e (.../ \("",,)

Using the assumption of a linear speed-density relationship as shown in Eqn 1.5, a parabolic flow-
density model can be obtained by substituting Eqn 1.5 into Eqn 1.4:

The general form of Eqn 1.6 is shown in Fig. 7. Note that in this figure the maximum flow rate,
qm' represents the highest rate of traffic flow that the highway is capable of handling. This is
referred to as the traffic flow at capacity or simply the capacity of the highway. The traffic density
that corresponds to this maximum flow rate is km, and the corresponding speed is um. Equations
for qm' km and um can be derived by differentiating Eqn 1.6 because at maximum flow,

dq = U
dk f
(1- 2k) =
k.
J
0

k k.
=-l..
m 2
 =!!:t.
2

=!!h
4

Returning to the linear speed-density model (Eqn 1.5), a corresponding speed-flow model can be
developed by rearranging Eqn 1.5 to
 --71
/ I
I
I k"""
I I
I
I
I
--~--
I
I

1<...... rlow
l:>.~,,,,,.s;T1 (v •.\....(k,,,,,,) (\i e l../\,.,,)

Figure 9. Flow-density, speed-density, and speed-flow relationships

(assuming a linear speed-density model).

The speed-flow model defined by Eqn 1.12 again gives a parabolic function as shown in Fig. 8.
Note that Fig. 8 shows that two speeds are possible for flows, q, up to the highway's capacity qm
(this follows from the two densities possible for given flows as shown in Fig. 7).

All of the flow, speed, and density relationships and their interactions, as we have discussed, are
graphically represented in Fig. 9.
There is considerable analytic value in modelling the time between the arrivals of successive
vehicles (ie. the notion of vehicle time headways presented earlier). The most simplistic approach
to vehicle arrival modelling is to assume that all vehicles are equally or uniformly spaced. This
results in what is termed a deterministic, uniform arrival pattern. However, actual observations
show that such uniformity of traffic flow is not always realistic. Thus a representation of vehicle
arrivals that goes beyond the deterministic, uniform assumption is often warranted.

Models that account for the nonuniformity of flow are derived by assuming that the pattern of
vehicle arrivals corresponds to some random process. An example of such a distribution is the
Poisson distribution (the limitations of which will be discussed later), which is expressed as

P(n) = (ltYe - At
n!

t = the duration of the time interval over which vehicles are counted
P(n) = the probability of having n vehicles arrive in time t, and
A is the average vehicle flow or arrival rate (in vehicles per unit time).

The assumption of Poisson vehicle arrivals also implies a distribution of the time intervals between
the arrivals of successive vehicles (ie. time headway). To show this, let the average arrival rate, Il,
be in units of vehicles per second so that

Il=-q-
3600

(qt I 3600Ye - qr /3600

P(n) = ------
n!
Note that the probability of having no vehicles arrive in a time interval of length t (ie. P(O)) is
equivalent to the probability of a vehicle headway, h, being greater than or equal to the time interval
t. So from Eqnl.15,

This distribution of vehicle head ways is known as the negative exponential distribution and is often
simply referred to as the exponential distribution (see Fig.JO).

- 'ttjs
I,~ .-0 bOO

~ .•. o.g P ( h >,.. ~) - e

~ 0"
<J'
0'4
'"
~
~ 0·1-
~
0 5 Ie IS" 2..0 2.S" '"'36

T,~~, -c
Figure 10. Exponentially distributed probabilities of headways
greater than or equal to t, with q = 360 veh/h.

e - A/(1 + At)
the average length of headways ;;:::t = A/
Ae-
 (l + At)
A
1
=-+t
A

1 te - Ju
the average length of headways < t = 1 - _ Ju
/l, l-e

1
of course, the average length of all headways = A

In many traffic situations, vehicles or pedestrians crossing through a major traffic stream must
select a gap in which to perform the desired crossing. Assuming that the negative exponential
distribution reasonably applies to headways in the major traffic stream, the following equations can
be derived for the case ofrninor stream vehicles (or pedestrians) crossing through a major stream :-

T
the average delay to delayed vehicles or pedestrians = Ae ~ AT -AT
I -e

the average delay to all vehicles or pedestrians = Ae ~ AT ~ - T

where T is the critical gap, or average minimum acceptable headway in seconds.

If a queue of minor road vehicles desires to enter or cross a major traffic stream, the first vehicles in
the queue will require the headway T seconds, but successive vehicles will normally have a smaller
follow-up headway, To seconds. In this situation, the total number of vehicles that can enter or
cross the main road stream

The formation of traffic queues in congested periods is a source of considerable time delay and
results in the loss of highway performance. Given this, it is essential in traffic analysis to develop
a clear understanding of the characteristics of queue formation and dissipation along with
mathematical formulations that can predict queuing-related elements.

The purpose of traffic queuing models is to provide a means to estimate important measures of
highway performance including vehicle delay and traffic queue lengths. Such estimates are critical
to highway design (eg. the required length of right -turning bays and the number of lanes at
intersections) and traffic operations control including the timing of traffic signals at intersections.

Queuing models are derived from underlying assumptions regarding arrival patterns, departure
characteristics, and queue disciplines. The previous section explored traffic arrival patterns where,
given an average vehicle arrival rate, two possible distributions of the time between the arrival of
successive vehicles were considered: (1) equal time intervals (derived from the assumption of
uniform, deterministic arrivals) and (2) exponentially distributed time intervals (derived from the
assumption of Poisson-distributed arrivals). In addition to vehicle arrival assumptions, the
derivation of traffic queuing models requires assumptions relating to vehicle departure
characteristics. Of particular interest is the distribution of the amount of time it takes a vehicle to
depart; for example, the time to pass through an intersection at the beginning of a green signal, the
time required to pay a toll at a toll booth, or the time a driver takes before deciding to proceed after
stopping at a stop sign. As was the case for arrival patterns discussed previously, given an average
vehicle departure rate, the assumption of a deterministic or exponential distribution of departure
times is appropriate. Another important aspect of queuing models is the number of available
departure channels. For most traffic applications only one departure channel will exist, such as a
highway lane or group of lanes passing through an intersection. However, multiple departure
channels are encountered in some traffic applications, such as at toll booths at entrances to
turnpikes or bridges.

Two options have been popularized in the development of queuing models: fIrst-in-fIrst-out
(FIFO), indicating that the first vehicle to arrive is the first vehicle to depart, and last-in-first-out
(UFO), indicating that the last vehicle to arrive is the first to depart. For virtually all traffic-
oriented queues, the FIFO queuing discipline is the more realistic of the two.

Queuing models are typically identified by three alphanumeric values: the first value indicates the
arrival rate assumption; the second value gives the departure rate assumption; and the third value
indicates the number of departure channels. For traffic arrival and departure assumptions, the
uniform, deterministic distribution is denoted D, and the exponential distribution is denoted M.
Thus a DID/] queuing model assumes deterministic arrivals and departures with one departure
channel. Similarly, an WD/] queuing model assumes exponentially distributed arrival times,
deterministic departure times, and one departure channel.

The case of deterministic arrivals and departures with one departure channel (DID/] queue) is an
excellent starting point in understanding queuing models because of its simplicity. The DID/]
queue lends itself to an intuitive graphical or mathematical solution.

The assumption of exponentially distributed times between the arrivals of successive vehicles (ie.
Poisson arrivals) will, in some cases, give a more realistic representation of traffic flow than the
assumption of uniformly distributed arrival times. Therefore, the MID/] queue (exponentially
distributed arrivals, deterministic departures, and one departure channel) has some important
applications within the traffic analysis field. Although a graphical solution to an MID/] queue is
difficult, a mathematical solution is straightforward. Defining a traffic intensity term, p, as the ratio
of average arrival to departure rates (AlJ.l), and assuming that p is less than 1, it can be shown that
 _ p2
Q=--
2(1- p)

w= p
2.u(1- p)

The average time spent in the system (ie. the summation of average queue waiting time and average
departure time, or, as it is frequently referred to, service time) is

t=_2_-_p_
2.u(1- p)

It is important to note that under the assumption that traffic intensity (P) is less than 1 (ie. A < .u),
the DIDII queue will predict no queue formation. However a queuing model that is derived based
on random arrivals or departures, such as the MlDII queuing model, will predict queue formations
under such conditions. Also, note that the MlDII queuing model we have presented is based on
steady-state conditions (ie. constant average arrival and departure rates) with randomness arising
from the assumed probability distribution of arrivals.

A queuing model that assumes one departure channel and exponentially distributed departure time
patterns in addition to exponentially distributed arrival times (ie. anMlM/I queue) is also useful in
some traffic applications. For example, exponentially distributed departure patterns might be a
reasonable assumption at a toll booth where some arriving drivers have the correct toll and can be
processed quickly, and others may not have the correct toll, thus producing a distribution of
departures about some mean departure rate. Under standard MIMII assumptions, it can be shown
that the average length of queue (in vehicles), again assuming that p is less than 1, is

_ p2
Q=-
1-p

A
w=---
.u(.u - A)

The average time spent in the system (the summation of average queue waiting time and average
departure time) is

- 1
t=--
.u-A
A more general formulation of the MIMI] queue is the MlMIN queue where N is the total number of
departure channels. MIMIN queuing is a reasonable assumption at toll booths entering turnpikes or
at toll bridges where there is often more than one departure channel available (ie. more than one toll
booth open). A parking lot is another example, with N being the number of parking stalls in the lot
and the departure rate, Ji, being the exponentially distributed times of parking duration. MIMIN
queuing is also frequently encountered in nontransportation applications such as checkout lines at
retail stores, security checks at airports, and so on.

The following equations describe the operational characteristics of MIMIN queuing. Note that,
unlike the equations for MIDI] and MIMI], which require traffic intensity, p, to be less than 1, the
following equations allow p to be greater than 1 but apply only when pIN (which is called the
utilisation factor) is less than 1.

1
P 0---------
N-l pnc pN
L-+---
n ! N!(l- piN)
nc=O c

p = pnPo for n ;:::N

n ~-NN!

- PopN+l[ 1 ]
Q = N!N (1- pi N)2

- p+Q
t=--
Il

_ p+Q 1
w=----
Il Ji
The probability of waiting in a queue (the probability of being in a queue, which is the probability
that the number of vehicles in the system, n, is greater than the number of departure channels, N) is

P P 0 pN+l _

n>N- N!N(l-p/N)

Some of the most severe congestion problems occur at highway bottlenecks, which can be
generally defined as a portion of highway with a lower capacity (qm) than the incoming section of
highway. This reduction in capacity can originate from a number of sources including a decrease in
the number of highway lanes and reduced shoulder widths. There are two general types of traffic
bottlenecks - those that are recurring, and those that are incident-induced. Recurring bottlenecks
exist where the highway itself limits capacity; for example, by a physical reduction in the number of
lanes. Traffic congestion at such bottlenecks results from recurring traffic flows that exceed the
vehicle capacity of the highway in the bottleneck area. In contrast, incident-induced bottlenecks
occur as a result of vehicle breakdowns or accidents that effectively reduce highway capacity by
restricting the through movement of traffic.

The most intuitive approach from which traffic congestion at bottlenecks can be analysed is to
assume DID!] queuing.
 TYPICAL EXAMPLES (for volume, speed and density)

3600
volume (veh/hr) = --------
mean headway (seconds)

3600
=
6

If the space mean speed of the traffic in problem 1 is 80 km/hr, what is the density?

Solution:

An aerial photograph of a 0.4 krn section of road shows that there are 10 vehicles on it. What is the
density of traffic?

Solution:

density = 1.0 x 10 = 25 veh/km

0.4
A peak hour traffic speed and volume count on a typical day on a 1.6 km section of a semi-rural
highway yielded the following results:

175 vehicles travelling at 40 km/h

279 vehicles travelling at 45 km/h
349 vehicles travelling at 70 km/h
192 vehicles travelling at 90 km/h
44 vehicles travelling at 100 km/h

Find: (i) A.A.D.T.

(ii) 85th percentile speed
(iii) time mean speed
(iv) space mean speed

(i) peak hour total = 175 + 279 + 349 + 192 + 44 = 1039 vph
AADT = 1039 x 100
15
(15% - peak hour volume is usually 8 - 15% of total daily
flow or 2 - 2.5 times the average hourly volume)

= 6927 vpd

175 = 16.8% 16.8% 40km/h

279 = 26.8% 43.7% 45km/h
349 = 33.6% 77.3% 70km/h
192 = 18.5% 95.8% 90km/h
44 = 4.3% 100.0% 100km/h

Note: Speed environment is numerically equivalent to the desired speed of the 85th
percentile on the section of road. That is, it is the speed at or below which 85% of drivers
would normally travel on the particular section of road.
I

L(v1 + v2 +...vn)

175 x40 + 279 x45+ 349 x 70+ 192x 90+44 x 100

=------------------
1039
 v ,..•
'"'
(~ '(0
! "5'c.

-
V
1-
;Lc.

~" '0
50
-J
> +0
~ '30

-<l
::l
£
.2.0

lo
:)

" 0

_ n
s=n(ll 1)
V
L I
-+-+
VI
...
+-
V2 Vn

1039
=-----------
175 279 349 192 44
-+-+-+-+-
40 45 70 40 100
= 57.3 krn/h

A section of highway is known to have a free-flow speed of 90 krn/h and a capacity of 3300 veh/h.
In a given hour, 2100 vehicles were counted at a specified point along this highway section. If the
linear speed-density relationship shown in Eqn 1.5 applies, what would you estimate the space-
mean speed of the 2100 vehicles to be?
 k. = 4qm
} U
f

4x3300
90

= 146.67 veh/km

k. 2
....l..u-k.u+q=O
Uf }

146.67 u2 -146.67u + 2100 =0

90

An engineer counts 360 vehlh at a specific highway location. Assuming that the arrival of vehicles
at this highway location is Poisson distributed, estimate the probabilities of having 0, 1,2, 3,4
and 5 or more vehicles arriving over a 20-second interval.

The average arrival rate, A, is 360 vehlh or 0.1 vehicles per second (vehis). Using this in Eqn 1.13
with t = 20 seconds, we find the probabilities of having 0, 1, 2, 3 and 4 vehicles are

Ix 20)0 - 0.1(20)
P(O) =. (0 e = 0.135
O!
(0 1x 20Ye - 0.1(20)
P(I) = . = 0.271
I!
(0 1x 20)2e - 0.1(20)
P(2) = . = 0.271
2!
(0 1 X 20)3e - 0.1(20)
P(3) = . = 0.180
3!
 (0 1x 20)4e - 0.1(20)
P(4) = . = 0.090
4!

(O•. I~O)

LO'I'3f'

(0 '0'10)

Co. 05

of ye/ .•• ·c,/e.,.s ,""'1"'\";".;';'" 0.. ••.

Se.e::. '.'_c.. ;",7 .••~ .••,' (p,.. •.!:.c..I.:J.hi.cS

P(5) = 1 - P(n < 5)

= 1 - 0.135 - 0.271 - 0.271 - 0.180 - 0.090

= 0.053

Traffic data are collected in fifteen 60-second intervals at a specific highway location as shown in
Table 7.1. Assuming the traffic is Poisson distributed and continues to arrive at the same rate as
that observed in the first fifteen time periods, what is the probability that six or more vehicles will
arrive in each of the next three 60-second time intervals (ie. 12.15 pm to 12.16 pm, 12.16 pm to
12.17 pm and 12.17 pm to 12.18 pm)?
 Observed number of
Time Period vehicles
12.00 - 12.01 3
12.01 - 12.02 5
12.02 - 12.03 4
12.03 - 12.04 10
12.04 - 12.05 7
12.05 - 12.06 4
12.06 - 12.07 8
12.07 - 12.08 11
12.08 - 12.09 9
12.09 - 12.10 5
12.10 - 12.11 3
12.11 - 12.12 10
12.12 - 12.13 9
12.13 - 12.14 7
12.14 - 12.15 6

Table 7.1 shows that a total of 101 vehicles arrive in the 15 minute period from 12.00 pm to 12.15
pm. Thus the average arrival rate, A, is 0.112 veh/s (1011900). As in Example 6, Eqn 1.13 is
applied to find the probabilities of exactly 0, 1, 2, 3, 4, and 5 vehicles arriving. With A = 0.112
veh/s and t = 60 seconds (At = 6.733), the application of Eqn 1.13 gives probabilities 0.0012,
0.008,0.027,0.0606,0.102 and 0.137 for having 0, 1,2, 3,4 and 5 vehicles, respectively,
arriving in a 60-second time interval. The summation of these probabilities (0.3358) is the
probability that 0 to 5 vehicles will arrive in any given 60-second time interval (ie. Pen ~ 5), so 1
minus 0.3358 (0.6642) is the probability that 6 or more vehicles will arrive in three successive time
intervals (i's) is simply the product of probabilities:
3

P (n ;::::6) for three successive time periods = TI Pen ;::::6)

i=!

= (0.6642)3
= 0.293

Consider the traffic situation in Example 6 (ie. 360 veh/h). Again assume that the vehicle arrivals
are Poisson distributed. What is the probability that the gap between successive vehicles will be
less than 8 seconds, and what is the probability that the gap between successive vehicles will be
between 8 and 10 seconds?
By definition, P (h < t) = 1 - P (h;::: t). Therefore Eqn 1.16 gives the probability that the gap will
be less than 8 seconds as

To determine the probability that the gap will be between 8 and 10 seconds, we compute the
probability that the gap will be greater than or equal to 10 seconds as

A drive-in bottle shop is to be erected at a mid-block position on a busy two-lane, two-way

roadway. The peak traffic volume in the lane adjacent to the development is 700 veh/hr and 500
veh/hr in the opposite direction. Traffic in and out of the development in the hour coinciding with
the peak traffic volume given above is 160 vehicles (80 in and 80 out).

(i) the probability that a departing vehicle will suffer delay;

(ii) the average delay to all departing vehicles;
(iii) the average delay to entering vehicles; and
(iv) the effect of the traffic entering and having the development on the capacity of the frontage
roadway

go ouT

---- -- ---~~-- -- -0;-

l.
L
R L-
Assume that left turning and right turning movements are in the same proportion as the directional
flows.

7
:.IL = 80 x - = 47 = 0L
12

5
IR = 80x- = 33= OR
12

A. = --700 = 0.194 veh/s

3600

1-p(gap>4sec)= 1-0.46

= 0.54

A. = 700 + 500 = 0.333 veh/s

3600
 and P (delay to departing vehicles) = 0.54 x ~ + 0.81 ~
12 12

1 1
= ---------4
0.194e - 0.194 x 4 0.194

1 1
---5
= 0.333e - 1.67 0.333

(iii) Average delay to entering vehicles:

(a) left-turn entries - unless the entrance is blocked either by a queue of preceding
vehicles, pedestrians or some other object, there is, in general, no delay to vehicles
entering by the left -turn manoeuver.

1 1
= 0.194e - 0.194 x 3.5
---3.5
0.194

Vehicles arrive at an entrance to a recreational park. There is a single gate (at which all vehicles
must stop), where a park attendant distributes free brochures. The park opens at 8.00 am, at which
time vehicles begin to arrive at a rate of 480 veh/h. After 20 minutes, the arrival flow rate declines
to 120 veh/h and continues at that level for the remainder of the day. If the time required to
distribute the brochure is 15 seconds, and assuming DID/1 queuing, describe the operational
characteristics of the queue.

)l = 480 veh/h = 8 vehlmin

60minJh

)l = 120 veh/h = 2 vehlmin

60 minJh

60 s/min
J1 = --- = 4 ve himm.
15 s/veh

Equations for the total number of vehicles that have arrived and departed up to a specified time, t,
can now be written. Define t as the number of minutes after the start of the queuing process (in this
case the number of minutes after 8.00 am). The total number of vehicle arrivals at time t is equal to

These equations can be illustrated graphically as shown in Fig. EIO. When the arrival curve is
above the departure curve, a queue condition will exist. The point at which the arrival curve falls
below the departure curve is the moment when the queue dissipates. In this example, the point of
queue dissipation can be determined graphically by inspection of Fig. EIO or analytically by
equating appropriate arrival and departure equations; that is,

Solving for t gives t = 60 minutes. Thus the queue that began to form at 8.00 am will dissipate 60
minutes later (9.00 am), at which time 240 vehicles will have arrived and departed (ie. 4 vehlmin x
60 min).

Another aspect of interest is individual vehicle delay. Under the assumption of a FIFO queuing
discipline, the delay of an individual vehicle is given by the horizontal distance between arrival and
departure curves starting from the time of the vehicle's arrival in the queue. So, by inspection of
Fig ElO, the 160th vehicle to arrive will have the longest delay of 20 minutes (the longest
horizontal distance between arrival and departure curves), and vehicles arriving after the 239th
vehicle will encounter no queue delay because the queue will have dissipated and the departure rate
will continue to exceed the arrival rate. It follows that with the LIFO queuing discipline, the first
vehicle to arrive will have to wait until the entire queue clears (ie. 60 minutes of delay).
 'Po;"'" £> f..
'tve. cJ e
cl;~si p••.
,.•.••..

'De1t J lOo"ti... ~c.\~

"'e •..•. -to o.••"', ...•~

"e"',~I"C.'t"vtU~ (20( 80)

00 40 ~o
T\'vv..e (1'V\.,~r")

Figure ElO. DID/l queuing diagram

The total length of the queue at a specified time, expressed by the number of vehicles, is given by
the vertical distance between arrival and departure curves at that time. For example, at 10 minutes
after the start of the queuing process (8.10 am) the queue is 40 vehicles long, and the longest
queue (longest vertical distance between arrival and departure curves) will occur at t = 20 minutes
and is 80 vehicles long (see Fig. ElO).

Total vehicle delay, defined as the summation of the delays of each individual vehicle, is given by
the total area between arrival and departure curves (see Fig. ElO) and, in this case, is in units of
vehicle-minutes. In this example, the areas between arrival and departure curves can be detennined
by summing triangular areas, giving total delay, Dr' as

1 1
Dr = -(80 X 20) + -(80 X 40)
2 2

Finally, because 240 vehicles encounter queuing delay (as previously detennined), the average
delay per vehicle is 10 minutes (2400 veh-min/240 veh).
After observing arrivals and departures at a highway toll booth over a 60-min period, an observer
notes that the arrival and departure rates (or service rates) are deterministic but, instead of being
uniform, change over time according to a known function. The arrival rate is given by the function
Aft) = 2.2 + 0.17t - 0.0032r, and the departure rate is given by /l(t) = 1.2 + O.Olt, where t is in
minutes after the beginning of the observation period and Aft) and J1{ t) are in vehicles per minute.
Determine the total vehicle delay at the toll booth and the longest queue assuming DIDII queuing.

Note that this problem is an example of a time-dependent deterministic queue because the
deterministic arrival and departure rates change over time. We begin by solving the time-to-queue
dissipation by equating vehicle arrivals and departures:

s; 2.2 + O.17t - 0.0032t dt 2

= S;1.2 + 0.07tdt
2.2t + 0.08Sr - 0.00107f = 1.2t + 0.03Sr

- 0.00107f + O.OSr + t = 0

which gives t = 61.8 minutes. Therefore, the total vehicle delay (which is the area between the
arrival and departure functions) is

r61.8 2 3 r61.8 2
Dt = Jo2.2t + 0.08St - 0.00107 t dt - Jr2t + 0.03St dt

3 161.8
= 1.1t 2
+ 0.0283t - 0.000267St4 - 0.6t2 - 0.0117t3 0

Q(t)= S;2.2+ 0.17t-0.0032t2dt- S;1.2 + 0.07t dt

= -0.00107f+ O.OSr+ t

dQ(t) = -0.0032lt2 + O.lt + 1 = 0

dt
 39.12
Q (39.12) = -0.00107f + 0.05r + t
o

= -0.00107 (39.12)3 + 0.05 (39.12)2 + 39.12

= 51.58 veh

Consider the entrance to the recreational park described in Example 10. However, let the average
arrival flow rate be 180 veh/h and Poisson distributed (exponential times between arrivals) over the
entire period from park opening time (8.00 am) until closing at dusk. Compute the average length
of queue (in vehicles), average waiting time in queue, and average time spent in the system,
assuming MJD/1 queuing.

Solution:

Il = _18_0_v_e_h/h_
= 3 vehlmin
60 minlh

60 s/min 4 hi .
p=--- = ve mIll
15 s/veh

Il 3
p=-=-=0.75
P 4

0.752
Q = 2(1- 0.75)

0.75
w=-----
2( 4)(1- 0.75)
 - 2-0.75
t=-----
2( 4)(1- 0.75)

Alternatively, because the departure time (or service time) is l/Ji (ie. the 0.25 min it takes the park
attendant to distribute the brochure),

- _ 1
t =w+-
Ji

1
= 0.375 +-
4

Assume that the park attendant in Examples 10 and 12 takes an average of 15 seconds to distribute
brochures, but that the distribution time varies depending on whether or not park patrons have
questions relating to park operating policies. Given an average arrival rate of 180 veh/h as in
Example 12, compute the average length of queue (in vehicles), average waiting time in queue, and
average time spent in the system, assuming MIMII queuing.

Using the average arrival rate, departure rate and traffic intensity as determined in Example 7, Eqn
1.33 gives the average length of queue (in vehicles) as

2
Q = 0.75
1-0.75

3
w=---
4(4 - 3)
 _ 1
t=--
4-3

At an entrance to a toll bridge, four toll booths are open. Vehicles arrive at the bridge at an average
rate of 1200 veh/h, and, at the booths, drivers taken an average of 10 seconds to pay their tolls.
Both the arrival and departure rate can be assumed to be exponentially distributed. How would the
average queue length, time in the system, and probability of waiting in a queue change if a fifth toll
booth was opened?

Using the equations for MIM/N queuing, we first compute the four-booth case. Note that J1 = 6
veh/rnin, A = 20 veh/rnin, and therefore p = 3.333. Also, because pIN = 0.833 (which is less than
1), Eqns,036tol.42can be used. The probability of having no vehicles in the system with four
booths open (using Eqnl.~1.)is

1
2
Po = 1 3.333 3.333 3.3333 3.3334
+--+---+--+----
I! 2! 3! 4!(0.1667)

= 0.0213

Q = 0.213(3.333/ [ 1 ]
4!4 (0.1667)2

_ 3.333 +3.287
t=-----
20

P = 0.0213(3.333)5
n > N 4!4(0.1667)

= 0.548
 1
Po = 1 3.333 3.3332 3.3333 3.3334 3.3335
+--+--+--+--+----
1! 2! 3! 4! 5!(0.1667)

Q = 0.213(3.333)6 [ 1 ]
5!5 (0.1667)2

- 3.333 + 1.967
t=-----
20
= 0.265 min

P = 0.0239(3.333)6
n > N 5!5(0.1667)

= 0.328

So, opening a fifth booth reduces the average queue length by 1.32 veh (3.287 - 1.967), average
time in the system by 0.066 min (0.331 - 0.265) and the probability of waiting by 0.22 (0.548 -
0.328).

A convenience store has four available parking spaces. The owner predicts that the duration of
customer shopping (the time that a customer's vehicle will occupy a parking space) is exponentially
distributed with a mean of 6 minutes. The owner knows that in the busiest hour customer arrivals
are exponentially distributed with a mean arrival rate of 20 customers per hour. What is the
probability that a customer will not have an open parking space available when arriving at the store?
Putting mean arrival and departure rates in common units gives Ji = 10 veh/h and A = 20 veh/h. So
p = 2.0 and, because pIN = 0.5 (which is less than 1), Eqnsl·Ol;to 1·"!1.canbe used. The probability
of having no vehicles in the system with four parking spaces is (using Eqn 1·3b)

1
Po = 2 22 23 24
1+-+-+-+---
I! 2! 3! 4!(0.5)

P = 0.1304(2i
n > N 4!4(0.5)

= 0.087

An incident occurs on a freeway that has a capacity in the northbound direction, before the incident,
of 4000 veh/h and a constant flow of 2900 veh/h during the morning commute (ie. no adjustments
to traffic flow result from the incident). At 8.00 am a traffic accident closes the freeway to all
traffic. At 8.12 am the freeway is partially opened with a capacity of 2000 veh/h. Finally, the
wreckage is removed and the freeway is restored to full capacity (4000 veh/h) at 8.31 am. Assume
DIDIl queuing to determine time of queue dissipation, longest queue length, total delay, average
delay per vehicle, and longest wait of any vehicle (assuming FIFO).

Let Ji be the full-capacity departure rate and Jir be the restrictive partial-capacity departure rate.
Putting arrival and departure rates in common units of vehicles per minute,

4000 veh/h 66 67 hi'

Ji = 60 min/h = . ve mm

2000 veh/h 33 33 hi'

Jir = 60 min/h = . ve mm

A = 2900 veh/h = 48.33 vehlmin

60 min/h
The arrival rate is constant over the entire time period and the total number of vehicles is equal to
At, where t is the number of minutes after 8.00 am. The total number of departing vehicles is

These arrival and departure rates can be represented graphically as shown in Fig. E16. As
discussed in Gt'0ev"."l!}, for DIDIl queuing, the queue can dissipate at the intersection point of
arrival and departure curves, which can be determined as

At this time a total of 3777.5 vehicles (48.33 x 78.16) will have arrived and departed (for the sake
of clarity, fractions of vehicles are used). The longest queue (longest vertical distance between
arrival and departure curves) occurs at 8.31 am and is

Qm = At - J.1r (t - 12)

= 48.33 (31) - 33.33 (19)

'Po; •.•T e.~

ll"ue.1o.:>
el •.ss iro--1".0 •.•.
Total vehicle delay, using equations for triangular and trapezoidal areas to calculate the total area
between arrival and departure curves is

Dr = 2.(12)(580) + 2.(580 + 1498.33)(19) - 2.(19)(633.33)

2 2 2

+2.(1498.33 - 633.33)(78.16 - 31)

2

The average delay per vehicle is 9.95 min (37604.2/3777.5). The longest wait of any vehicle (the
longest horizontal distance between arrival and departure curves), assuming a FIFO queuing
discipline, will be the delay time of the 633.33rd vehicle to arrive. This vehicle will arrive 13.1
minutes (633.33/48.33) after 8.00 am and will depart at 8.31 am, thus being delayed a total of 17.9
min.
 Chapter 2
Components Of The Traffic System
The characteristics of traffic are the result of many varied and complex interactions among the
four primary elements of the traffic system - road users, vehicles, roadways, and controls.

Traffic engineers have little to no control over the first two elements - drivers and vehicles.
They do, however, design the other two elements - roadways and control devices - in ways
intended to optirnise the safety and efficiency of drivers and vehicles in traffic.

Effective traffic engineering requires a basic understanding of these four elements. This chapter
treats some of their more critical characteristics, which have substantial influence on traffic
performance.

A number of critical characteristics of road users are quantifiable and can be taken into account
in traffic engineering decisions. These include perception and reaction times and visual acuity,
which are measurable and can be related to critical traffic analyses.

Clearly, driver perception-reaction time varies with a number of complex factors. In general, it
increases with

Age
Fatigue
Complexity of cue or task
Physical impairments
Presence of alcohol or drugs

The American Association of State Highway and Transportation Officials recommend the use of
a 2.5 second perception-reaction time in computations involving stopping or braking reactions.

Visual Acuity and Driving

Figure 1 illustrates the three primary fields of vision that affect the driving task. The field of
clear or acute vision, in which character and legend messages may be discerned, extends only 3
to 5 degrees around the centre line of the eye.

Periph~,,"_1 &

Vis-Io t'\ ('2.e.!. I €O )

A wider field ofjairly clear vision, extending about 10 to 12 degrees around the centre line of
the eye, provides for recognition of shape and colour but not specific legend.

The field of peripheral vision for most people extends 120 to 180 degrees around the centre line
of the eye. In this range, neither colour nor shape can be discerned. Drivers use peripheral
vision to recognise the presence of objects moving through the peripheral field. Studies have
shown that peripheral vision is a critical factor in drivers' estimation of speed.

The influence of other driver characteristics, such as hearing, on the driving task is not well
understood. Deaf drivers are not restricted, and statistics do not show any clear deterioration in
driving record (accidents, violations) with hearing impairments. Although certain handicaps
require modified vehicles, many handicapped individuals drive with acceptable accident and
violation records. In addition, psychological factors are clearly important, but are not well
understood.

Just as the characteristics of drivers vary considerably, so too do the characteristics of vehicles
using streets and highways. Highway systems accommodate a wide variety of sizes and types
of vehicles, from the smallest sub-compact passenger car to the largest double and triple tractor-
trailer combinations. The operating and performance characteristics of these vehicles vary
almost as much as their size and weight, and traffic engineers must consider these factors
explicitly in design and analysis of traffic facilities.

For road design purposes it is usual to adopt standard design vehicles, i.e. a typical design
passenger car, a typical design single unit truck and a typical design articulated vehicle (or semi-
trailer). These are shown in Figure 2.

When vehicles travel round a curve they occupy a greater width of road than when travelling on
a straight. This extra width is not significant on the larger-radius curves and on the wider
pavements (or lanes), but on the smaller-radius curves, and particularly at turning roadways at
intersections, it should be allowed for in the placement of traffic islands, road furniture and road
signs. Figure 3 shows the configuration of a single unit vehicle and of an articulated vehicle
when travelling on a circular curve.

1. the rear wheels track inside the path of the front wheels;
2. the front of the vehicle projects or overhangs outside the p~th of the front wheels.
 I' :I
II JEtI III I
bJ
King Pin }.sm

1.6m ~ I. 4.Sm ~lo.4m

.
'g.Om
---.,' 28m

1.Sm --1" I'

IX
2.4m l I.

S.18m

I- -/ to. 18m

II
~I I.
I 11'51t93~ 3.0Sm
. 1~·27~1
t O.18m

Figure 2 - Typical design vehicles

(SCU'l'Ca: IoolAASa.-A, I CfS" )
 db
Front overhang, C =1{

The articulated vehicle shown in figure 3. has dimensions as follows: w = 2.5 m, b = 4.0 m, d
= = =
1.5 m, g 8.0 m and h 0.5 m. Calculate the front overhang and the inside tracking of the
rear wheels when the vehicle is turning on a 30 m radius turn.

db
Front overhang, C =1{

_1.5x4.0_
- 30 02 -. m

Inside tracking, t = R - ~ - ~ (a + ~ )2 + h2 _ g2

where a is given by R2 = b2 + (a + w)2. In this case, R, band w are given, and can be
calculated directly, giving a = 27.2 (89.3 ft).
 t=30- i
2 5
-~(27.2 + 2;}) 2 + 0.52 _ 8.02

= 1.4 m

Acceleration capability is dependent on vehicle mass, the resistance to motion and the available
power, although most drivers will only use a proportion of the total engine output in any
selected gear for acceleration purposes. Typical acceleration rates in normal use are shown in
Table 1.

km/h/s rn/s2
medium cars 3-8 0.85 - 2.20
sports cars 12 - 16 3.33 - 4.5
~-2
commercial vehicles 4 0.21 - 0.56

In highway safety and design the most critical performance characteristic of vehicles is braking,
or deceleration. The time and distance required to stop a vehicle is a primary consideration in
virtually every aspect of traffic-system design and operation.

In general, the distance required to decelerate a vehicle from one speed to another (with the use
of brakes) is given by the formula:

v2 - u2
db = 254 (f + g)

db = braking distance (ft)

v = initial speed of vehicle (krn/h)
u = final speed of vehicle (krn/h)
f = coefficient of forward rolling or skidding friction
g = grade, expressed as a decimal
254 = units conversion factor

Consider a vehicle travelling at 97 krn/h on a roadway with a coefficient of forward skidding

friction of 0.40. If the grade is level, what is the braking distance to decelerate to 48 k:rn/h?
What is the braking distance to stop?

(972 - 482)
db (to 48 krn/h) = 254 (0.40 + 0.0) = 70 m
(972 - 02)
db (to stop) = 254 (0.40 + 0.0) = 92.6 m
Table 2 shows the normal range of coefficients of forward skidding friction used. The value
varies with type of pavement, running speed, tyre condition, and wet versus dry pavements.

Dry Surface Wet Surface

Surface New Standard Tyres BadlyWom Recommended by
Description Tyres AASHTOfor
Design
Running speed = 18 km/h
Dry bit. conc. 0.74 0.61 -
Sand asphalt 0.75 0.66 -
Rock asphalt 0.78 0.73 -
Port. cern. conc. 0.76 0.68 -
Running speed = 30 km/h
Dry bit. conc. 0.76 0.60 0.40
Sand asphalt 0.75 0.57 0.40
Rock asphalt 0.76 0.65 0.40
Port. cern. conc. 0.73 0.50 0.40
Running speed = 50 km/h
Dry bit. conc. 0.79 0.57 0.36
Sand asphalt 0.79 0.48 0.36
Rock asphalt 0.74 0.59 0.36
Port. cern. conc. 0.78 0.47 0.36
Running speed = 65 km/h
Dry bit. conc. 0.75 0.48 0.33
Sand asphalt 0.75 0.39 0.33
Rock asphalt 0.74 0.50 0.33
Port cern. conc. 0.76 0.33 0.33
All pavements Running speed = 80 krn/h 0.31
Running speed = 95 krn/h 0.30
Running speed = 110 krn/h 0.29
Running speed = 130 krn/h 0.27

In general, braking distance must be combined with perception-reaction distance to fmd the total
distance travelled between the initial observation of a hazard and the stopping of a vehicle. The
total stopping distance for a vehicle may be expressed as :

v2 - u2
ds = 0.278 vt + 254 (f + g)

where ds is total stopping distance (m), dp is the perception-reaction distance (m), and all other
terms are as previously defined.
The sample calculations that follow illustrate various applications of braking and reaction
considerations in traffic engineering :

Consider a highway with a design speed of 110 km/h. Table 2 indicates that a coefficient of
skidding friction of 0.29 should be used. This is indicative of wet-pavement conditions.
AASHTO further recommends the use of 2.5 sec for reaction time. The safe-stopping distance
is then computed:

1102
ds = 0.278 (110) (2.5) + 254 (0.29)

= 241 m
Thus, the highway must be designed such that drivers in either direction have a minimum of
241 m of clear sight distance at all times.

A critical aspect of signal timing is the change or clearance interval provided between conflicting
phases. A green signal cannot be changed instantaneously from one street to another, as
vehicles too close to the intersection when the light changes will not be able to stop safely. If
vehicles are instantly released from the conflicting street, accidents will occur frequently. To
avoid this, a clearance and/or change interval is provided between signal phases in the form of a
YELLOW signal and/or a brief period when all signals are RED - referred to as an ALL RED
signal.

The length of this change interval is related to safe-stopping distance. Any vehicle further away
than one safe-stopping distance from the signal when the YELLOW is flashed is assumed to be
able to stop safely. Any vehicle just at or closer to the signal than one safe-stopping distance
when the YELLOW is initiated will not be able to stop safely. Such a vehicle must be permitted
to pass safely through the intersection before the conflicting flow is released.

Consider the case of an intersection with approach speeds of 50 km/h, a coefficient of friction
of 0.45 and an assumed driver perception-reaction time of 0.5 sec. The safe-stopping distance
is calculated thus :

502
ds = 0.278 (50) (0.5) + 254 (0.45)

= 28.8 m

For a vehicle to safely clear the intersection from a point just at the safe-stopping distance, it
will have to travel the stopping distance, plus the width of the street, plus one car length (to
clear the rear of the vehicle). If the street is 12 m wide and a car is taken to be 5.5m long, such
a vehicle must safely traverse 28.8 + 12 + 5.5 = 46.3 m before vehicles from an opposing
approach are released.

. 46.3 m 333
hme = 50 km/h x 0.278 =. sec

The YELLOW and/or ALL RED signal should be 3.33 seconds long to accommodate the safe
clearance of vehicles unable to stop when the light changes.
Sign placement involves many issues that have been discussed in this chapter, including the
human eye's visual field. Consider the placement of a sign indicating "TOLL PLAZA AHEAD
- PREPARE TO STOP". How far in advance of the toll plaza should such a sign be placed,
given that it can be seen and read from a distance of 90 m and that queued vehicles from the toll
plaza rarely extend more than 46 m from the gates? Approach speed is 95 kmIh, the coefficient
of friction is 0.35, and reaction time is 2.5 sec.

Clearly, the sign must be seen in time to allow vehicles to stop safely before the end of the
vehicle queue at the toll plaza. Again, the safe-stopping distance is the key to the solution and is
found by the safe-stopping distance equation :

952
ds = 0.278 (95) (2.5) + 254 (0.35)

= 167.5 m

The vehicle queue extends 46 m from the toll gates. Thus, the driver must see the sign a
minimum of 167.5 + 46 = 213.5 m from the gates. The sign itself, however, may be read from
=
90 m. Thus, the sign must be placed a minimum of 213.5 - 90 123.5 m in advance of the toll
gates.

,
Li' 19,,-;.,.. _

------------
A
--
~
- --- -- ---C

- ~ second/word for reading and comprehending

- 1 second for pictorial sign
 v = speed of the vehicle (m1s)

L = legibility distance = 0.6 m per mm letter height (6 m per cm)

H = minimum capital letter height (mm)

S = offset distance to centre of sign, m. (In the case of a multi -lane highway, the
vehicle should be assumed to be in the fast lane).

0.37 NV + SCot e
H = L mm

Note - For sign mounted by the side of the road, e = 10°

For sign mounted overhead of the road, e = 5°

1. A rural road, with 85-th percentile speed of 90 km/h contains a 150 m radius curve, with
a superelevation of 0.06. Determine how far in advance of the TP (the point at which
straight and the circular curve joint) should the appropriate warning sign be located:

v2
Rmin = 12 7 (0(,max + f max)

v2
150 = 127 (0.06 + 0.25)

v90= 90km/h

= 25 m1s

offset (s) = 2.5 (shoulder) + 3;} (lane width) + 0.6 (clearance to edge of sign)
+ 0.5 (half sign width - assumed 1 m wide)

reading = 1.5 s (assumed)

 (25)2 - (21.4)2
= 2 x 1.5

37·~(~)
~e.oI.'J J.jf,.,.u-\ 1Ze..e/h",,- d,r"~~ r~,JcI~~ cP~
.(..Jl~""'t::~

2. The diagram below shows the plan and longitudinal section of aT-junction between a
local and a major rural road.

Given that the design speed of the local road is 60 km/h, calculate how far in advance of
the junction the pictorial "STOP SIGN AHEAD" sign should be placed on the local road
approach, 2.5 m from the pavement edge.

State any assumptions made.

 I

6!~.2_---- - - __U

Lo t,J6-.IT ul> \ N-A L

I
x'-<-7io rJ

f
-;

h2 = height of object in metres

=Om

= (illS + 10)~20f ~ ~60

= 72.506 m

u2 - v2
S= 2 a s
 o _ (60 )2
S- 3.6
- -2 x 0.15 x 9.81

S>D
94.38 > 'n.506 m

:. Read and comprehend distance = 1 x ~~.

3. An urban freeway, designed for a speed of 80 kmIh and with lanes 3.7 m wide, contains
a fork at the approach to an interchange.

Calculate the distance ahead of the fork at which an overhead direction sign consisting of
a three-word message should be placed 5.3 m above the road surface so as to enable a
motorist to change lanes, if necessary, to select the appropriate lane for his destination.
Calculate also the height of the letters for this sign.

State any assumptions made.

Solution:
 80
:. S = 3.7 x 3.6

~UAt:7,.,,,.. d,J: gS (rn

=::J. X ....E.. ~ It ,oj! r,:.y fa'" t!.
t 3·'
= ~i.I·Zf'l (-) ".C-ft-,j...~ .'
SA.7 'Ie f_j
/7

Accident investigators often make use of measured skid marks to estimate vehicle speeds before
an accident. Such measurements are used together with knowledge of friction coefficients,
estimates of collision speed, and the braking-distance formula to approximate the initial speed of
vehicles. .

Consider the following example. A vehicle hits a bridge abutment at a speed estimated by
investigators as 25 krnfh. Skid marks of 30 m on the pavement (f = 0.35) followed by skid
marks of 60 m on the gravel shoulder approaching the abutment (f = 0.50) are observed. The
grade is level. What was the initial speed of the vehicle?

This problem involves only the braking distance, as skid marks indicate only the distance
travelled after the brakes were engaged. Thus, perception-reaction time and distance are not
factors in these computations.

Two braking distances are known: 30 m on the pavement and 60 m on the shoulder. Each has
an initial and a final speed. As the only known speed is the final collision speed of 25 krnfh, the
second braking distance (on gravel) is considered first:

v2 - 252
db (gravel) = 60 = 254 (0.50)
This, however, is not only the speed at the beginning of the gravel skid, but the speed at the end
of the pavement skid. Thus, for the pavement skid:

v2 - 90.82
db = 30 -254 (0.35)

= 104.5 km/h
The speed of the vehicle just prior to skidding on the pavement was, therefore, 104.5 km/h.
This information, along with other aspects of the accident investigation, can be used to help
determine whether excessive speed was a contributory cause of the accident.

Streets and highways provide for two distinct and very different functions:

Through movement, or mobility

Land access

Highway facilities are classified by the relative amounts of through and land-access service they
provide. There are three primary classifications :

These facilities are provided to serve primarily through-traffic movement. While some
land-access service may be accommodated, it is clearly a minor function, and all traffic
controls and the facility design are intended to provide efficient through movement.

2. Collectors.

These facilities, the "middle" classification, are intended to serve both through and land-
access functions in relatively equal proportions. For long through trips, such facilities
are usually inefficient, though they are frequently used for shorter through movements
associated with the distribution and collection portion of trips.

These facilities primarily serve land-access functions. Their design and control
facilitates the movement of vehicles onto and off the street system from land parcels.
Through movement is difficult and is discouraged by both the design and control of the
facility.

Figure 9 illustrates the relationship between the three primary highway classifications and the
type of service provided. Note that the relationship is a continuous one, and that no magic
boundaries separate one class from another. In the final analysis, highway classification
requires substantial judgement on the part of the traffic engineer.
 -
..;..... -.•
~-r" '::;,
.;. ·t:.. ~'
~
, ,~
. - ."

Geometric Features of Streets and Highways

To understand traffic engineering, one requires a basic appreciation for the geometric design
features of highways, which contribute to operational characteristics and which can be part of
comprehensive solutions to traffic problems.

Geometric design has four basic elements :

1. Horizontal alignment
2. Vertical alignment
3. Cross-section design
4. Channelization

The fourth critical element of the traffic system is traffic control. It involves the communication
of traffic laws and regulations to drivers by means of control devices. These devices are of
three types :

Markings
Signs
Signals

Because these devices are the only means of transmitting operational rules to drivers, they must
be clear, easily interpreted and commanding of attention.
Chapter 3
Sight Distance
Sight distance is the distance for which a diver can see without obstruction. The
necessity for a driver to see sufficiently far ahead to enable him to assess developing
situations and to take appropriate action is obvious. The most frequent occasions that
arise are those (a) calling upon him to stop when approaching a stationary or moving
object, (b) requiring a decision regarding overtaking, and (c) requiring an assessment
of the course action to be taken at an intersection. The sight distances needed in the
circumstances are discussed below under the headings of:
* Stopping sight distance,
* Manoeuvre sight distance,
* Intermediate sight distance,
* Overtaking sight distance, and
* Intersection sight distance
When designing for sight distance, assumptions must be made about driver eye height,
object height and driver reaction time. Commonly used values are shown in Table 1
and 2.

Driver eye Object Oncoming Reaction

height (m) height (m) vehicle time (s)
height (m)
Australia 1.15 0.2 1.15 2.5
USA 1.07 0.15 1.3 2.5
UK 1.05 0.26 2.0 2.0

Height of Driver's Eye Height of object

(m) (m)
Stopping Sight Distance 1.15 0.20 l \

Overtaking Sight Distance 1.15 1.15 r H5 (F&>~ t!OY:Y ."ld A-'"~.-1

Intermediate Sight Distance
Intersection Sight Distance
1.15
1.80
-"0
0.15
0
2 (~t-r vc::r1.·"C_\ v.\-lij)

(Trucks)
Intersection Sight Distance 1.15 1.15
(Cars)
Intersection Approach 1.15 0.00
Sight Distance
Sight Distance through 1.80 0.60
Underpass

Stopping sight distance is the minimum distance at which a driver must be able to see
an object on the roadway so that he is able to stop before colliding with it. The time
taken to stop is made up of reaction time and deceleration or braking time.
=
stopping Distance Reaction Distance + Braking Distance
Reaction Distance:
Reaction Distance = Reaction time x design speed in m/s
= RT v/3.6 (m)
Where v is in KmJh
Braking Distance:
Friction force resisting motion = wd
Also "Grade force" either assists or resists motion parallel to road
surface force = w sin B, which is approximately = w tan B .
But tan B = GIlOO, where G = grade %
=
Therefore grade force 0.01 w G

Force resisting motion = algebraic sum of wd + 0.01 wG

By conservation of energy:
Force x distance = 0.5 M v2
distance = [0.5 (wig) v2] I force
= [0.5 (wig) v2] I (wd + O.OlwG)
= v212 x 9.8 x 3.62 (d + O.OIG)
Braking distance = v21254 (d + O.OlG)
Total stoppine distance = RTv/3.6 + v#254 Cd + O.OlG)

where; G is the per cent grade, being + for an uphill and - for a
downhill.
RT = driver reaction time (sec)
v = initial speed of vehicle (lcm/h)
d = coefficient of longitudinal deceleration

The value of the coefficient of longitudinal deceleration between the vehicle tyres and
the road surface, d, depends on factors such as the speed of the vehicle, the tyre
condition and pressure, the type of road surface and it's condition including whether it
is wet or dry. Typical design values of longitudinal friction for bituminous and
concrete surfaces are shown in Table 3.

Initial speed Coefficient of Longitudinal Deceleration

Ckm/h) Australia USA
50 0.52 0.35
60 0.48 0.32
80 0.43 0.30
100 0.39 0.29
120 0.35 0.28
Source: Austroads (1989), Table 6.1 & AASHTO (1984), Table 111.1
Table 3 Design values of longitudinal deceleration

Note that Table 6.1, AUSTROADS, assumes sealed surface and values are less for
high speed because of the reduced weight due to aerodynamic uplift. Smaller values
should be used for gravel surfaces. Table 6.2, AUSTROADS, gives rounded values of
stopping sight distance and assumes 2.5 (s) reaction time for high speeds, 2.0 (s) for
low speeds and 1.5 (s) at low speeds where drivers are expected to be alert.

Manoeuvre Sight Distance

This is for where the diver can swerve around an object rather than stop for it. Values
are given in Table 6.3 (AUSTROADS).

The minimum distance at which drivers must be able to see each other when
approaching head on so that both may stop before they collide with each other.
Intermediate sight distance is equal to double the stopping sight distance. It is
measured between the eyes of the drivers of the two vehicles that are approaching each
other.

Overtaking sight distance is the distance required for a drive of a vehicle to safely
overtake a slower moving vehicle travelling in the same direction before meeting an
oncoming vehicle. It is measured between the driver's eye and an oncoming vehicle as
in Figure 1.

Figure 1 Overtaking sight distance

AASHTO recommends that the minimum overtaking sight distance is the sum of the
following distances, which are shown in Figure 1.

l.d 1 - the distance travelled during the perception-reaction time, and during the initial
acceleration to the point of encroachment on the left lane.

dl = 0.277 tl [v -m + (a tl /2) ]
Where, tl = Time of initial manoeuvre, sec;
a = average acceleration, m/s2;
v = average speed of overtaking vehicle, km/h;
m = difference in speed of overtaken vehicle
and overtaking vehicle, kmIh.

2.d 2 - the distance travelled while the overtaking vehicle is in the opposing lane.

d2 = 0.277 v t2
Where, t2= Time passing vehicle occupies the
right lane, sec;
v = average speed of overtaking vehicle, kmIh.

3.d 3 - the distance between the overtaking vehicle at the end of it's manoeuvre and the
oncoming vehicle.

d3 = 0.277 (2v) t3
Where, t3 = Time between overtaking vehicle and the
oncoming vehicle at the instant the
overtaking vehicle has returned to its own
lane. One suggested value has been 1.5 s.
v = average speed of overtaking and
oncoming vehicle, kmIh.
4.d 4 - the distance travelled by the oncoming vehicle for two-thirds of the time that the
overtaking vehicle is in the opposing lane.

Speed Group (kmIh) 50-60 65-80 80-100 100-110

a = average acceleration (m/s2) 0.626 0.640 0.657 0.670
tl = time (s) 3.6 4.0 4.30 4.5
=
t2 time (s) 9.3 10.0 10.7 11.3
t3 = time (s) suggested values has been 1.5 s

Austroads has taken a different approach, based on work at the Australian Road
Research Board. In the Australian method:-

1.Establishment sight distance (ESD) is derived from the size

of gap accepted by a potential overtaking driver. It only
applies to two-lane, two-way roads. ESD depends upon
design speed and varies from 330 m at 50 km/h to 1500 m
at 130 krn/h (see Table 6.4, AUSTROADS).

2.Continuation Sight distance (CSD) is derived from the time

taken to complete the manoeuvre once the overtaking
vehicle is wholly in the opposing lane. CSD is less than ESD
(see Table 6.4, AUSTROADS).

3.Re-establishment distance is related to times spent trailing a vehicle before an

overtaking manoeuvre is attempted.

4.The overtaking vehicle and an oncoming vehicle are both assumed to travel at the
design speed.

5.The overtaken vehicle is assumed to travel at a lesser speed, taken as the mean speed
for it's direction of travel.

6.The sight distances are used with the 1.15m to 1.15m eye height and object cut-off
height.

Note:
Study the visibility diagram in Figure 6.3 of AUSTROADS:
* As long as the curving line is above the ESD line, overtaking can
commence
* As long as the curving line is above the CSD line, overtaking can
continue.
* So in the vertical alignment we see a line OA which indicates the
length of road for overtaking.

For the design case there is no point in providing CSD unless ESD has been available.
However CSD would be enough for overtaking a very slowly moving vehicle. Table
6.4 (AUSTROADS) is based on approximately 14% difference in speeds whereas in
practice much greater differences can occur.
Intersection Sight Distance

Sight distance at intersections is given in AUSTROADS " Guide to Traffic Engineering

Practice, part 5 Intersections at Grade".
For safe traffic operation at any given intersection, two or more of the following sight
distance considerations may be applicable, depending on the particular circumstances.

1. The sight distance along each approach.

2. Unobstructed sight triangle.

5. Vehicles on both approaches to adjust speed.

6. Traffic signal control.

The Si~ht Distance Alon~ Each Approach

In this case there are two requirements, and they apply at all intersections. Theyare:-

l.There must be stopping sight distance on each approach to the intersection, measured
from the driver's eye height to an object on the pavement. f0f t"hp relevant design speed
for each approach. Stopping sight distances are given ip Table b<2.. (A0S'l'2.o ,4 'D'S.") ,

2.The road pavement within the intersection proper must be visible on each approach
for~ance at least equal to the stopping distance for the relevant design speed, given
in al,le :and measured from the driver's eye height to zero on the pavement, so that
the contlict area itself, pavement markings and traffic islands can be clearly seen by the
approaching driver. If this sight distance cannot be provided, signs or other devices
should be used to make the precise location of the intersection clearly apparent to
approaching drivers.

Unobstructed Sight Triangle

It is desirable that vehicles approaching an intersection on conflicting paths, and b. 2-

travelling at the design speed of the approach roads, be mutually visib~e . t~es~
equal to at least the stopping sight distance for each approach given i Table :sotfiat if
necessary they can both stop before reachin~e:rsection. The des sight
distance to satisfy this condition is shown in Figure 3, sight distances being measured
between the height of the driver's eye in the cross-road vehicle and the height of the
main road vehicle (see Table 1). It is not always practicable to provide this
unobstructed sight triangle, in which case one of the following cases should be
considered.

sr·rp,,,,,,,
d.As. 't.~CA...
F-o-:B
Under this condition, the driver of a side road vehicle is required to come to a stop at
the intersection, and then must be given sufficient visibility along the main road to
enable the vehicle to move from the stopped condition and to cross or turn safely. Main
road vehicles travel through the intersection at normal speed. In the case of a crossing
movement, it may be assumed that a side road vehicle will stop with the driver's eye
about 3m clear of the outer edge of the main road pavement as in Figure 4.

S=0-3m
L = 5 m for car, 9 m for truck
a = 0.125 g for a car, and 0.06 g for a truck

I
I
I
~I
r-i'i
r.:
'D

---lj---i5M ~~_.
r.... VM (I:.•••, I.)

Figure 4 Intersection controlled by a stop sign

II II

Under this condition, the driver of a side road vehicle is required to give way to main
road vehicles to avoid a collision. Main road vehicles travel through the intersection at
normal speed. The position is illustrated in Figure 5.

Where, YZ = the distance travelled by the approaching vehicle

whilst the driver perceives and reacts to the
situation, and the required braking distance to bring the
vehicle to rest.
u = approach speed (m/s)
tl= perception - reaction time (s)
d = allowable deceleration (m/s2)
(typical value = 3.5 m/s2 or 0.356 g)
xz = the distance travelled by the major road vehicle during
he perception - reaction time tl of the other vehicle, the
time t2 for minor road vehicle to accelerate from point A to
reach point B clear of the junction, and a safety clearance
time t3.
= V (tl H2 +t3)
tl= perception - reaction time (typical value = 2 s)
 t2 is dependent on length AB and this includes the braking
distance (u2I2d), the width of the major road and the length
of the minor road vehicle. If the rate of acceleration of the
minor road vehicle is a rn/s2 (typical value 0.125 g for a car,
0.06 g for truck). Then:
t2=[(u2+2aS)1/2 -u]/a
S = distance AB (m)
t3= safety clearance time (s)

B~[
Figure 5 Intersection controlled by a " GIVE WAY" sign

Vehicles on Both Approaches to Adjust Speed

When a crossing does not have unobstructed sight triangles in all quadrants, and where
there is no stop, give way or signal control, a vehicle approaching an intersection,
whether it be on the main road or on the side road, must be able to perceive and react to
a potential hazard such as another vehicle and to adjust speed to avoid it. In this case, it
may be assumed that the distance at which a driver can first see a potential conflicting
vehicle should be equal to at least the distance travelled in a 3 second travel time at the
design speed. This results in sight triangles as in Figure 6.

B
~I-----r

Ur't - ob~T•.",J<..J
C;""T .l.i~_t.. .
r
,
I
I
I
Ql.--- I
AI

Figure 6 Intersection sight triangle - both vehicles to adjust speed

In Figure 6, distances d] and d2 are those traversed by vehicles A and Bin 3 seconds,
when travelling at the design speed. However, it should be noted that intersections
that provide this standard of sight distance need not necessarily be safe because it could
turn out that, even though the drivers of both vehicles change speed, they might not be
able to avoid a collision. In general, this standard should be adopted only on relatively
low volume rural roads where it is impracticable to achieve greater sight triangles.
Particularly as traffic volumes increase, if it is not possible to provide unobstructed
sight triangles, it is preferable to adopt the more positive stop or give way conditions,
and to sign them and provide the appropriate sight triangles.

At intersections controlled by traffic signals, the basic requirement is that drivers must
be able to see the signals early enough to take the appropriate action safely. However,
where practicable, it is desirable to provide sight triangles at least equivalent to those
required for stop conditions on the side road as an increased safety measured - having
regard to the possibility of signals not working or of vehicles travelling through the red
signal.

Example 1:

Calculate the stopping sight distance for a vehicle travelling at 110 km/h on a level
section at bituminous pavement. what would be the stopping sight distance on a down-
grade of 10 percent?

Solution;

Stopping Sight Distance = RT v/3.6 + v2/254 (d + O.OIG)

RT = 2.5 sec V= 100 km/h and f= 0.39 (From table 3)

On level section:
stopping Sight Distance = 0.7 x 100 + 1002/254 (0.39 + 0.01 x 0)
= 170m

On 10% grade:
stopping Sight Distance = 0.7 x 100 + 1002/254 (0.39 + 0.01 x 10)
=205m

Example 2:
In the example 2, what would the intermediate sight distance be in each case?

Solution;

Intermediate sight distance equals twice the stopping sight distance. Hence,
intermediate sight distances are 340m and 41 Om.

Example 3:
The attached sketch shows the plan of an existing uncontrolled street intersection in a
suburban residential area, where the increase in traffic over the recent years has resulted
in an increase of minor accidents and near misses. A speed survey of vehicles
approaching the intersection resulted the 85 percentile speeds to be 63 km/h in the E-W
street and 56 km/h in the N -S street. Recommend and design an appropriate measure to
remedy the situation.

u= 56km/h
V= 63 km/h
YZ = uq + u2/2d (refer to Figure 5)
Assume q = 2 see
d = 0.356 g

YZ = (56/3.6) x 2 + (56)2/2 x 0.356 x 9.81 x (3.6)2

=66m

XZ = V ( tl + t2 + t3 )
= V {tl + [ (u2 + 2 a S)0.5 - u / a] +.t3 }

where: Assume t 1=2 see and t3= 2 see

S=AB
= (u2/2d) + 11.2 + 5
= [ (56/3.6)2/2 x 0.356 x 9.81 ] + 11.2 + 5
= 50.84 m say 51m

XZ=(63/3.6) { 2 + [ «56/3.6)2 x 0.125 x 9.81 x 51)2 - (56/3.6)] /0.125 x 9.81 + 2}

= 121.42 m say 122m
2. "Stop" sign

Dm = (11.2 + 5 + 2 )
=18.2 m

DM = 0.4 VM [ (W + S + L) / a ] 1/2
= 0.4 x 63 (18.2/0.125 x 9.81)1/2
=97m o.k.

Example 4:
Calculate the minimum sight distance required (AASHTO method) for safe overtaking
to take place under the following road conditions.
Design speed of road, 100 kmlh
Speed of vehicle to be overtaken, 80 km/h

Solution:
q= 4.5 s
t2= 11.3 s
t3= 1.5 s
a = 0.67 m/s2
v= l00kmlh
m=20kmlh

d1 = 0.277 tI [ v -m + (a q /2) ]
= 0.277 x 4.5 [ 100-20 + (0.67 x 3.6 x 4.5 /2) ]
=106.48 m say 107 m

d2 = 0.277 v t2
= 0.277 x 100 x 11.3
313.01 m say 315 m

d3 = 0.277 (2v) t3
= 0.277 (2 x 100) 1.5
83.1 m say 84 m
d4 = (2/3) d2
 = (Z/3) 315
210m

d2+d3+d4
. +315+84+210
:::716 m say 720m
 Chapter 4
Horizontal Alignment Of Highways
Horizontal Alignment

The horizontal alignment of a road is usually a series of straight (tangents) and circular
curves connected by transition curves. Where a transition curve is omitted, the tangents
connect with the circular curve.

Figure HI illustrates how two tangents are joined by a circular curve and shows some
related circular curve terminology.

Tangent: T = R tan (M2)

Chord: C = 2 R sin (M2)
Mid-ordinate: M = R [1- cos (M2)]
External: E = R [sec (M2) -1]
Arc: L = (Il/180) .1. R
L = 30.5 (.1. / D) (alternative)
Arc subtended by an angle q: A = (Il/180) 8R
Abscissa of any point on the X = R sin e
circular curve:
The ordinate of any point on the circular
curve corresponding to the abscissa X:

Radius is a measure of the tightness 'of a curve. Another commonly used measure is the
degree of curvature (D), which is defined as the angle subtended by a 30.5 m (100 ft)
length of curve.
imperial units
The degree of curvature.: D = 1746/ R D= 18000/ R
(where R is in metres) (where R is in ft)
Arc: L = 30.5 (MD) L= 100 (MD)
Some basic horizontal curve relationships can be derived by summing forces parallel to the
roadway surfaces;

Where:- Wp is the weight of the vehicle parallel to the roadway (N)

Ff is the side frictional force (N)
~ .is the centrifugal force acting parallel to the roadway (N)
lCop
From basic physics this equation can be written as
n
[witb E: = F (WI')+ Fep)],
f . W v2 Wv2
W Sin a + F (W Cos a +-R Sin a) = -- Cosa
g y g Ry

Where :-F is the coefficient of side friction (which is different from the coefficient of
friction term, f, used in stopping-distance computations).
g is the gravitational constant
v is the velocity speed (m/s)

Dividing both sides of equation H2 by W Cos a gives;

v2
Tan a + F = -R (l - F Tan a)
g y
The term Tan a is referred to as the superelevation of the curve and is denoted E (the
number of vertical metre rise per one metre horizontal and has units of m/m). The term F
Tan a in equation H3 is conservatively set equal to zero for practical applications due to the
small values that fs and a typically assume. With E = Tan a, equation H3 can be arranged
such that:-

The maximum super-elevation to be adopted should range from 0.06 m/m in flat country up
to about 0.12 m/m in mountainous terrain, although many authorities tend to limit this
upper value to 0.1 mlm where practicable (See AUSTROADS, page 8). On urban roads,
with closely spaced intersections and access points to abutting property, maximum values
of 0.04 to 0.05 mlm are desirable.

In general, studies indicate the maximum sideway friction factors developed between new
tyres and wet concrete pavements range from about 0.5 at 32 kmIh to about 0.35 at 96
k:mIh. For normal wet concrete pavements and smooth Tyers the value is about 0.35 at 72
km/h(See AASHTOI984, page 165). On sealed bituminous pavements slightly higher
figures have been reported, with a range of from about 0.67 at 32 kmIh to about 0.3 at 110
k:mIh.For gravel or unsealed pavements, typical design values for the coefficient of
sideways friction range from 0.12 at 50 kmIh to 0.08 at 120 kmIh (Road Construction
Authority, RCA, Road Design Manual, 1984, Melbourne). However, safety and comfort
considerations require that lesser values be used for design. Typical design values are set
out in tables HI.

Table HI Design values for sideways friction

sealed concrete pavements (RCA)

Design Speed Coefficient Sideways friction

(krnIh) AUSTROADS(l989) ASSHTO(1984)
(Table 3.1) (Table 111.6)

50 0.35 0.16
60 0.33 0.15
80 0.26 0.14
100 0.12 0.12
120 0.11 0.10

In the design process, a design speed is chosen and a minimum value of radius of circular
curve, for that speed is calculated (see Examples on horizontal alignment). The radius
should preferably be larger than the minimum but it is helpful to find out what the smallest
value is. American Association of State Highway and Transportation officials (AASHTO)
provides general guidelines for the selection of E and F for horizontal curve design, as
shown in Table H2.
 Design Speed Maximum Maximum
(km!h) E F
30 0.04 0.17
50 0.04 0.16
65 0.04 0.15
80 0.04 0.14
100 0.04 0.12
30 0.06 0.17
50 0.06 0.16
65 0.06 0.15
80 0.06 0.14
100 0.06 0.12
105 0.06 0.11
110 0.06 0.10
30 0.08 0.17
50 0.08 0.16
65 0.08 0.15
80 0.08 0.14
100 0.08 0.12
105 0.08 0.11
110 0.08 0.10
30 0.10 0.17
50 0.10 0.16
65 0.10 0.15
80 0.10 0.14
100 0.10 0.12
105 0.10 0.11
110 0.10 0.10

A curve inserted to transition from the straight or tangent section of a road to a circular arc
is known as a plan transition curve.

The primary purpose of transition curves is to enable vehicles moving at high speeds to
make the change from the tangent section to the curved section to the tangent section of a
road in a safe and comfortable fashion. The proper introduction of transition curves will
provide a natural easy-to-follow path for motorists so that the centrifugal force increases
and decreases gradually as the vehicle enters and leaves the circular curve. This minimises
the intrusion of vehicles on to the wrong lanes, tends to encourage uniformity in speed and
increases vehicles safety at the curve.

Desirably all curves with a design speed of 60 km/h or greater should be transitioned,
except in the following cases:
1. In hilly or mountainous terrain where there is insufficient tangent
distance between intersection points.
2. When there are circular curves of 1800 m or greater, although
transition curves may often be used with advantage to improve
the appearances of up to 6000 m radius.
The essential requirement of any transition curve is that its radius of curvature should
decrease gradually from infinity at the tangent-spiral intersection to the radius of the circular
curve at the spiral-circular intersection. Various forms of curves are used for this purpose,
each having its own special advantages. The most common of these are the lemniscate, the
spiral (or clothoid) and the cubic parabola. In practice there is very little difference between
the results obtained by all three procedures. However, in recent years, the spiral transition
curve has become widely accepted by highway engineers, so it will be briefly discussed
here.

A circular curve joined to two tangents by spiral transition curves is shown in Fig. H2.
1..~
"

R = radius of the transition

IP = point at which the two straights join
TS = tangent - spiral intersection
=
SC junction of the spiral and circular curve
CS = intersection of the circular arc and second spiral
TS = second tangent point
PC = The point on the circular curve (extended) at
which the radius if extended would be
perpendicular to the straight
I = intersection angle
=
Lls spiral angle in degree
T = tangent distance (m)
S = external distance to the shifted curve
L = length of transition curve from TS to SC (m)
Lc = length of circular curve from SC to CS (m)
1 = distance along the transition to any point B from TS (m)
x = abscissa of any point B on the transition with reference
to the straight and TS (m)
y = ordinate of any point B on the transition corresponding
to the abscissa x (m)
p = shift (m)
¢
- -:l.
?,
=. -115'D
Z.
(oL~5'('U-~)
7\ (,RI-
 I

- \ 8° L
--- - (cl~y~)
7r' 6R

Figure H2. Basic properties of the spiral transition curve

=
T (R + p) tan (II2) + K
S = (R + p) sec (II2) - R
Ls = PIlSO ( I - 2 1l.s) R
x = 1 - [15/40 (RL)2]
Y = [13/ 6 (RL)] - [17 / 336 (RL)3]
P =L2/24R
=
K (L / 2) - L31240 R2
=
1l.s (180/ P ) (L/2) R

• To locate the tangent point T.S. :

(i) Calculate the shift p from the expression

L2
P = 24 R

IM = (R + P) tan 2I
(iii) Since A M =~
I L
I A = (R + P) tan 2 + 2"
 y = 6rR (cubic spiral)

x3 .
Y = 6LR (CUbICparabola)

or calculate the deflection angle d for particular distance 1 from T.S. using the fact that 0 = *
and so since

12
=( 2RL) (180/ 1t)

The final deflection angle locates Dl is ~s when 1= L.

. L2
Smce<1>s=2RL radians

L
=2R radians

L
Then Omax = 6R radians

L 180
=6Rx- degrees
1t

C
Ocirculararc = 1718.6 R min

where C is the chord length for the circular arc. The angle subtended by the circular arc D}
Dz at the centre of that arc is (I - 2 <1>s).

1t
Length of arc Dl DZ = R (I - 2 <1>s)
180
LenWI of Transition Curve to be Adopted

The length of transition curve to be adopted usually should satisfy the following:

1. rates of change of lateral acceleration;

2. pavement rotation and relative grade (AUSTROADS3.8.1 and 3.8.2);
3. appearance.

Usually, and largely for convenience in design and in the field, transition curve lengths are
rounded up to the next 5 or 10m length.

Rates of Change of Lateral Acceleration

As a vehicle passes along the transition curve its centrifugal acceleration changes from zero
at TS to v2JR at Sc. The transition length over which this change takes place is equal to the
vehicle velocity v multiplied by the travel time t. Thus

The minimum superelevation development length to satisfy the rate of rotation criterion can
be determined from the following expression:

Note AUSTROADS 3.8.1 has a / sign missing out of line 5, and the 0.09 results from
0.025 x 3.6 ie. converting V (kmIh) to v (m/s).

A high rate of rotation may caused instability in the steering/suspension- system and may
cause discomfort to the passengers. AUSTROADS 3.8.1 limits the rate of rotation to
0.025 radians per second for higher speeds (80 km/h and above) and 0.035 ( for design
speeds to 70 km/h). Assuming normal crossfall is 0.03 and constant speed, this fixes the
length of superelevation transition as in AUSTROADS Table 3.3 .
The rate of rotation method is normally used in the following cases:
1. Single-lane roads, when the edges of the pavement or shoulders
are not kerbed.
2. Two-lane two-way roads when the axis of rotation is about the
centre line of the pavement and when the edges of the pavement
or shoulders are not kerbed.
3. In other cases when the width between the axis of rotation and
the edge of the pavement does not exceed 5 m.

Relative Grade

The relative grade method should be used in all cases other than those for which the rate of
rotation method is applicable. In this method, the maximum relative grade for two-lane
pavements should be selected from the values given in Table H4. Length of super-
elevation development can be calculated from the following equation:

where L = the length of super-elevation development

w= width of pavement from the point of rotation to the outer edge
E= super-elevation
g= relative grade read from Table H4

AASHTO (1984) indicates that, in practice, a maximum length equal to 1.5 times the length
required for a two-lane pavement may be used on a four-lane pavement.

Table H4. Relative grade of the pavement edge with respect to the axis of
rotation for two-lane pavements (AASHTO, 1984)

Design speed
(Km/h)

Maximum relative grade

between the edge of two-lane
pavements and the centre line

For reasons of appearance, it is desirable to keep the ratio of the length of transition to the
length of circular curve generally within the range 1:2 to 1:4.

The actual attainment of superelevation requires that the road cross-section be tilted, the
amount of movement varying with its location on the transition or circular curves. This
tilting of the road can be carried out by anyone of the following three ways (see Fig. H3):
1. Rotation about the centre line of the road.
2. Rotation about the inside edge of the road.
3. Rotation about the outside edge of the road.

Method 1 is commonly used as it leads to minimum displacement of the road edges.

Method 2 is preferable where drainage problems may arise and method 3 leaves the outer
edge free.
 m
il)~\f.l~ II
centre line I I
I I -~f~~e~eJ=--i----
(a) ROTHlOII ABOUT CEllTRE lIl1E I I - - _I- _
1\ I I
I I
II, \ I ~ ~~~t. I
m ..., ", , 'I;~\
, I I I -t----
II ""', \ (b) ROTATlOII ABOUT IIlSIOE EIlliE I ceft~t.-I- - I
I~ ~
~ •••
~,\

==--=:..i.?.J -= C ='1
I I ....l-- -
~n~ ~o!..l L _
I
n
=~=.~, I
I·--t--J
I Outeredoel
~fr. I
I
I
ill I I - r ~
--.1r-~er IIfle I
~_ I
. II (e) ROTUIOII ABOUT OUTSIDE EDGE I "~I -- - -1--- - -
~ I I I , "I.... I
I II m II "" I
,-----

Various methods of introducing super-elevation are used. The discussion to follow is

typical of good practice, but for more specific details of methods adopted by different
authorities the relevant design manuals should be consulted:

When super-elevation is developed on plain circular curves without transition curves, it is

desirable, wherever possible, for the length of the circular arc to be at least twice and
preferably four times the length required for super-elevation development in order to give a
satisfactory visual appearance.

Typical design practice is to placed approximately two-thirds of the length of development

of super-elevation on the straight prior to the tangent point, and one-third on the circular
curve, beyond the tangent point. On this basis, the cross-section will vary uniformly from
normal cross-falls on the pavement and shoulders at a Point two-thirds of the length of
development before the tangent point to full super-elevation at a point one-third of the
length of development beyond it.

Typical methods of introducing super-elevation are shown in Figures H4 and H5 for

Undivided and divided pavements respectively.
SN,uldcr Pov-t~·~ I st-""Idl:'t"",
I I M e.ok-e>...... I -po." •.....;i I
I
I I
I I
I I sJ...c,ulJ('r--,-r--I---~'
I I I I
I I c?l
AT S·C:;.I I I ,::
.•. I
I I
I I
I
I I
L-e. 'I ~ I I I I I I
I S I --------- --_--!:!
1~ fiT T.S. I I
I I~
I I I I
,
I 1 I
I
I I IS 1
I I 1"1
I
't>:s T _ec- ~t. ~~<:: Tk.t. "T- S .

Fig. H4 Introducing super-elevation Fig. H5 Introducing super-elevation

on an undivided pavement on an divided road

Super-elevations are developed, in these examples by rotating the pavement centre line, as
follows:

N = normal pavement cross-fallon the straight

=
S shoulder cross-fallon the straight
=
E super-elevation to be developed
=
L length of transition curve.

Note: AUSTROADS suggests that the starting point for the super-elevation development can either
be established by projecting the rate of change of the outer lane back from the tangent to spiral point (TS,
in Figure 3.3 AUSTROADS) until it reaches super-elevation value of -n, or it can be fixed at a value
approximately 0.4 V d in advance of tangent to spiral point.

2. The outside traffic lane and the outside shoulder cross-falls are unifonnly rotated so
they become level at TS, the cross-falls of the inside pavement and shoulder remaining
unaltered.

3. Between TS and SC, the pavement and shoulders are unifonnly rotated so that full
super-elevation of both pavement and shoulder is achieved at Sc.

Horizontal curves should be checked to ensure that the normal sight distance requirements
are satisfied. Stopping sight distance should always be provided and where practicable
intermediate sight distance is desirable. Stopping sight distance should be measured along
the centre line of the inside lane. If necessary, the formation may need to be widened
and/or clearing and benching outside the formation may be required to provide stopping
sight distance. The most usual method of calculating the offset distance necessary to secure
the required horizontal sight distance is illustrated in Figure H6.
 !
/
/h -I
I.f'';;: CC5 (Q- ""')
(1<.-n)

D= 2. eft (R-I'"\)
cp \ "v- r 0... .,,{,\ s
••••.•••••.

\'lrh--\
: I!.
ii···~
Figure H6 Sight distance consideration on a horizontal curve

Application of Geometric Desi2n Principles to Route Desi2n

The Fundamental in the highway geometric process is the establishment of the new
highway's centreline and cross sections in relation to the terminal points and to the
topography through which the highway will pass. The vertical and horizontal alignment of
the centreline determines the amount of cut and fill. cross section details, drainage design,
construction and user cost, and environmental impacts.

Preliminary Route Layout and Geometric

The major activities may be divided into:

1. design controls;
2. establishing an initial alignment;
3. balancing cut and fill; and
4. refinement of the design.

* Traffic volumes
* Design vehicle
* Design speed
* Maximum grade
* Lane and shoulder widths
As well as the design control listed above, it is necessary to specify several other variables
that are inputs to establishing a preliminary route. These variable are described as follows:

1. Minimum radius of horizontal curves, based upon the design and the
permissible super-elevation.

2. Minimum length of vertical curves, based upon design speed and difference
intersection grades.

3. Maximum grade at any point on the highway, determined from

consideration of road classification, truck traffic, and terrain.

4. Maximum grade in proximity to existing intersections. The vertical

alignment of the proposed route should allow for a minimum to moderate
grade approaching the intersection with the existing highway in order to
assist in safe stopping on downhills and improved sight distance on uphill
approaches. Ideally, this grade should be no more than is required for
adequate drainage. Because this approach grade may intersect with a
vertical curve ascending or descending the hillside, a vertical curve may
be required at this location. The length of this curve is determined by
consideration of the design speed and intersection grades. For
preliminary design purposes, it is suggested that within a distance of
about 30 m from the intersection's stop line, the grade be no more than
2%.

5. Minimum grade at any point on the highway to ensure adequate

drainage. At minimum of 0.5% is suggested for preliminary design
purposes.

6. Maximum horizontal approach angle at intersections. The horizontal

alignment should feature as nearly as possible a right-angled intersection
with the existing highway (desirable intersection angle between 700 -
90°). It is suggested that for preliminary design purposes the proposed
approach to the intersection be a tangent section for a distance of at least
30 m to aid drivers' visibility at the approach.

7. Maximum depth of excavation and height of fill. The design details of the
cut will depend on the type of material excavated and upon the depth
from the original grade level. Although no guidelines have been
established, in general, a maximum of 10m should be about the greatest
depth of cut for most highways, with 6 m being a more desirable
maximum. An example of selected characteristics of highways
constructed in cuts is shown in Figure H7. However, the design of all
slopes should be based upon detailed engineering soils analysis.

As with the cut section, it is usually desirable to keep the height of the fill
section to 10m or less, with 6 m being preferred maximum. Above this
height, depending upon the topography, the classification of the highway,
and affected land uses, it may be more economical to construct a bridge.
Usually, the side slopes should be no steeper than 1 in 2 for regular fill
material from soil stability, surface vegetation, and maintenance
consideration.
 -- \~-
Original ground level

CL
..~-R1-_=--~--,-.---
L=F

__ -.1 c,-/ [o(~)

,Original ground level

_R-L b

Development of the alignment is a trial and error process involving first defining a trial
alignment, then checking to see if it complies with the horizontal and vertical controls, then
modifying it in successive iterations until all the controls are complied with. One approach
to this process is illustrated by the problem example shown in Figure H8.

Problem: Connect point A and B below with a highway having maximum grade of 6%, minimum
grade(for drainage) of 0.5%, a minimum horizontal radius of 240 m, minimum vertical curve length of 120
m, and maximum cut and fill depth of 6 m.

STEP 1: Examine the contours along the

shoulder route and estimate the steepest
ground slope on this route is between
?'ShO:~ ~
t;
I
. ---
Existing Highway

~--!.!///1'(
""-

II/II
"

points X and Y and features ( by examination _ b7 _ (~pothetical) "J / J /

of the height differences indicated by the - - - - _.... I J/ / I
contour lines) a rise of approximately 38 m -73 ...."/// ;
in 240 m horizontal - a slope of over 15%. ; / I
This is more than twiCe as steep as the allowable - 7'1 - - - - - _ .... I ~ :j
grade and would result in excessive depths of cut 5 - - - - C\\ ./ ""'"
and height of fill. Therefore, a less steep route - ~ ~o - 305.... "0 I j

should be examined (step 2, below). scale: CV"')

STEP 2: Sketch a new trial alignment with a

reduced ground slope. The one here is sketched //1/1
freehand and includes a curve, the radius of which //lJJ
is approximately the same or larger than the 240 m //11/
minimum specified. Notice the ground between X and - _..,.. " I

Y of Step I will be less steep because the horizontal

alignment of the highway crosses the contour lines -73 - - -
at a greater angle than Step 1. -7&f-----
_ f,'7 ~ - -
o
I
3""5 I
" LO
STEP 3: Convert the sketch of Step 2 into a
dimensioned tangent and circular curve alignment
by scaling and the use of compasses or a template - 73 - - -
for constructing and measuring the curve. It is also -7Cf - - - - -
usually desirable to ensure that the intersection of
the proposed highway with the existing road is within -~?---
about 15 degrees of a right-angle in order to assist oI
30S
I

driver sight distance requirements.

STEP 4: Construct a profile of the ground levels as a

basis for defining a profile of the highway pavement ql -
which does not exceed the 6 % slope and the 6 m cut PROFILE
and fill requirements. This is usually done to balance -%-
the cut and fill amounts also, and an ?cceptable(though 1
not necessary unique) solution is shown here, for the
alignment of Step 3 above. It is also usual1y desirable H~hway
rofile
to ensure that the grade of the proposed road where it
intersects the existing road is a maximum of about 2%
to assist stopping and sight distance requirement. Note 67 ~
also that the minimum grade requirement of 5 % has
been complied with. {,I
A -~
0.5%
I I vc I
340 ''34

The alignment selection process can now be followed for a number of alternatives. There
are several reasons for doing this. For example, the shortest highway which is feasible in a
technical sense may not be the least expensive. If possible, at least three alternatives should
be defined, all of which are technically feasible and conform to the specified design
controls.

The last phase of the design process is the production of a set of working drawings for the
construction. The set would usually include plan and profile drawings (see working
drawing - sheets 1 to 3).
 f1 = 57°35'23"
R = 712 m S}'iV
T = 424.84 m 1( 2s:
Lp = 67 III

S = 0.26 III

A = 648.65 III

[I = 100.76 m

6. = 27°44'14" I-'
I-' (Xl
'-J -!'>
(Xl .f.:>
R 1528 III
1-- W 'J
0
T = 410.75 III NN
Of-'
N

NN
L 67 III
P 'J 11l
00
S = 0.12 m
A = 672.70 m
t 46. 00 m
I~
I:E:: :1:
E ,= 0
""1
>-.
G1
("I ~(.ul:i v A :J:
-'. :E:
0(0'/\ ::J
<.0
)::>
-<
CJ 0
""1 Pl
llJ Vl
::;: ......•
-J.
G1
Scale 1: 10000 ::J Z
<.0
l/l
Can tours: III A. A. 0
Vl
::r
rt>
ro
rt-

f-'

----------------------"""'--
 ~ ::c
.,
0 >-<
G>
7' ::c
-'. ~
~ );>
\L:l -<
.,
0
AI
0
rr1
VJ
~
G>
~ :z:
lCl
1I1

VJ
::r
ro
-"- :
'"-'
ro
r+
> iC N
----/ -.......
220.0 /'
""
/ V~
210.0
200.0 ~ - -- ~,'
~~~ ~

LONGITUOINAL SECTION 190.0

/~
- ~~ /
on C.L. Carriageway
180.0 / ~ -
.//
Scales: 1:10000 horlz.
1: 1000 vert. 170.0
?"
~
~
- --...... ------
Z
:>K.
160.0 l--
t
m A.O.O. fJ 150.0 I
::l
E.G .L. m A.o.o. '" ~ r.
~ ~ ~
.
I

~ 'S .;.
to..
~ l?'
F.L. on C.L. c'way m A.D.O.
~ . . !!
IJl
~
0
<-l
~
III
1::
~
~ ~
~
<i

GradIent %
.•. 5·8'<- _____- ~'86
---- ---- -.- ----_±_?::~:::>

'I.e. 81~"", -\--- ---:i_.c. 71' •.....

Vertical Alignment

---- c;.:_~'_A:::._ ~r-------

T 5 C.R. 7°4<1'14_'__ .. _-~ S S ?5'.2.5~ S T
liar lzon ta I Alignment c~ l-- ------ ----~-------~
- t:-
19 ~ ": 1Il
~
()
.;, :;; V)
t;- c t;-
'" t. 1II

Chalnage m 0 I'l
«l ...
~
t<J

~ ~ ••~ -..•.- a C'/

't
<T
<J)
~
--
tl ..•§
<!
c,
o-l '"
~
..•
~ '"
:!; ~
0

~
 Working Drawings (Sheet 3)
~
Typical Cross sections
I
2.5m 7.01m 2.5m
rave 1 sealed tr.affic:·grave.
s auld r lan~s hould r
-+- 3% 3%+

;;.-"\

Table drain

~~;"" -..•.
..•..

j 4:1

~
Table

.
7.0 1m
sealed traffic ravel
lanes s oulde
.•.. 3°/
• /0
:

12• 5 m 7.0 J m 2•5 mI

ravell sealed traffic ravel
s~oulde~ lanes 's aulde
-+- 5% -+- 5°/ iI
10
2 01
/0
+

//'-' ///
An undivided two way road, carriageway width 7 Am contains a
curve of 200m radius.

Calculate the clear distance from the inner edge of the carriageway to
any lateral sight obstructions, to provide the minimum stopping sight
distance for a design speed of 90 km/h.

(90)2
Ds = 0.7 x 90 + 254 x 0.4 1

1 (R-m)
<I> = cos- (R - n)

D
<1>=2 (R - n)

141
- 2 (200 - 1.85) = 0.356 radians

= 180 x 0.356
1t

Also:
1 200 - m
<I> = cos- (200 - 1. 85)

200 - m
:. 0.93737 = 200 - 1. 85
 Therefore, 10.56m must be cleared from the inside edge of the
inside lane.

5488
1t R

50 = 5488 or R = 349 m
1t R

Calculate the mInImUm radius of a circular curve having a design

speed of 80 km/h and a superelevation of 0.10 m1m. Use a design
value of sideways friction of 0.14.

v2
E+F=127 R

802
0.10 + 0.14 = 127 R

What length of transItIOn curve should be adopted for a 500 m

radius circular curve having a design speed of 100 km/h and a
super-elevation of 0.06 m/m? Assume that the circular curve is 300
m long, and that the road is a two-lane-two way rural road.
Solution :
From table H3, the value for the rate of change of lateral
acceleration is 0.45 m/s3, and the minimum length to satisfy
this requirement is given by :

0.0214 V3
A = R L

0.0214 x 1003
:. 0.45 - 500 L

As the road is a two-lane two-way rural road, assume that the

cross-section will be rotated about the centre line. From
AUSTROADS Table 3.3 , the minimum length to satisfy the rate of
rotation requirement is 67 m (by interpretation). For reasons
of appearance, the ratio of the length of transition to the length
of circular curve generally should be within the range 1:2 to
1:4, i.e. between 75 and 150m. Therefore adopting a rounded
length of 100m would satisfy all requirements.

For the transition curve III example 4 how would you introduce the
super-elevation?

See figure H4. Assuming a normal pavement cross-fall of 0.03

m/m, development of superelevation would begin a distance of
NL/E or 0.03L/0.06 m before TS, i.e. 50m in the case of the
100m transition curve. Full super-elevation would be
developed at SC.

Calculate the transition curves Lp and Le for a roadway given the

following data :
 e 1 - e2
Le - . X v
rotatIon

0.07 - (- 0.025) 2
= 0.03 x 2.2

The plan transition curve Lp will be the distance from where

the cross-fall in the outer lane has been reduced to zero until
the junction of the spiral with the circular curve,

. L 0.07 - (0.0) x 22.2

l.e. p - 0.030

For convenience, let Lp = 50m.

v2
R=(e+f)g

(22.2)2
- (0.07 + 0.14) 9.81

L
P = 24 R
 (50)2
-24x240

15
x = 1 - 40 (RL)2

13 17
Y = 6 (RL) - 336 (RL)3

Table E6 was prepared.

T a bl e E6 T abl e f or settm~ out 0 e transItIOn curve
Arc 1 Straight x Offset y Tangent to Deflection
Curve <I> Angle 8
10 10.000 0.0139 0° 14' 0° 4' 46.48"
19.44"
20 20.000 0.1111 0° 57' 0° 19' 5.92"
17.75"
30 29.996 0.3749 2° 8' 54.93" 00 42'
58.31 "
40 39.982 0.886 3° 49' 1° 16'
10.99" 23.66"
50 49.946 1. 7348 5° 58' 5.92 1° 59'
21.97"

Three straights, AB, BC and CD have whole circle bearings of 30°, 90°
and 45° respectively. AB is to be connected to CD by a continuous
reverse curve formed of two circular curves of equal radius together
with four transition curves. BC, which has a length of 800 m, is to be
the common tangent to the two inner transition curves. Determine
the radius of the circular curves if the maximum speed is to be
restricted to 80 kmlh and a rate of change of radial acceleration is
0.3 m/s3 obtains. Give (i) the offset, and (ii) the deflection angle,
with respect to BC, to locate the intersection of the third transition
curve with it's circular curve.

Note - Whole-circle bearing are measured clockwise from

north, from 00 to 3600.

y3
Length of transition curve, L = R A

y3
:. A = 0.3 = L R

80 x 1000 1000
y - 3600 = 45 m/s

1000 x 1000 x 1000

L=45x45x45xO.3R

36580
= R m

L2
P = 24R

55 760 000
= R3

L L
:. BC = (R + P) tan 30 + 2 + 2 + (R + P) tan 22.5
· 800 - (R 55 760 000) 300 36 580 (R 55 760 000) 2250
.. - + R3 tan + R + + R3 tan.

36 580
an d L = R

L3
-6 L R

(48.23)3
=6 x 36 580

T
B
1800L 180 L
1t R min or -;- 6 R degrees

1800 x 48.23
1t x 758.5
It is required to Jom two straights having a total deflection angle of
180 36' right by a circular curve of 450 m radius, having cubic spiral
transition curves at each end as shown below. The design velocity is
70 km/h and the rate of change of radial acceleration along the
transition curve is not to exceed 0.3 m/3. Chainage of I is 2524.20m.
Determine the following :-

(i) Length of required transition curve;

(ii) Shift;
(iii) Tangent length IT;
(iv) Cubic spiral transition curve with 10m chord lengths;
(v) Circular arc Tl Tz.

c:> /
to 3.b

700
- 36 m/s

A = Rate of change of radial acceleration = 0.3 m/s3

v3
AL=R
 L2
Shift = 24 R

(54.44 )2
-24 x 450

(1°1°1°) T angent Iengt h IT = (R + P) tan - 2"

I +L
2

18° 36' 54.44

= 450027 tan 2 + 2 m

(iv) Deflection angle to locate a point on the transition distance 1 m

from T

0= 1800 12
1t R L

= 1800 12
1t x 450 x 54.44
 Chord (m) I (m) Chainage Deflection angle
0 0 2423.23 0
6.77 6.77 2430.00 0.02339 x 6.772 =
01.07'
10.00 16.77 2440.00 0.02339 x 16.772 =
06.58'
10.00 26.77 2450.00 0.02339 x 26.772 =
16.76'
10.00 36.77 2460.00 0.02339 x 36.772 =
31.062'
10.00 46.77 2470.00 0.02339 x 46.772 =
51.17'
7.67 54.44 2477.637 0.02339 x 54.442 =
69.31 '

(v) Circular arc Tl T2

<1>s
8max =""3
:. <1>s = 3 8max
= 3 X 69.31'
= 3° 27.93'

1tR
Length of circular arc = 180 (I - 2 <1>s)
450
= 18 0 X 1t xlI. 669

= 91.65 m
:. Chainage at end of circular curve;

= 2569.32 m
:. Chainage at tangent point U = 2569.32 + L
= 2623.76 m
ChapterS
Vertical Alignment
The vertical alignment of a road usually consists of a series of grades (or straights) joined to each
other by vertical curves, although for very small changes of grade, vertical curves may be omitted.

The purpose of vertical curves is to provide:

1. a safe, comfortable transition from one grade to the next;

2. adequate sight distance across the junction of two grades; and
3. a satisfactory appearance.

Grade is the longitudinal slope or steepness of slope going along the road and is usually measured
in percentage.

Modem day passenger cars can travel on up-grades of up to about 7 percent or so without
appreciably reducing speed, whereas on long sustained grades in excess of about 3 percent, the
speeds of trucks may be significantly reduced.

Grades < 3% : High levels of comfort, safety, and efficiency. (3% is the desirable
maximum for freeways).

Grades 3% to 6% : Still good for design speeds to 90 km/h.

Grades> 9% : Not to be used except in difficult circumstances; however, particularly on

relatively low volume access roads in both rural and urban areas and hilly
terrain, grades of up to 20 percent and steeper have been used when there is
no feasible alternative.

Minimum grades are usually fixed by drainage considerations. Earth table drains should preferably
be no flatter than 0.5 percent. When a kerb and channel or other form of lined or paved drain is
used, drainage grades as flat as 0.3 to 0.2 percent may be acceptable, provided drainage can be
discharged to suitable outlet drains. ~

The usual form of vertical curve is the parabola. The general form of the parabolic equation, as
applied to vertical curves, is :

where y is the roadway elevation, x stations from the beginning of the vertical curve (i.e. the
PVC). By definition, c is the elevation of the PVC, since x = 0 corresponds to the
 ~=2aX+b

where Gl is the initial slope as previously defined. Also note that the second derivative of equation
1 is the rate of change of slope and is :

dZy
dx =2a

The rate of change of slope (2a) can also be written as AIL. However, the average rate of change
of slope, can also be written as :

dZy _ Gz - Gl
dx - L

GZ - Gl
a= 2L

Pvr
I
-t

Figure 1 - Offsets for vertical curves.

Some additional properties of vertical curves can now be formalized. In Figure 1, Y is the offset at
any distance, x, from the PVC; Yill is the mid curve offset; and Y f is the offset at the end of the
vertical curve. From the properties of an equal tangent parabola, it can be readily shown that :

A Z
Y=200L x
where Y is the offset in feet, A is the absolute value of the difference in grades (G2 - Gl expressed
in percent), L is the length of the vertical curve in m, and x is the distance from the PVC in m. It
follows from Figure 1 that:

AL
Yf= 200

Computation of the high or low point on a vertical curve

The location of curve high and low points (if applicable) are important for drainage considerations;
for example, on curbed streets catch basins must be installed precisely at the drainage low point.

It was noted earlier that the slope was given by the expression

Figure 2 shows a sag vertical curve with a tangent drawn through the low point; it is obvious that
the tangent line is horizontal with a slope of zero; that is :

Had a crest curve been drawn, the tangent through the high point would have exhibited the same
characteristics.

A
2a=L

~
X= A

t~ eve. 1.ow poi..•

T
j, '" --
-- ---
1. Compute the algebraic difference in grades; A = G2 - Gl.

2. Compute the chainage of the PVC and PVT. If the chainage of the PVI is known, l/2L is
simply subtracted and added to the PVI chainage.

3. Compute the distance from the BVC to the high or low point (if applicable) :

and determine the station of the high/low point.

4. Compute the tangent grade-line elevation of the PVC and the PVT.

6. Compute the midpoint of chord elevation:

Elevation of PVC + elevation of PVT

2

_ difference in elevation of PVI and midpoint of chord

d- 2

Tangent offset = Y = Y m (I52)2

where x is the distance from the PVC or PVT (whichever is closer) to the required station.

9. Compute the elevation on the curve at each required station by combining the tangent offsets
with the appropriate tangent grade-line elevations. Add for sag curves and subtract for crest
curves.

Table 7.1 (Section 7 of the AUSTROADS) sets out details of appearance criteria for vertical curves,
namely the maximum grade change not requiring a vertical curve and the minimum length of
vertical curve.

Apart from the appearance criteria mentioned above, the main design consideration at crest vertical
curve is that of sight distance.
1. when the required sight distance is less than the length of the vertical curve; and

2. when the required sight distance is greater than the length of vertical curve.

Sight distance less than length of vertical curve (D < L)

This case is illustrated in Figure 3.

L= AI)2
100 ("'"2 h I + "'"2h2)2

If the dangerous object is assumed to be

at road level, then:

AI)2
L = 200hI

If hI = h2, then:
+-__ L_

AI)2
L = 800hI

If hI = 1.15 m and } (SSD as per Table

h2 = 0.2m } 6.2 AUSTROADS) Where A = percent change of
grades
lA
e = 800
L = length of vertical curve
D = required sight distance

200hI
If h2 = 0, then L = 2D - ~

800 hI
IfhI=h2,thenL=2D- A

462
IfhI = 1.15, and h2 = 0.2, then L = 2D - A
The decision as to which condition should be used at a particular site can be made by solving either
of the equations :
(9 - p) L _ (p - 9) D
800 e - 800

depending upon whether L or D is the known value.

In either case, if e is found to be greater than hI, then the equation for the first condition, i.e. when
L is greater than D, should be used. If it is found that e is less than hI then the equation where L is
less than D should be used.

Overtaking requirements
The general formulas for crest vertical curves are applied, but the 1.15 m height of object results in
the following specific formulae with the same term as above:

AD2
when D < L, L = 920

920
when D > L, L = 2D - A

Eye height hI = 1.15m

Object cutoff h2 = 1.15 m.

Generally, it is impracticable to design crest vertical curves to provide for overtaking sight distance
because of high cost where crest cuts are involved.

There are at least four widely accepted criteria for determining the minimum lengths of sag covers.

1. Motorist comfort.

2. Vehicle headlight sight distance.

3. Highway passes underneath a structure such as a bridge.

4. Drainage.

5. General aesthetic.

Length of vertical curve determined by riding comfort

Centripetal acceleration:

v2 y2
a= r= 13 R rn/s2
 100 Lv
:. r= m
~

. y2 y2
an d , Since a = 13 R' r = 13a
100 L y2 V2 ~
~ = 13a and L = 1300 a

:. Rate of change of gradient: ~ = 13~ a

Rate 0f c h ange 0f gra d·lent n1

-/0
/
100 ~ -- 13 ay2000
100 m : 100 n1
-/0
/ 100 m

It is good practice to provide headlight sight distance on unlit roads, i.e. the distance that can be
seen by the headlight beam should be at least equal to the stopping sight distance for the relevant
design speed. Typical assumptions are indicated in Figure 6, and in particular :

Based on the above assumptions, expressions for the length of vertical curve L in metres in terms
of the required stopping sight distance D in metres and the change of grade A gl - g2, expressed =
as a percentage, are as follows :
 Figure 6 - Typical assumptions for calculation of headlight
sight distance on sags
Sight distance less than length of vertical curve.

In this case,

AJ)2
L= 150 + 3.5D

L = 2D _ 150 + 3.5D
A

In general, a sag vertical curve designed for headlight sight distance will usually provide for a
maximum vertical acceleration of about 0.02 g rn/s2. Thus, such a curve would normally be longer
than one based on the comfort criterion.

At sag vertical curves, overhead obstructions such as overpass structures, sign gantries and
overhanging trees may limit the available sight distance, particularly in those cases where it is
desired to provide greater than stopping sight distance. If these types of obstruction are likely to be
present, adequate checks should be carried out to ensure that the intended sight distance is in fact
available.
 _ h, . , . ~ h, _

~ I~=-I
I
I . L I I' .

DZA
L = --~-=--------==--=---=---
200 H H - hi + "'./H - hZ]Z

---I
L = 2D _ 200 [v!H=hi .: "'./H - hz]Z

Where H = height of overhead obstruction;

hi and hz are eye height and object cut off height respectively; and
D = SSD as per Table 6.2 of AUSTROADS.

Typical value for hi = 1.8 m (driver eye height for commercial vehicles) and
hZ = 0.6 m (vehicle tail-light or stop-light).

Parabolic vertical curves are sometimes specified in terms of K values, where K is defined as the
length of vertical curve required for a 1 per cent of grade. Thus:

L=KA
L = length of vertical curve in metres
A = per cent change of grade
K = the K value appropriate to the design speed and to the particular sight distance considerations.
Aids for Calculation of Vertical Curves
While the length of vertical curves required under different conditions can be calculated directly by
use of the appropriate equations given above, various authorities usually have the following types
of aids to calculations:

1. Design charts which enable the required length of vertical curve to be read directly for any given
design condition ( such as, for example, stopping sight distance, design sight distance and change
of grade).

2. tables of K values for given design conditions and design speeds, so that the length of curve
required can be obtained simply by multiplying the appropriate k value by the grade change.
Values of K determined for the various sight distance design situations are given in tables 7.2,7.4
and 7.5 of AUSTROADS.

The criteria which decide the minimum length allowed are (see Section 7 of the AUSTROADS for
more details):
- Appearance,
- Comfort and
- Various sight distances.
Sight distances are in turn dependent upon design speed.

* Draw in rough grade lines.

* Determine levels and chainages of IPs by scaling off drawing.
* Calculate grades.
* Alter if any grade is steeper than 9% .
* Calculate change of grade A at each IP.

Input to Tables 7.1, etc (AUSTROADS ) - They all give minimum lengths so the bil:l:est
minimum is the valid minimum.

Table 7.1 - If good appearance is more important than usual, use the longer minimum i.e. at 60
km/h use a 50 m not 40 m.

Table 7.2 - For high speed or high class roads use the a = 0.05 g column, i.e. at 100 km/h
Lmin = 16 A. The comfort criterion controls over humps or through spoon drains or similar cases
where sight distance is not obstructed. One other example is where irrigation channels cross under
the road in flat country, the road usually humps up and down again abruptly without obstructing
sight distance.

Table 7.3 - The normal value of C for Stopping Sight Distance (SSD) cases on crests is 461, and
the normal value of C for Continuation Sight Distance (CSD) cases on crests is 920. They can be
checked or calculated for other cases on crests using the formula top right of page 35
(AUSTROADS). Foe sags with headlights and sag under overhead bridges, the C formulae are on
page 36 (AUSTROADS).

Table 7.4 - Stopping sight distance for two object heights, continuation sight distance and
establishment sight distance requirement are tabulated in terms of bKA. Note that for a 10%
change of grade and a design speed of 130 km/h, the minimum length of vertical curve to provide
establishment sight distance is 25 km. Not likely to occur in practice.

Section 7.5 - Some values have been tabulated in Table 7.4 but for other values and for the sag
under overhead bridge case we need to work the formulae. Both equations (1) and (2) should be
evaluated, and only one of them will be valid.

Table 7.5 - Comfort criteria are repeated as they appeared in Table 7.2, unnecessarily. Up to 90
krnIh the headlight sight distances are the same as normal stopping sight distance but note that the
resulting length of sag vertical curve is more than three times the crest curve length for the same
change of grade.

1. The design speed of the road in both the horizontal and vertical directions should be of the
same order.

2. Horizontal and vertical sight distances should be considered together. While each may be
adequate when taken separately they may be deficient in combination.

4. Particularly on two-lane two-way roads the need to provide adequate overtaking

opportunities may override considerations of co-ordination of alignment.

4. Sharp horizontal curves should not be introduced at or just beyond the top of a vertical
curve as the change in direction may not be obvious to a driver, particularly at night.

5. Provision of reverse horizontal curves in association with crest vertical curves is

undesirable particularly if the horizontal curves are short, because the reverse curvature may
not be obvious to the driver.

6. As far as practicable, crest vertical curves should be located away from intersections, road-
rail level crossings and the like.

7. Adequate visibility should be provided at the transition between undivided and divided
roads and on the approaches to intersections. In such situations if crest vertical curves
cannot be avoided, they should be longer than the minimum requirement.

Guidelines that relate more specifically to appearance considerations include:

1. Desirably a horizontal curve should be longer than any associated vertical curve, and it
should begin before the start of the vertical curve.

2. A sag vertical curve should be located on the horizontal curve rather than on the adjacent
tangent close to the start of the horizontal curve.

4. Rolling grades on isolated straights between curves should be avoided.

5. Short tangents between sag curves should be avoided.

6. Broken-back crest or sag curves should be avoided.

7. A disjointed effect occurs when the start of a horizontal curve is obscured by an intervening
crest while the continuation of the road is visible in the distance.

8. A short sag curve on a straight is undesirable, whereas a long sag curve may provide a
pleasing appearance.

9. A tight-radius curve which is visible at the end of a long straight appears to look sharper
than it is. In this type of situation a radius as large as practicable is desirable for both safety
and aesthetic reasons.

Examples on Vertical Alignment

Example 1:
The road is on a straight grade: chainage 13 789.642, RL 124.781 to
chainage 13 927.558, RL 115.989

Grade = 100 (difference in level) / ( difference in chainage)

= 100 (115.989 - 124.781) / (13927.558 - 13789.642)
= - 6.37 %

The ease with which vertical curves can be computed is shown in the following example. Given
that L-=90m, gl = -3.2 %, g2 = +1.8%, PVI at 9 + 9, and elevation = 142.01, determine the
location of the low point and elevations on the curve at some relevant stations, as well as at the low
point.
Solution: q 0 (-.)

1. A = 1.8 - (-3.2) = 5

2. PVI - 0.5 L = PVC

PVC at ( 9+9 ) - 45 = 8 + 64
PVI + 0.5 L = PVT
PVI (9+9) + 45 = 9 + 54
PVT -PVC=L
(9 +54) - (8 + 64) =90 Check

3. Elevation ofPVI = 142.01

45 m at 3.2 % = 1.44 (see Figure El)
Elev. PVC = 0.81
Elevation PVT = 142.82

x = G~ L (12~ .
= 3.2 x 90/5 = 57.6 (from the PVC)
 Example: Elevation at 9+00 = 143.45 - (0.032 x 36)
= 142.30

6. Mid-chord elevation:
(143.45 + 142.82) /2= 143.135

7. Tangent offset at PYI (d):

d = (difference in elevation of PVI and mid-chord) / 2
= (143.135 - 142.01) /2
= 0.5625 m

8. Tangent offsets are computed by multiplying the distance ratio squared, (xl L/2)l, by the
maximum tangent offset (d). See table E1.

Table El- Parabolic Curve Elevations by Taneent Offsets

Station Tangent Elevation Tangent Offset Curve Elevation

PVC 8 + 64 143.45 (0/45)2 x 0.5625 =0 143.45

8 + 80 142.94 (16/45)2 x 0.5625 = 0.0711 143.01
8 +00 142.30 (36/ 45)2 x 0.5625 = 0.3600 142.66
PVI 9+09 142.01 (45/45)2 x 0.5625 = 0.5625 142.57

Low Point 9 + 21.6 142.24 (32.4 /45)2 x 0.5625 = 0.2916 142.53

9+40 142.57 (14/45)2 x 0.5625 = 0.0544 142.62
PVT 9 +54 142.82 (0/45)2 x 0.5625 =0 142.82

9. The computed tangent offsets are added (in this example) to the tangent elevation in order
to determine the curve elevation.

An up-grade of 6% and a down-grade of 2% are connected by a 400m vertical curve as shown

below. The intersection point of the two grades has a reduced level of 100m. Calculate the
reduced level at each 100m along the vertical curve.

R\..'=\OO (W')
/'
e---_.
\

l-= 400 t-) ~

A Vertical Curve connecting two grades.

 L
e = 800 (g 1 - g2)

In this case, L = 400m

gi = +6%
and =
g2 -2%

400
e = 800 [6 - (-2)]

4ex2
y= L2

4 x 4 x 1002
= 4002

Horizontal distance Reduced level on Offset to Curve (m) Reduced level on

from A (m) tangent (m) curve (m)
0 88 0 88
100 94 1 (= y) 93
200 100 4 (= e) 96
300 98 1 (= y) 97
400 96 0 96

Example 4:
A highway is being designed with a 113 krnIh design speed and, at one section, an equal tangent
vertical curve must be designed to connect grade of + 1.0% and -3.0%. Detennine the minimum
length of vertical curve required assuming provisions are to be made for minimum SSD and
desirable SSd.

assumed average running speed = 93 krnIh

If the conservative worst-case value of G is used (i.e., -3.0%),
VI2
SSD = VI x 0.7 + 254 (f + 0.01 G)
 (93)2
= 93 x 0.7 + 254 (0.28 _ 0.01 x 3) = 201.30 m

Lm= AD2
200 (-{hI + ..J h2)2

_ 4 (201.30)2
- 200(~ + ..J 0.2)2

= 350.96m
Since 350.96m > 201.3 or the assumption that SSD < L is valid.

For desirable SSD,

(113)2
SSD = 113 x 0.7 + 243 (0.28 _ 0.01 x 3)

= 280.19m

. L 4 (280.19)2
Ie -
. m - 200 (~ + ..J 0.2)2

= 679.95 m

Again the assumption that SSD < L is valid.

Example 5:
Design the vertical curve for this crest for design speed of 110 km/h, spray sealed road.
Design for stopping sight distance:

(1) { h2 = 0.2 m
{h2 =0m
{ hI = 1.15 m
(2) Overtaking ESD (Establishment Sight Distance) hI = h2 = 1.15 m
y2
Ds = 0.7 V + 254f @ 110 km/h, f == 0.35

(110)2
= 0.7 (110) + 254 (0.35)
Assume Ds < Lv
L= Al)2_
200 ({hI + ...j h2)2

:. L = 7.35 x (213.1)2
200 ({[15 + ...j 0.2)2
=723 m

L= Al)2_
200 ({hI + ...j h2)2

_ 7.35 (213.1)2
- 200 (1TTI + -{O)2

Lv = 1451 m
(2) Establish Sight Distance

(from Table 6.4 AUSTROADS )

7.35 x (1100)2
920
Lv = 9667 m
This is very high, hence is not a good practice to provide overtaking sight distance at this point.

Calculate the minimum length of vertical curve over a crest in a road to satisfy the stopping sight
distance requirements, if the approach gradients to the crest are 6% from one direction and 5.75%
from the opposite direction, and the design speed of the road is 65 km/h.

y2
Ds = 0.7 V + 254 f

= Q.7 x 65 + (65)2/ (254 x 0.468)

=81 m
 18

(6 + 5.75)(81)2
= 462

An elevated urban road is to pass over the lowest part of a sag vertical curve in an existing local
road, 90m long and linking gradients of 4.5% and 3.5%. Calculate the minimum clearance of the
new overpass over the existing road level for SSD requirements.

y2
SSD = 0.7 x V + 254 x f (f = 0.455 @ 70 krn/h)
(70)2
= 0.7 x 70 + 254 x 0.455
= 91.4 m (say 92 m)

L = 90 m ie. < 92 m (ie. L < Ds)

. L - 2D _ 200 {\/H - hI + --J H - h2)2

.. - A

90 = 2 x 92 _ 200 (--JH - 1.8 + --JH - 0.6)2

4.5 + 3.5

A down-grade of 2% and an up-grade of 6% are joined by a sag vertical curve. How long should
the vertical curve be to provide a headlight sight distance of 200 m?

Solution:

AD2
L= 150 + 3.5D
 8 x 2002
-150+3.5x200

= 376 m

A vertical curve is to be constructed between an ascending 3.5 % grade and a descending 4 %

grade. The required safe stopping sight distance is 300m, the dangerous object is considered to be
the carriageway level, and the motorist's eye height is 1.15 m. Determine the minimum length of
vertical curve that will satisfy this sight distance requirement.

Solution:

q=-4, p=+3.5, D=300m

therefore, e = (p sJJ D (-4 8~6 300 = - 2.82
But hI, the eye height, is equal to 1.15 m, therefore the equation for L>D will be valid.
When L>D, the length of the vertical curve is given by

L= AD2
100 (" 2h 1 + "2h2)2

= [7.5 (300)2] / 100 [ (2 x 1.15)0.5]2

= 2935 m

Determine the minimum length of curve required to connect a descending 4 % grade to an

ascending 3 % grade. The design speed of the road is 100 km/h and acceptable radial acceleration
is 0.3 m/sec2 .

V211
L = 1300 a
= (100)2 (7) / 1300 (0.3)
= 180m

Find minimum curve length where approach grade is +4 % and departure grade is -5.9 % at 120
km/h and overtaking is not required.
Sight distance must never be less than SSD (table 6.2 AUSTROADS), and the minimum allowed at
120 km/h is 250 m. This agrees with the tabulated value in table 7.4 (AUSTROADS) so we can
use a K value of 135.
Grade change = A = -5.9 - (+ 4.7) = - 10.6
L = KA = 135 (10.6) = 1431 m (ignore sign)
check table 7.1 (AUSTROADS). A = 10.6 does need a curve, minimum length = 100 m to 150 m
not critical. Appearance o.k at 1431 m.
Therefore minimum length =1431 m.

Approach grade + 0.865%, departure grade + 0.043%. Determine minimum length for
continuation sight distance at 100 km/h.

These are both up hill grades but the curve is still called a crest curves.
A = + 0.043 - ( + 0.865) = - 0.822
Table 7.4 (AUSTROADS) shows K = 200
Lmin. = KA = 200 x 0.822 = 165 m.

Example 13:
Overhead bridge has clear height of 0.43 m. What sight distance will be available (1.8 m to 0.6 m)
at 70 km/h with a grade change of 7 % and curve length of 80 m ?

A=7, L= 80
C = 200 ( H - hI + H - h2 )2
= 200 (4.3 - 1.8 + 4.3 - 0.6)2 (check Table 7.6, AUSTROADS, shows C between 2200 and
2600)

For L<D, Eqtn (1) L = 2D - CIA, D = (L + C/A)12

= ( 80 + 2456.6 17) I 2
= 215.5 m L< D , therefore valid

For L>D, Eqtn (2) L = LC/A

= ( 80 x 2456.6 ) I 7
= 167.6 m L< D , therefore not valid

Sometimes the results are confusing so it is best to work both equations, although in this case the
first answer was correct.
Chapter 6
Climbing Lanes
Climbing lanes are extra lanes on upgrades to enable vehicles to overtake slow moving
vehicles.

Overtaking lanes can also be used on flat sections and the shape of the road in plan view may
be the same in both cases. Overtaking lanes (including climbing lanes) are particularly
important for 2-lane, 2-way roads but climbing lanes are often used for divided roads on long
upgrades (see AUSTROADS 9.2).

Volume warrant: for climbing lanes use Table 9.1 (AUSTROADS) but reduce the AADT
(Average Annual Daily Traffic) volumes by 20%. For example, for moderate overtaking
opportunities for the last 5 km and 10% of slow vehicles, the AADT becomes 2240 instead of
2800.

AADT is for both directions. Percentage of slow vehicles would have to be determined by a
speed survey on an upgrade as empty modern semi-trailers are more likely to overtake other
vehicles than be overtaken and hence just counting trucks and trailers would not be sufficient.

Speed drop warrant : based on whether the design truck's speed drops to less than 40 km/h
from an assumed approach speed of 80 km/h. See chart Fig. 9.1. (AUSTROADS). Lower
approach speed to be used if road design speed is lower.

Minimum length warrant : the resulting design length of the full width climbing lane should
be at least the minimum values given in Table 9.3 (AUSTROADS), (minus the taper lengths).
The full width lane is at least 90m longer than the length for which the speed is below 40
km/h. See 4. and 5. under Design Procedure.

Special case warrants: where the traffic volume is not met there may still be need for a
climbing lane as for example in the following cases :-

where a quarry is located near an upgrade and many slow-moving trucks join the
traffic;

where the upgrade is long and the road geometry has provided few other chances for
overtaking (ESD and CSD not often available); and

where there have been too many accidents apparently attributable to unsafe
overtaking.

2. Approximate the vertical curves with a series of straight grades about 100m long and
hence obtain a list of values and lengths of grades, e.g. 100m of 1%, 100m of 3%,
600m of 5%, etc.
3. Determine the positions on the road (the chainage) where the speed of the design
truck drops to 40 k:m/h and where it rises to 40 km/h again later. Known as the V 40
positions. See "Finding V 40 positions" following.

4. The full width climbing lane starts 60m before the first V40 position and ends 30m
after the second V 40 position.

5. Adjust the start and finish positions to suit the conditions. End on a straight if
possible. If road becomes steep again later consider continuing the climbing lane to
join both upgrades, etc. See 9.5.1 (AUSTROADS).

6. At the approach end there is a taper of length (0.33 x 85th percentile speed of all
traffic x lane width) metres, and on the departure end there is a taper of length (0.5 x
85th percentile speed of all traffic x lane width) metres.

7. Determine the location of signs: "Overtaking Lane 500 m", "Keep Left Unless
Overtaking", "Left Lane Ends 200m", "Form One Lane". Road paint is standardised
but has to be shown on the drawings.

For an isolated hill there may be need for climbing lanes up both sides. These will overlap at
the top giving a carriageway 4 lanes wide plus shoulders. Often this is just at the point of
deepest cut through the hardest rock, so the installation is not cheap.

Truck reaches first upgrade at 80 k:m/h and begins to decelerate. Hence use the
DECELERATION chart.

Follow the curve for the first grade from 80 km/h for the same horizontal distance on
the chart as the grade extends for on the approximated longitudinal section. This
gives the truck's speed at the end of that grade. Note that for an approach speed of 70
k:m/h the distance measurement on the chart is not from the beginning of the chart but
from where the 70 k:m/h line hits the grade line.

Transfer this speed to the next grade line, i.e. if the first grade reduces the truck speed
to 71 k:m/h, then enter the second grade at 71 km/h and follow it.

Note the distance where the speed drops to 40 km/h but keep on following the
deceleration grade lines until the grades begin to decrease.

Transfer the slowest speed reached across to the ACCELERATION chart and enter
that first grade line with that slowest speed.

Continue following the grade lines for the distance of each that the appropriate grades
occur on the longitudinal section, until the truck speed reaches 40 km/h again.

A two-lane, two-way road has restricted opportunities for overtaking in the 5 kIn preceding
an upgrade. AADT is 1000 vpd and 11% of the traffic is slow vehicles. Design speed is 90
km/h.
0%, 100 m of 1%, 100 m of 3%,500 m of 6%, 100 m of 3%,100 m of 1%, 100 m of -1 %,
400 m of -3%.

Volume warrant: On the "restricted" line of Table 9.1 we have AADT 1330 which reduces to
1064, at 10% slows. 1000 is 6% better than 1064 and 11% slows is 10% worse than the table
10% slows figure. Hence our situation is slightly worse than the warrant, i.e. the warrant is
satisfied.

1. Enter the deceleration chart at 80 km/h and follow the 1% grade line down until the
horizontal distance is 100 m from the first point. Speed reached is 78 krn/h.

2. Enter 3% grade line at 78 km/h and follow it down as before. Speed reached is 75
km/h.

3. Enter 6% grade line at 75 km/h and follow it for 500 ill horizontally. On the way we
pass the 40 km/h line and the first V 40 position is 300 m past the start of the 6%
grade. At the end of the 6% grade the speed reached is 23 km/h.

4. Enter the acceleration chart on the 3% upgrade line (down grade lines are shown with
a - sign) at 23 km/h and follow it for 100 ill. Speed achieved is 30 krn/h.

5. Enter the 1% upgrade line at 30 km/h and follow it for 100 m. Speed achieved is 39
km/h.

6. Enter the -1 % (downgrade) line at 39 km/h and follow it until 40 kmJh is reached-
another 10 or so metres.
Chapter 7
Roadway Capacity
Capacity analysis is fundamental to the design of roads, and provides the basis for
determining the number of traffic lanes required at any point in a road network, having regard
for the volume and composition of traffic and the prevailing roadway and traffic control
conditions. Alternatively, for a given number of traffic lanes, capacity analysis provides a
means of determining the traffic carrying performance of a road or any element of it under
the prevailing roadway, traffic and control conditions.

This chapter defines some basic road capacity terms, and presents specific information
relating to the capacity of two-lane rural roads, multi-lane roads, freeways, signalised and
unsignalised intersections and urban arterial roads.

Capacity is the amount of traffic a road or roadway component can accommodate. The
capacity of a road or intersection is the maximum rate at which vehicles can pass along the
road or through the intersections. The capacity of a single traffic lane or intersection depends
primarily on the ability of a driver to follow the vehicle ahead at a minimum headway
consistent with safety and ability.

Traffic flow may be classified into two broad categories :-

(i) Uninterrupted Flow - In this condition vehicles are not required to stop by any cause
external to the traffic stream, such as traffic control signals, that c~use interruption to
traffic flow.

(ii) Interrupted Flow - In this condition, vehicles are required to stop by causes outside
the traffic stream, such as signs, traffic signals or significant friction from abutting
development.

A term that is used to classify the varying conditions of traffic flow that take place on a
highway is level of service. The various levels of service range from the highest level, which
is found at a flow where drivers are able to travel at their desired speed with freedom to
manoeuvre, to the lowest level of service, which is obtained during congested stop-start
conditions, see Figure 2.1. The levels of service are:-

"A" - high speed can be safely sustained, unrestricted safe overtaking at will

"B" - high speed can be maintained most of the time, overtaking frequently possibly with
little delay
"C" - stable flow at more or less constant speed dictated by slower vehicles, overtaking
opportunities limited

"D" - stable flow, speed restricted, lane changes mostly forced

"E" - flow at capacity, flow becoming unstable, frequent speed changes

"F" - continuous stopping and starting, forced flow, volume well below capacity, in the
extreme total standstill, volume and flow both zero

The practical speed range will vary with the category of road but, in general :-

(i) L.S. "A" cannot exist unless the geometry of the road can allow a sustained speed of
110 km/h.

(ii) If geometry of the road restricts the sustained speed to no more than 60 km/h, only
L.S. "E" can be provided.

Each level of service represents a RANGE OF CONDITIONS, defined by the operating

speed and degree of congestion (vie ratio). BOTH parameters must be within their ranges or
given L.S.

For road capacity analysis, the general terrain of a road is classified as follows :-

(i) Level terrain - any combination of grades and horizontal and vertical alignment
permitting heavy vehicles to maintain about the same speed as passenger cars.

(ii) Rolling terrain - any combination of grades and horizontal and vertical alignment
causing heavy vehicles to reduce their speeds to substantially below those of
passenger cars, but not causing them to operate at crawl speeds for any significant
length of time.
(iii) Mountaineous terrain - any combination of grades and horizontal and vertical
alignment causing heavy vehicles to operate at crawl speeds for significant distances
and/or at frequent intervals.

The material in this chapter is based on the traffic carrying concept, namely the ability of
traffic facility to carry traffic under given roadway, traffic and control conditions. However,
two other concepts, the environmental capacity and the economic capacity, may also be used.
The environmental capacity of a road is a measure of the volume of traffic it can
accommodate while still maintaining acceptable environmental standards. These standards
may relate to traffic noise, air quality, safety, visual intrusion, freedom of local movement
including pedestrians and bicyclists, and the like. This concept is usually applied to
residential streets. While there are no clear-cut guidelines for environmental capacity, it is
generally agreed that the environmental capacity of residential streets is of the order of 3000
to 4000 vehicles per day.

The economic capacity of a road is a measure of the volume of traffic that is required to use it
to justify it's provision economically. Alternatively, it is the volume of traffic required to
justify any given improvement economically. All significant proposals for road
improvements should be subject to the appropriate level of economic analysis.

Hourly volume is the actual hourly demand volume for the highway in vehicles per hour,
given the symbol V. Generally the highest 24 hour hourly volume (i.e. peak-hour volume) is
used for V in traffic analysis computations.

The peak-hour factor accounts for the non-uniformity of traffic flow over the peak hour. It is
denoted PHF and is typically defined as the ratio of the hourly volume (V) to the maximum
15 minute rate of flow (V 15) expanded to an hourly volume. Therefore,

Service flow is the actual rate of flow for the peak 15 minute period expanded to an hourly
volume and expressed in vehicles per hour. Service flow is denoted S.F. and is defined as :-

V
PHF

2.3 Freeways

A freeway can be regarded as being composed of three different types of components, namely
basic freeway segments, weaving areas and ramp terminals. Figure 2.2 shows the various
types of freeway components.
 \
(t:\) ouT~,~e
Ra..-p
~
OT"
t~lu~
'Nt..Gto...'''j
",p
"""ew'\~\J\l~""..s lE) lSol..:l' _00\ 7 o-fF - ..,..~
.
we", '1M' A 'f'~

~<0
-----
(8) Me. •.•~~ Aor~
--
~,'o\..Ja...cJ 1 y
-" "
(f) c..ot'\sec.uT"\';'~ e"" - .,..~

Di,,~'5~

~2=$f
<..C:)
/"/
Ot'\ _ fl-.p ~ 110....,J
~"'\."
"7 e.j: F -
r~ "" i rt..... a.vX i1"r7 (AM.~

0,,')
7
0\'\- rA~ follo •.••.~
\ 1:>1
(!)f~ - yo ov-vp '"-l i r1-.. V' eo

Basic freeway segments are located outside the influence area of any weaving area or ramp.
In general, the influence area of weaving areas and ramps may be taken as follows :-

(i) Weaving areas - 150m upstream of the merge point marking the beginning of the
weaving area, and 150m downstream of the diverge point forming the end of the
weaving area.

(ii) Entrance ramps - 150m upstream and 750m downstream of the ramp junction.

(iii) Exit ramps - 750m upstream and 150m downstream of the ramp junction.

These influence areas are shown in Figure 2.3. They apply for stable flow conditions.
During congested or flow breakdown conditions, they can be much greater.
~Q.. ~'-L ~V.-c..-,
£ --, --......-T

.I

t
-:t,.... I~__c... L.. e\ y ~ S G /- <.
/- •••.• c.. ....,.. 7 Co ..--.f' C> "'" ......,:;;fi
Figure 2.3
A basic freeway segment is defined as a section of a divided highway having two or more
traffic lanes in one direction, full access control and traffic that is unaffected by merging or
diverging movements near ramps or lane deletions.

The term maximum service flow, M S F, can be defined for a given level of service i as the
highest service flow that can be achieved, while maintaining the specified level of service i,
assuming ideal roadway conditions. Since ideal conditions specify the presence of passenger
cars only, and it is desirable to make the maximum service flow rate independent of the
number of highway lanes, M S F i is in units of passenger cars per hour per lane (pcphpl).
Maximum flow rates and accepted level-of-service criteria for freeway segments with design
speeds of 110, 100 and 80 km/h are given in Table 2.1. along with approximate densities and
operating speeds for each level of service. One of the basic relationships underlying Table
2.1 can be expressed as:-

v
Co x (_)0
:J C 1

maximum service flow rate per lane for level of service i under ideal
conditions in pcphpl.

capacity per lane under ideal conditions for a freeway with a design speed j,
i.e. 2000 pc/h/lane for j = 100 km/h or 110 km/h and 1900 pc/h/lane for j = 80
km/h.
V
(-)i
c = maximum volume/capacity ratio which can be accommodated at level of
service i, obtained from Table 2.1.

The concept of a maximum service rate provides an important bench mark for determining a
highway's level of service but, since ideal conditions are seldom realised in practice, some
method of converting the maximum service flow rate into an equivalent service flow rate
(which accounts for actual prevailing conditions) is needed. In calculating service flow rates
under prevailing conditions, corrector factors are used along with the number of directional
lanes giving:-

total service flow rate in one direction under prevailing roadway and traffic
conditions for level of service;
 adjustment factor for lane width and/or lateral clearance restrictions, obtained
from Table 2.2;

f _ 1
HV - 1 + PT (ET - 1) + PB (EB - 1) + PR (ER - 1)

PT,PB and the proportion of trucks, buses and recreational vehicle respectively in
PR ;;: the traffic stream expressed as a decimal;

ET,EB and the average passenger car equivalents for trucks, buses and recreational
ER ;;: vehicles obtained from Table 2.3; and

Fp=

The methodology is most often used in one of three applications :-

(i) Operational analysis - An operational analysis is an analytic evaluation of operations

on an existing freeway segment. The same type of analysis may be applied to
evaluate probable operating conditions on a future facility.

The following step-by-step procedure can be used :-

(a) convert all volumes to peak 15 minute flow rates;

(b) adjustment factors and passenger-car equivalents for prevailing conditions are
obtained from the appropriate tables :-

fw (Table 2.2)
ET (Table 2.3, 2.4, 2.5 or 2.6)
ER (Table 2.3 or 3.7)
EB (Table 2.3 or 2.8)
fHV (compute from equation 2.7);

determine the MSF or ~ ratio;

compare the MSF or the V ratio to the criteria of Table 2.1 to determine level of
c
service. MSF or ~ must be less than the tabulated criteria to fall within a given level
c
of service; and

using the MSF or V ratio, Figure 2.7 is used to find the approximate density of the
c
traffic stream, and Figure 2.8 is used to find the approximate average travel
speed of the traffic stream.
(ii) Design - A design analysis is made to determine the number of lanes required on the
freeway to provide the desired level of service for the forecasted traffic volume and
traffic characteristics.

(a) convert the directional design hour volume, DDHV to an equivalent peak flow rate,
which is set equal to the service flow, SF :-

_DDHF
SF
-PHF
(b) find all adjustment factors and passenger car equivalents, based on forecast traffic
characteristics and select design standard;

(c) select a design ~ ratio, or corresponding MSF, from Table 2.10; and

N-- V
SF
[Cj x (c) x fw x fHVX fp]

N- SF
- [MSF x fwxfHV x fp]

2,000 pcphpl for 100 and 110 kmIh freeway elements, and 1,900
pcphpl for 80 kmIh freeway segments.

(iii) Planning - the objectives of a freeway capacity analysis at the planning level are
principally the same as those of a design analysis. The primary difference between
design and planning analysis is the amount and detail of information available as
inputs into the analysis.

The following steps are involved in conducting a planning analysis:-

(a) convert AADT to DDHV using

DDHV=AADTxKxD

AADT= forecast average annual daily traffic, in vpd;

DDHV= directional design hour volume, in vph;
K= percentage of AADT occurring in peak hour; and
D= percentage of peak hour traffic in the heaviest direction.

Values of K and D should be based on local or regional characteristics. If such

information is unavailable, the following approximation may be used:-
 Suburban Freeways 0.10 - 0.15
Rural Freeways 0.15 - 0.20

Urban Circumferential Freeways 0.50

Urban Radial Freeways 0.55
Rural Freeways 0.65;

(b) select an appropriate value of SFL, the service flow rate per lane, from Table 2.11 for
the prevailing truck percentage and terrain, and for the desired LOS; and

(c) compute the number of lanes that would be required in each direction of the freeway
using :-

DDHV
N = [SFL x PHF]

On many long and/or steep upgrades, it is necessary to consider adding a climbing lane for
trucks and other heavy vehicles.

CT= 2~~

CT= 1~~

SFT=

CT=

V
(-)i = ~ ratio for loss i, from Table 2.1 for operational analysis, or from
c
Table 2.10 for design.

Procedure for Detennining Passenger-car equivalents of trucks on composite

upgrades

In general, the following steps describe the solution for equivalent grade (refer to Figures 2.4,
2.5 and 2.6) :-

(i) Enter the appropriate truck acceleration-deceleration performance curves with the
initial grade and length of grade. Find the speed of trucks at the end of the first grade,
which is the speed at which they enter the second grade;
(ii) Find the length along the second grade which results in the same speed as found in
Step 1. This is used as the starting point along the second grade;

(ill) Starting with the length found in Step 2, add the length of the second grade and find
the speed at the end of the second grade;

(iv) If there are additional grades, repeat step 1 through 3 for each subsequent grade until
the fmal speed is found; and

(v) Enter the truck performance curves with the final speed of trucks and the total length
of composite grade to find the equivalent uniform grade percent, which may be used
in finding ET.

2.1 A six lane urban freeway (three lanes in each direction) is on rolling terrain with 110
km/h design speed, 3m lanes, with obstructions ofO.6mfrom both roadside and
median edges. A directional weekday peak-hour volume of 2200 vehicles is observed
with 700 vehicles arriving in the most congested 15 minute period. If the traffic
stream has 12 percent trucks, 10 percent buses and 2 percent recreational vehicles,
determine the level of service.

2.2 Considering the conditions described in problem 2.1, how many additional vehicles
can be added to the peak hour before the freeway reaches capacity?

23 An older four-lane urban freeway with a 100 km/h design speed serves a directional
peak hour volume of2100 vph with 6 percent trucks and a PHF of 0.95. Thefreeway
has 33m lanes, obstructions immediately at the pavement edge at both the roadside
and median, and generally rolling terrain. Evaluate the level of service at the facility.
Determine how much additional traffic could be accommodated before reaching
capacity. field studies of average travel speed indicate that during the peak 15
minutes offlow, speed is 55 km/h.
2.4 A six-lane freeway with a 110 kmIh design speed carries a peak hour volume of 3500
vph in one direction, with 5 percent trucks and a PHF of 0.85. Thefreeway has 3.7
lanes, a 6m clear median, and rock cliffs 0.6mfrom the pavement edge.
The freeway segment in qestion is the composite grade illstrated in Figure 2.1.
Determine the level of service at which the freeway operates during peak periods -
upgrade and downgrade.

800 (WI) -'6ca lM) \ f, 0 D (\rY\)

~
PIt' _ ~~ I
2.5 An extended section of freeway in level terrain in an urban area is to be designed to
operate at level-of-service C. The section is expected to carry a directional design
volume of4500 vph, with 12 percent trucks, no buses or RV's, and a PHF of 0.90. The
driver population consists primarily of commuters. Determine the number of lanes
which must be provided through the section.

2.6 A long segment of rural freeway is to be designed for level-of-service B. The DDHV i
s 2200 vph (weekday), including 20 percent trucks and a PHF of 0.95. A 13 km
segment of level terrain is followed by a 3 percent sustained grade of 1.6 km. How
many lanes will be required on both the level terrain and sustained grade segments?

2.7 Afreeway is being planned to service a radial route in an urban area. It is expected
to heave an AADT of 80,000 vpd, with approximately 10 percent trucks. A PHF of 0.9
is anticipated. The region through which it will travel has generally rolling terrain.
Determine the number of freeway lanes that will likely be required to provide for LOS
C?

The approach that will be taken to determine level of service will be to compute the volume-
to-capacity ratio (vie) of the roadway and to compare it with the maximum volume to
capacity ratios for specified level of service, as given in Table 2.1. To arrive at the roadway's
volume-to-capacity ratio, Equation 2.6 is rearranged giving :-

vie = S_F _
Cj xN x fwxfHVXfp
 SF = V 15 x 4 = 700 x 4 = 2800 vehlhr
and Cj = 2000 PCPHPL (110 krn/h design speed)
fp = 1.0 (weekday, table 2.9)
fw = 0.85 (3.0m lanes, obstruction 0.6m on both sides, assumed with 0.6 shoulder,
Table 2.2)
ET = 4.0, EB = 3.0, ER = 3.0 (rolling terrain, Table 2.3)

From Equation 2.7 :-

f _ 1
HY- 1 + PT (ET -1) + PB (EB - 1) + PR (ER - 1)

1
fHY= 1 + 0.12 (4 - 1) + 0.10 (3 - 1) + 0.02 (3 - 1)
so, substituting, we find that
vlc _ 2800
- 2000 x 3 x 0.85 x 0.625 x 1.0

which gives a LOS D from Table 2.1, since the maximum vlc for LOS Cis 0.77 and the
maximum vie for LOS D is 0.93 (i.e. 0.77 < 0.878 < 0.93).

The freeway segment in question is the composite grade illustrated in Figure 2.1. Determine
the level of service at which the freeway operates during peak periods - upgrade and
downgrade.

The key to the upgrade solution is to find an equivalent grade of 3.2 kID in length which
results in the same final speed of trucks as the sequence of grades illustrated in Figure 2.1.
This is done using the performance curves for a 200 lb/hp standard truck.

The performance curves are entered by constructing vertical line 1 at 800 m and finding the
intersection with the 2% deceleration curve. A horizontal line drawn through this point to the
vertical axis indicates a speed of trucks at 79 krn/h.

Vertical line 2 is constructed from the intersection of the 79 krn/h horizontal line and the 3%
deceleration curve, indicating that trucks enter the 3% grade as if they had been on it for 300
m. Vertical line 3 is drawn at the 300 + 800 or 1100 m mark and carried to the intersection
with the 3% deceleration curve. A horizontal line through this point to the vertical axis
indicates a speed of 64 krn/h at the end of the 3% grade.

The 64 krn/h horizontal line, however, does not intersect with the 1% deceleration curve.
This is because trucks entering a 1% curve from a 3% curve would be expected to accelerate.
Thus, vertical line 4 is drawn from the intersection of the 64 krn/h horizontal line with the 1%
acceleration curve, indicating that trucks enter the grade as if they had travelled on it for 640
m.

Verticalline 5 is constructed from the 640 + 1600 = 2240 m mark. The intersection of this
line with 1% acceleration curve yields the final speed of trucks of 80 krn/h.
The solution for an equivalent grade is now an unknown percent grade of 3.2 Ian that results
in a final truck speed of 80 km/h. This, however, would be misleading. The minimum truck
speed of 64 kmlh is reached at the end of the 3% grade segment, and it is at this point that
trucks would have the maximum impact on operations. Therefore, the solution point sought
should be an unknown grade of 800 + 800 or 1.6 Ian that results in a final speed of trucks at
64km/h.

This is given by the intersection of vertical line 6 (constructed at 1600m) and the 64 kmlh
horizontal, and yields an equivalent grade of 2.8 percent, which will be taken at 3%, 1.6 Ian
long, for the analysis.

Then :-

vie SF _
- [Cj x N x fw x fHVx fp]

3500 .
SF = 0.85 = 4118 vph (gIven);

Cj = 2000 pcphpl (Table 2.1)

N = 3 (Given)
fw = 0.97 (Table 2.2, 3.7 m lanes, obs.one side at 0.6 m)
fp = 1.00 (Table 2.9)
ET(upgrade) = 7 (Table 2.4, 3% grade, 1.6 Ian length, 6 lanes)
ET (downgrade) = 1.7 (Table 2.3, level terrain);
1 1
fHY (upgrade) = 0.77 (fHV= 1 + Pr (ET - 1) = 1 + 0.05 (7 -1) = 0.77)
1
fHY (downgrade) = 0.97 (fHV = 1 + 0.05 (1.7 -1) = 0.97)

vie (upgrade) = 4118/[2000 x 3 x 0.97 x 0.77 x 1.00] = 0.92

vie (downgrade) = 4118 I [2000 x 3 x 0.97 x 0.97 x 1.00] = 0.73

From Table 2.1, the respective level of service are D for the upgrade and C for the
downgrade.

Figure 2.7 and 2.8 may be entered with the above vie values to obtain approximate speeds
and densities for the upgrade and downgrade conditions described. For the upgrade, speed is
74 km/h and density 25 pc/km/h; for the downgrade, speed is 87 km/h and density is 17
pc/km/h.

The relatively high value of vie for the upgrade might suggest consideration of a truck
climbing lane for this location.
 88
80
""\0.

12·
---1 I
,~ -1--------------------,%
5~

.~
48
L:
~ '+0
a
w
w
Q.. 32
(/)

:14

16 - - - - acceleration curve
--- deceleration curve
8

0·3 o· 6 o.~
~:~ 800
tl-0"'0 THOUSANDS~ OFt"IETER. OF GRADE
\- Fil:ure J.-4 I'er!ormatrt"c curw:s fur a standard truck (200 lb/I,p).
 ~~1·
w
z
:5 43·
i'
~ 37·1.

J.$ pC,!V.••..(t.M ~ 31-D

>-
l-
V; ,.,.g
z
w
0/
8,,, -----_._----

2 4 6 8 10 12 14 16 18 20 VOL./LN(lOOpeph)
(0.1) <0.2) (0.3) (0.4) (0.5) (0.6) (07) (0.8) (0.9) (1.0) ,·/e. Rolio •.•.
« capacity
••'" v/c fOlio bosed. on 2000 pcphpl valid only (or -100 and I/o KM?H design speeds

q{; t 8-Lones

g 1c.~~( £...)
80

7q((M/t.~.
::s bl:r
a
w
ILl
0...
(/)

u146
~
i= "
w
~
2
ILl
3Z
>
<:(

2 4 6 8 10 12 14 16 18 20 VOL/LN(lOOpcphl
(0.1) (0.2) (0.3) (OA) {D.5) (0.6) (0.7l (0.8l (0.9) (1.0) v/c Retio •.•.

« capacity .
••'" vIe re lio based on 2000 pcphpl valid only (or-IOO end 110KMPH design speeds

IdeAl Cond,.-t/i::>"S
~
a) 3.7 m m';"~v",,, lilt,,"! W',~7t.
b) 1.8 m m,·,,"""'''- I~~,..Q./ Cleetl"D"Ce be7~<.c..., T1..~e.oIJe of ~~
-t~t!l.vel 1''''Ie! a.,d -r/.,•. 1)ea.rcsT obs7we objec..7 0 •.• 7it-e nn"..Js/de. •
or
c) All P;".sse",~r c.",rJ .'-- 7A~ -r••••ff/c S7,.~~,..,..,.
d) [)r/..,_r c/'tJlrf1>.c,-re".is 'Trc ( -t5flc•1 "...vl~,- 7'#,,, 57,c.UH'
of ~&.,ktl()..7 c.o•.•••
 8a
80

12-
I
I

b~ 1/1 I
/11 I
II II
5& 1/
/1 /

.X
48 II{I/ / /
y- 1/ / /
1/ /'
C 40 ~I /
'/ / .,/
0 // --------------------------- 6 %

W 1// ./
W /"
7%
Cl.. 32 /,
11/
/

<J)

.?~ '/
II
, /
/"

/
1//

I G I"I - - - - acceleration curve

I
J --- deceleration curve
8

:t: S6

t o Lt8
w
w
'e; Lt-o

- - -- acceleration curve
--- deceleration curve

1 I I ~
..,.2. 4.5' "i·B
 ~> ---:------~

/ //./'
/ --::::?~------
------~-------l %
~/ I >Y': ./
_-------------, 1%

t:: If;l/»-~ - --~-~-----------'

~ 40
/l

'1///
\/~
>~_-~~---------------r---
---3%
2%

if // /
fJ)

32 l!i~/~:----- ---- -----------------~:I

iiij~/ -- 6%
!~!r::.--- -------·-------------7 °/0
16 !;V
"

- - - - acceleration curve
-- deceleration curve
 _l.f~.
w
b1'0 --

55·g -- ---c--=-
:z:
:5 ""3'
i:
:::.: 37.2-
...•.
u
::: 3/'0
>-
f-
Vi ~L/·8 ---- ---- ---
:z:
w
0
18,,, ---- --- --_ .. _-- --

2 4 6 8 10 12 14 16 18 20 VOL/LN{JOOpeph}
(Ol) (02) (03) (04) (OS) {0.6} (O?) (O.8) (0.9) (1.0)vIe, R atio~"
••.eapaei ty
••.•. vIe ro I io based. on 2000 pepllpl valid only for -10 0 and 110 KMPH design speeds

J. a-Lanes
q&
I 4- r
La S

80 ----~-I-
-s.
-'-
~ bL!-
a
w
w
0..
Vl

Ld 48
~
~
W
<..?
~
w
Z>2. ______ - _1--- --- ---j-,,1:--1-ci.1
>
<i

____________ 1
~s
."oU'-"co!.-i/C:=J/
I --I
IU
.. I"
I
___--,----~----r---I------1-- !: I
I

2 4 6 8 10 12 14 16 18 20 VOLI LN (lOOpeph)
(0.1) (0.2) (0.3) (0.4) (0.5) (0.6) (Q.?) (0.8) (Q.9) (LO) vie Ratio~"

•. eapaei ty
•.•. vie ralio based on 2000 pephpl valid only for-IOo andl!O KI-1PH design speeds

Idea.l Co"d,.T,o"1.s
~
Q) 3·1 Yn "";,,-,,.,.v..... fA" -e w-,'''/'Tt.
b) 1.8 m m/,,''''''v- Ip.Ttorq / clee:t.,-p." c e be 7l-V~e..., rf.~ e.oIJe ()I- /'h-e.
~"'t1I.\lel I •• nc.s lJI.,d ft.•. ne ••,.csT 66S7f>..c.le. or C'bjec..70..., 71te t'"'DDo..Js/de. •
c) t:l11 P().$senJ~'- C."rS ::... ~~ T"'",ff'"c S7,.~'t,.,..,.
d) [)t'"',·.nr c.J,,,.1't1-e-"1'e,,./s1" cf "'!Jf,·c.1 ~f 1J"C.Ktlo.7 c.l> •••••..,..,~7~,-7'#,~51re.It,...,~.
f
 Design Speed Design Speed Design Speed
I Level Density 110 kmIh 100 km/h 80 km/h
of
Service pc/krn/lane Speed· v/eb MSfC Speed vie MSF Speed vie MSF
km/h km/h km/h
A ~7.5 ~96 0.35 100 - - - - - -
B ~12.5 ~ 91 0.54 1,100 ~80 0.49 1,000 - - -
C ~ 18.8 ~ 86 0.77 1,550 ~75 '0.69 1,400 ~69 0.67 1,300
D ~ 26.3 ~74 0.93 1,850 ~67 0.84 1,700 ~64 0.83 1,600
E ~ 41.9 ~48 1.00 2,000 ~48 1.00 2,000 ~45 1.00 1,900
F >41.9 <48 d' d <48 d d <45 d d
I

I
a. Average travel speed.
b. Volume/capacity ratio ..
c. Maximum' rate of service' flow per lane under ideal conditions, rounded to the nearest 5OpclMane.
d. Highly variable.

Table 2.2
Adjustment Factor for Restricted Lane Width and Lateral Clearand for Freeways

Adjustment Factor, fw
Distance from Obstructions on One Obstructions on Both
Travelled Side of the Roadway Sides of the Roadway
Pavement Lane Width (m)
(m) 3.7 3.3 3.0 2.7 3.7 3.3 3.0 I 2.7

4-Lane Freeway (2 Lanes each direction)

~2.0 1.00 0.97 0.91 0.81 1.00 0.97 0.91 0.81
1.5 0.99 0.96 0.90 0.80 0.99 0.96 0.90 0.80
1.2 0.99 0.96 0.90 0.80 0.98 0.95 0.89 0.79
0.9 0.98 0.95 0.89 0.79 0.96 0.93 0.87 0.77
0.6 0.97 0.94 0.88 0.79 0.94 0.91 0.86 0.76
0.3 0.93 0.90 0.85 0.76 0.87 0.85 0.80 0.71
0.0 0.90 0.87 0.82 0.73 0.81 0.79 0.74 0.66

6- or 8-Lane Freeway (3 or 4 Lanes each direction)

~2.0 1.00 0.96 0.89 0.78 1.00 0.96 0.89 0.78
1.5 0.99 0.95 0.88 0.77 0.99 0.95 0.88 0.77
1.2 0.99 0.95 0.88 0.77 0.98 0.94 0.87 0.77
0.9 0.98 0.94 0.87 0.76 0.97 0.93 0.86 0.76
0.6 0.97 0.93 0.87 0.76 0.96 0.92 0.85 0.75
0.3 0.95 0.92 0.86 0.75 0.93 0.89 0.83 0.72
0.0 0.94 0.91 0.85 0.74 0.91 0.87 0.81 0.70

Vehicle Type Type of Terrain

Level Rollinq Mountainous

Trucks, ET 1.7 4.0 8.0

Buses, EB 1.5 3.0 5.0

Recreational 1 .6- 3.0 4.0

Vehicles, ER
GRADE LE:"GTH . PASSENGER·CAR EQUIVALE:'<"T. Er
(%) (KyV ~-LASE FREEWAYS &-8 LA1"iE FREEWAYS

i .-\..it
0.0- o.~
1°.8-
1•6
I ;:: \·6
10'0 - 0.4-·
1°·4..,.0'8
:0'8- \.2
;\'2 - ;;..",
I ;::..z .l.f
10- 0''10
i o.!t- o·g
1°'&_ 1'6
11•6- .(.q.
i ;::.? .l+
io -c''+-
!o·Lf-o./?
10'8- l·b
i \·b ;::
Ie - ·0'4-
!o.'t- o·£'
o.<g-I.b
/ ;:: 1·£
0 - 0·4- i
0.Lt- o·~ i
1
,0''8 - \.;(, i
, ;:: \.2, i

:--lOTE; [f ~ Icng'th o{ gr::a.dc falls on :l Oounaary ..:onoicion. ~hc equiva.icL'\t for the: !on,;C( grade C:1CcgOry 1$ u.s.ed. FoC'":lay g.n.dc: neeO'er tha.n the po:rc<:nt1~<: shown. use the: next higilc!
;n.dc category.

GRADE PASSENGER·CAR EQUIYALE"'''T. Er

(%) ~!..A..""'E :'REEW A YS &-8 LA.1'o"'EFREEWAYS

0-0 - C''f-
0-.,+ ~O'B'
0'8 _ \.2
\-'2- '·b
> \·b
o~ 0 - O·Lt

0,4- o.g
0,8-1· 6
> I·b
6-0-0,4-
O',~-I'"
> l'b
o . 0- colt
0''1--1·6
> l·b

Sou •...
ce: -rR 8 (J 985)
GRADE PASSEt'GER·CAK EQU!VALE:-''T. ET
(%) ~LA?'."E FREEWAYS 6-8 LANE FREEWAYS

<! AI!
0.0-0',+
0·4- 0·8
0:8 - \'2-
\.:2. - \'0
H'-~'4
>.Q.q.
0:0 - 0·L4
o·lf- Q.g
OJl?_ 1·2-
\.2. - I- b
\. b -.2 .'t
>..:<,.4
0'0
0'4 - 0'8
- 0'41'
0'8 -
\·2 -
> l'b
1,' I
\.2. I

0'0 - 0·4 I
t>'4 -
0,8-
0·8
\.2-
I
\,:2. - ',6 I
>\·6 I
0·0 - 0·4
c.'t - 0·6
0·8 - "2
> \'"2-

~o- 0·4- i 9
O.'t-o'~1 17
> 0.8 28

LE.~GT:'"!
PASSENGER·CAR EQUIVALEo.•.•
'T. E.
GRADE
(%) (I::M) ~-LANE ~EWA YS 6-8 LANE FRE:EWA YS

Ali
0'0 -0.8
:::.
o·B
0'0 -0·4
0·4 -1.2-
;:: \.:1-

=- ~:11
~:~;::'\'2 ,

0'0-
o. 't-
0't.241
:::.\.2.
 Driver Population Factor

Commuter, or other regular user 1.000

Recreational, or other non-regular user 0.75 - 0.90·

• Engineering judgement and/or local data should be used

in selecting an appropriate value.

So urc ~ >TR- B ·~("t6 5)

TA.BLsZ-10. VALUES OF VOLlJME-To-CAPACITY RATIO FOR USE IN DESIGN

MSF'
(PCPRPL) DENSITY SPEED
(PC/I<"/LN) (KHM

\ \0 KM!tl ELEMEl'ITS

0.30 60J A 6· 5- 97
0.35° 700 A :;Z.S" en
0.';0 800 B 8'7 '15
0.50 1.0c0 B \0-9
0.54c 1.100 B \2·,+ 'l3
9z.
0.60 1.200 C
13·0
0.70 1,.;00 C
15' 5
10
0.77· l,550 C
16·6 ~9
87
0.80 1,6OJ D /'1.0 8l+
\ 00 KM/H ELEME:-'-rs

$Lf-
0.30 00:J B 71·5
O.c.o 800 B q·b 8'+
0,490 1.0c0 B \2-<.+ 80
0.60 1,200 C
[..;00 15'5 77
0.69° C 18· b' 76
0.80 1,600 D 2'3 ·3 b~
60l<M/H ELE~IE~-rs

0.30 550 C 8-1 76

0.40 750 C , o·b 7{,
0.50 950 C \3-7
0.60 C
7Z'
t.150 \ b·8
l.300 7/
0.67C C \8·f, 6'3
0.70 USO D ~\·I 66
0.80 1,500 D 2.6'" {,If
V:!lues round~d to the neares~ 50 pcphpL
::l
= Design may be v..i[hin LOS bounds. >lot necessarily a~ maximum condition for LOS. SOVl"'Ce: IRB ( '~85)
: ~a}.imum permissible value for the LOS shoQlll.

PE:l.CE:-'"T ,RUCKS
LEVEL OF
SERVICE a to 15 20

A 700 650 650 60J 600

B !.l00 1.050 LOCO 950 950
C [,550 1,500 1,~50 USO l.300
D !.S50 l,S00 1.700 1.6OJ !.SSO
E 2.0c0 1.900 !.S50 t.750 LiCO
.
..." 700 600 550 500 ~50
B 1.100 950 S50 750 700
C U50 l.350 I.lOO l.050 i.OCO
D i,350 l,6OJ l..;.()() 1.300 U50
- 2.000 1.750 U50 l,~ [,250

-'A 700 500 ~OO 350 250

B i.l00 80J 650 550 .;.()()

C l,550 U50 900 750 550

D 1.850 !,350 l,lOO 900 650
E 2.0c0 UOJ l.200 l,OCO 7CO
3.ue 3.SSum!'uol"'oS :QC' T3btc

ph design sOoeed
AU neavy ve!ucic::s are trucb
L.:ln,= •••.Idths :lrC' 3" 7 .,...
~ou •...
ce TRB (fq85)
L:Lte:a! C~C:3r3.nce:s > , " as ~
Weaving is the crossing of two or more traffic streams travelling in the same direction along a
significant length of road without the aid of traffic control devices, such as when an entrance
ramp tenninal is closely followed by an exit ramp terminal. The area over which weaving
occurs is called the weaving area.

Weaving areas may be simple or multiple as shown in Figure 2.9. A simple weaving area is
formed by a single merge point followed by a single diverge.

2 ---
3
~--# ::::::::> •••
•.

~-~-~
~==================;;>-------_=..7
5 •.
------------------------

In the simple weaving area, flows I and 2 weave, whereas in the multiple weaving area flow 3
weaves with flows 1 and 2 and flow 4 weaves with flow 2. These are termed weaving flows.
Flows that need not cross other flows are termed non-weaving flows. For instance flow 3 in
the simple weaving area and flow 5 in the multiple weaving area are non-weaving flows.

(ii) Configuration - which is related to the relative placement and number of entry and exit
lanes for the section. The three types of configuration, namely A, B and C, are
illustrated in Figure 2.10. With the type A configuration each weaving vehicle makes
one lane change to perform the desired movement, while the type B and C
configurations require more complex manoeuvres.

(iii) Weaving width - which is the width of the weaving area measured in terms of the
number of traffic lanes in the section.

(iv) Type of operation - in general, vehicles in a weaving area will make use of all the
available lanes in a way such that all component flows take place at approximately the
same average running speed, but with the weaving flows slightly slower than non-
weaving flows. However, sometimes the configuration of the weaving area limits the
ability of weaving vehicles to occupy the necessary proportion of the available lanes to
achieve this balanced operation. When this occurs, the operation of the weaving area is
said to be constrained, and in this situation non-weaving vehicles will operate at a
significantly higher speed than weaving vehicles. When configuration does not restrain
weaving vehicles from occupying a balanced proportion of the available lanes, the
operation is termed unconstrained.
 _~===- -:xz -~
-----------

~=====~
~ CONFIGURATION TYPE B ~

--------~
Z = =:-

Evaluation of the level of service in an existing or projected weaving area is

accomplished using the following computational steps :-

(a) Establish Roadway and Traffic Conditions (Table 2.12)

(b) Convert all traffic volumes to peak flow rates under ideal conditions - all
component flows must be converted to :-

v - ---------
v
- PHF X fHV x fw x fp

v= flow rate for peak 15 minutes, in pcph under ideal conditions;

V= hourly volume, in vph, under prevailing conditions;

fHY, fw the adjustment factors for heavy vehicles, lane width and lateral
and fp == clearance and driver population, obtained from Tables 2.2, 2.3, 2.4,
2.5,2.6,2.7,2.8 and 2.9 in Basic Freeway Segment Section;

C= Construct Weaving Diagram (Figure 2.11);

d = compute unconstrained weaving and non-weaving speeds (Table 2.14);

e= check for constrained operation (Table 2.15);

f = check weaving area limitations (Table 2.16); and

g = determine the level of service (Table 2.17).

 The analysis of multiple weaving areas involves the construction of appropriate
weaving diagrams for each subsegment of the area using Figures 2.12 and 2.13. Once
these diagrams are established, each sub segment may be analysed as a simple weaving
area, according to the procedures listed above.

Figure 2.12 Weaving flows in a multiple weave formed by a single merge followed by two
diverges

Flow ~
A-'I..
~- ~
Pt-'{
g-'(
c.-x.
C. - Y

Figure 2.13 Weaving flows in a multiple weave formed by two merge points followed by a
single diverges
2.8 The weaving area illustrated in Figure 2.2 serves the following
traffic volumes :-

A-C 1815 vph

A-D 692 vvh
B-C 1037 vph
B-D 1297 vvh

Traffic volumes include 7 percent trucks, and the PHF is 0.91. The
section is located in generally level terrain, and lane width are
3.7m. There are no lateral obstructions. The driver population is
composed primarily of commuters. At what LOS will the section
operate?

C 0 /"A
"'::-
L=4bO~

>/ -=-~-==--~-==~./<
...--- - - - - -- -- -- -- -- - --...,
"-

1) ~ /---------- ~" 8

2.9 The weaving section shown in Figure 2.3 serves the traffic flows
indicated. Lane widths are 3.7m and the section is located in level
terrain. There are no lateral obstructions. For convenience, all traffic
flows are given in terms of peak flow rates for ideal conditions,
expressed in passenger cars per hour. At what LOS will the section
operate? 4000

~ 1- - ---'---- - - ----~

~ ~
pi 'tJr~ :z. 3
2.10 The ramp-weave section shown in Figure 2.4 serves the following
demand volumes ."-

A-C 975 vph

A-D 650 vph
B-C 520 vph
B-D o vph

Traffic includes 10 percent trucks, is composed of daily

commuters, and the PHF is 0.85. 3.7m lanes are provided with no
lateral obstructions and the section is located in generally rolling
terrain. What is the expected LOS for the section?

~--------------~
J)~ ~B

2.11 Figure 2.5 shows a multiple weaving area. Peak flow rates In
passenger cars per hour for the section, are ."-

(a) A-X 900

(b) B-X 400
(d) A-Y 1000
(e) B-Y 200
() C-X 300
() C-Y 100

All geometric conditions are ideal, and the terrain is generally

level. At what level of service would the section operate?

X ,,~ L2. = 4bo",,", L, = >'00 ""

>-" ~-==- __ -=-_-= ~~_-f\

y//~ 0': ~ c B
 fl'i""","~

TABLE.). .12 CO~FJGTJRATIO:-; TypE. VS.ONUMBE.ROF REQUIRED LAKE

CHANGES

. NUMBER OF ltEQ'D
LANE CHANGES FOR
WE.••VING MVT. a
m'''!'UMBER

0
:><
OF REQ'D LANE CHAJ'GES FOR
WEAVING MVT. b
I > 2
- ..
0 Type B Type B Type C
1 Type B Type A -
~ 2 Type C - -

L . Lengdl oi weaving area, in m.

LH ••••••••••••••••••••••••••••••••••• Length of weaving area, in hundreds oi m.
N . Tocal number.of lanes in che weaving are:!..
N•..•................................ ~umber of lanes used by weaving vehicles in ch~' weaving ue3..
;Y•••.••.•. , .•..••...•...•............ ~umber of lanes used by non weaving vehicles in ehe weaving area.
Tot:li flow rate in the weaving area, in pa=nger eu equivalents. in pcpe:.
Total weaving Dow rate in the weaving z-:C3." in passeoger car equivalents, in pepa.
Weaving Dow rate for the larger of th<: two ""<:aving ·110ws. in pass<:nger car equivol<:ntii. in ?eph.
We:.wing flow rate for the smaller of che cwo ""caving !lows. in passenger C:2.requiv:?iencs. i::. pcph.
Toc.al nonwe:!.ving Dow race in che ""eaving are:!.. in passe:"!ger C:J.: equiV2.iencs. in pcpr..
VR . Volume ratio vj'l.
R . \V ea vin; ratio v _1/ v_
S•................................... ";'ver2.ge running slXed of ""caving v<:i-.icle:sL.., the we2.ving area. in I<m / h
.-\ve:'ag~ running speed of r:.onwe2ving vehicles in ~he wC2.ving :lre:a. inKm/ h

CONST..•..
:-.-rS FOR CO·~STA:-.-rS FOR
WE..-\VING :-;Ot'WEA VlNG
TY?~ OF S?::':ED, S. S?::':ED. S••
a (, c
COKFIGURA"TlON a '0 c d

T~£A.
Uncons:r"in<:d 0.226. . :.:2 1.00 0.90 0.020 ~.O l.30 \.00
Constrained I 0.280 ', i.oo 0.90 0.020 ~.O O.SS 0.60

l":"PE B
Unconstrained
I 0.100 l.2 0.77 0.50 0.020 2.0 L42 0.95
Constr:!.in<:d I
I
0.160 ~.: 0.77 0.50 0.015 :.0 !.:;O 0.90

Ty?£ C
Unconstrained
..I, 0.100 LS O.SO 0.50 O.DlS l.S t.l0 0.50
Constrained O.too ~.O e.ss 0.50 O.Olj l.a UXl 0.50
 SO. OF LASES REQ'D FOR tJSCONSTIU..J:-.'EO :.lAX. :-.'0. OF
-rYPE OF
CONFIGURA TIOS OPERATIOI'<. Nw WEA Vl:'1G LANES • .v_ (max)
Typ~ A I 3.575 N(VR)o'S1ItLy)o'.23Lf /U'w)o'438
. r. }
Typ~ B 11"..1\0.0 8S + 0'703 lV~) + \7\.57(1-) -0.0,12 (Snw - So",,)

Typ~ C IN {o.7bJ - o·'3{,o'l LH - o· 00311 (:S~W - ~w) + o.",If? C.VI<) }

~ All v3nabks 3rt as defmed in Ta.ole .
0' For 2.slded wt.aving .1re3~.all free .....ay L1ne$ m."y DC us~d :lS WC:l.vtng l.1.ncs.
~OTE: When .v~~ ':.J' .• (ma.x.). o~r':Hlon 15 unconstr:unC'd. ..
\Vhcn -N. > .'( (max). OpCr;JClon IS constr;uned.

:.lAXI~{U:.l ~{AXIMUM MAXIMU~1

TYPE OF WEA VING CAl'ACrrr'. VOL. ;(ATIO. WEAVING WEAVlNG
CONFlGURA TION ~{AXIMUM '_ o VR .- llAnO, K" LENG..-Y..L
1.900 VR
pcphpl l.00
OA5
0.35
0.22
'1.900
pcphpl

1.900
ocphpl

Source; iR.B (/98;5')

J #;00 Pi
Level of Minimum Average Speed km/h c · ~---
Service Weaving Non-Weaving ~~~
1)~-----~B
88 96
80 86
n 77
M 67
48/56 48/56
< 48/56 < 48/56

•.•.'"? _ VOLv:--":: ?..,;:::::O ~ •.• /7 (800/2iOO _~ C.1Sti)

~ _ "~n~r~~ ~;7:Q - ~.2/V~ {JOo/aoe • 0.375)

(a) The existing geometries and traffic volumes are stated in the
description. Note that the section is Type B configuration (see
Table 2.12). Weaving movement B-C may be made without a lane
change, while movement A-D can be made with single lane
change.

(b) All volumes must be converted to peak flow rates under ideal
conditions, expressed in passenger cars per hour.

v-·
v
- PHF x fHV x fw x fp

PHF = 0.91 (given);

ET = 1.7 (Table 2.3)
1
fHV = 0.95 (computed as [1 + 0.07 (1.7 - 1)]
fw = 1.00 (Table 2.2); and
fp = 1.00 (Table 2.9).

1.815
A-C = (0.91 x 0.95 x 1.00 x 1.00) = 2 100 pcph
692
A-D = (0.91 x 0.95 x 1.00 x 1.00) = 800 pcph
1 037
B-C = (0. 9 1 x 0.95 x 1. 00 x 1. 00) = 1 200 pcph
1297
B-D= [0.91 x 0.95 x 1.00 x 1.00) = 1 500 pcph

(c) A weaving diagram for the calculation is now constructed, using

the converted flow rates of step (b)
(a) The existing geometries and traffic volumes are stated in the
description. Note that the section is Type B configuration (see
Table 2.12). Weaving movement B-C may be made without a lane
change, while movement A-D can be made with single lane
change.

(b) All volumes must be converted to peak flow rates under ideal
conditions, expressed in passenger cars per hour.

v- .
v
- PHF x fHV x fw x fp

PHF = 0.91 (given);

ET = 1.7 (Table 2.3)
1
fHV = 0.95 (computed as [1 + 0.07 (1.7 - 1)]
fw = 1.00 (Table 2.2); and
fp = 1.00 (Table 2.9).

1.815
A-C = (0. 9 1 x o. 9 5
xl. 00 xl. 00) = 2 100 pcph
692
A-D = (0.91 x 0.95 x 1.00 x 1.00) = 800 pcph
1 037
B-C = (0. 9 1 x 0.95 x 1. 0 0 x 1. 00) = 1 200 pcph
1297
B-D= [0.91 x 0.95 x 1.00 x 1.00) = 1 500 pcph

(c) A weaving diagram for the calculation IS now constructed, usmg

the converted flow rates of step (b)
Critical ratios may also be computed for use in analysis (refer to Table
2.13 and Figure 2.11) :-

Vw (total weaving flow rate in the weaving area) = 1200 + 800 = 2 000 pcph
v (total flow rate in the weaving area) = 2 000 + 2 100 + 1 500 = 5 600 pcph

. . Vw2 800
R (weavmg ratIO = V w )= 2 000 = 00400
. Vw 2 000
VR (volume ratIO = y-) = 5 600 = 0.357

(d) The unconstrained speeds of weaving and nonweaving vehicles in

the section may be estimated by using the equations for Type B
configurations from Table 2.14.

80.5
S w or Sn w = 24.15 + 1 + a (1 + V R) b (V / N ) c / (3.28 L ) d

80.5
Sw = 24.15 + 1 + O. 10 (l + V R) 1. 2 (V / N) 0.77 / (3.28 L ) 0 . 5

80.5
= 24.15 + 1 + O. 10 (l + O. 3 5 7 ) 1. 2 (5 600/4 ) 0 . 77 / (3. 28 x 460) 0 . 5

S - 24 15 + 80.5
nw -. 1 + 0.02 (1 + 0.357)2.0 (5600/4)1.42/ (3.28 x 460)0.95

(e) Using the unconstrained estimates of weaving and nonweaving

speeds, the number of weaving lanes required to achieve such
operation is computed and compared with the maximum value of
3.5 lanes for a Type B configuration. The equation is obtained
from Table 2.15.

71.57
Nw = N [0.085 + 0.703 (VR) + ( L ) - 0.0112 (Snw - Sw)]

71.57
= 4 [0.085 + 0.703 (0.357) + (460 ) - 0.0112 (63.75 - 64.76)]
Therefore, the section will operate in the unconstrained mode, and the
speeds computed in step (d) are the final solution.

(f) All values for the calculation are below the limits established in
Table 2.16, and the operation is expected to be as indicated in the
computations in previous steps.

(g) Comparing to the criteria of Table 2.17 shows that the level of
service for weaving vehicles is D, while the level of service for
nonweaving vehicles is E.

A multiple weaving section is analysed as two separate simple weaving

areas. The initial step in the analysis is to construct weaving diagrams
for the two subsegments of the multiple weaving area. Because all
demands are stated in peak flow rates under ideal conditions and no
conversion computations are required, this is done immediately. The
weaving area under study is of the type illustrated in Figure 2.13, as
follows :-

400 300
R= 1400 = 0.286 R = 1500 = 0.200

1400 1500
VR =2500 = 0.560 VR =2900 = 0.517

Note that both segments of the weaving area are Type B configurations.
In segment 1, movement A-X may be made with no lane changes while
movement B-X requires one lane change. In segment 2, movements A-
Y and B-Y may be made with no lane changes, while movement C-X
requires a single lane change.
Computations for speed are now done for each segment. Note that the
first three steps of the procedure have been completed in the
establishment of weaving diagrams for the two segments.

Segment 1
(a) Unconstrained speed equations for Type B configurations are
selected from Table 2.14:-

S - 24 15 + 80.5
w -. 1 + 0.10 (l + 0.56)1.2 (2500/3)0.77 / (3.28 x 300)0.5

S
nw
= 24 . 15 + 1 + 0.02 (l + 0.56)2.0
80.5
(2 500/ 3)1.42 / (3.28 x 300)0.95

(b) The number of lanes required by weaving vehicles for

unconstrained operation is computed using the appropriate
equation from Table 2.15, and compared to the maximum value of
3.5 lanes, also obtained from Table 2.15, for Type B
configurations:-

(c) None of the limitations of Table 2.16 is violated. From Table 2.17,
the nonweaving Los is E, and the weaving Los is D.

Segment 2
(a) Using the same equations as for segment 1, because both are Type
B configurations :-
S = 24 15 + 80.5
w . 1 + 0.10 (l + 0.517)1.2 (2 900/3)0.77 / (3.28 x 460)0.5
 S = 24 15 + 80.5 '
nw . 1 + 0.02 (l + 0.517)2.0 (2 900 I 3)1.42 I (3.28 x 460)0.95

71.57
Nw = 3 {0.085 + 0.703 (0.517) + (460) - 0.0112 (69.8 - 67.8)}

(c) None of the limitations of Table 2.14 is violated. From Table 2.17,
the level of service is D for both nonweaving and weaving
vehicles.

The analysis indicates that the entire weaving area will operate in
the range of 64 to 69 km/h, a range which stradles the boundaries
between levels-of-service D and E.
 ~\",,\ .•....~

TABLE 2..12 CO:"FIGURATION TYPE VSh,,rUMBER OF REQUIRED LAI'E

CHANGES

. NUMBER OF REQ'D
LANE CHAI'GES FOR
.
:x
mll".I'UMBER OF REQ'D LANE CHA1'GES FOR
WEA VING MVT. b
WEAV1NG MVT. a 0 o·
1 ?; 2

0 - Type B Type B Type: C

1 Type B Type A -
?; 2 Type C - -

Leng.ch of weaving area, in rn.

LH• •• ••••••••• ••••••••••••••••••••••• Lengch of weaving area. in hundreds of m.

N . Total number .of lanes in the weaving ar~.
N~ . Number of lanes used by weaving vehicles in the weaving area.
N.w •••••••••• • ••••••••••••••••••••••• Number of lanes used by non weaving vehicles in the weaving area.
Toc3.1 flow rate in che weaving area, in p=nger car equivalents. in pcph.
Tocal weaving !low rate in che weaving are2. in p"assenger car equivalencs. in pcph.
\Veaving !low race for che larger of the cwo weaving ',lows. in passenger car equivalencs. in pcph.
We:?oving now race for the smaller of the two weaving clows. in passenger car equivalents. i:1 peph.

Toc.a! nonw:::l.ving flow r:?otein the ""eaving ar~. in passenger car equivalerm. in pcpC-.
VR . Volume racio v.lv.
R .......................•............ \Veaving racio v.,/v_
5•................................... Aver2.ge running s?<,ed of "'eaving vehicles in che weaving area. in J<m / h
Average running ~peed of non weaving ve!licles in the weaving area. in Km/ s-.

CONSTANTS FOR CO",·STA:-.-rSFOR

WEAV1NG :-;ONWEAVlNG
TYl'E OF S?EED.S_ S?EED.S ••
CONFIGURATION a b c d :?o b c d

TYPE A
Unconstrained 0.226. .·:!.2 1.00 0.9.0 0.020 ~.o l.30 1.00
Conscrained 0.280 :.2 l.00 0.90 0.020 ~.o 0.88 0.60

TYPE B
0.100 1.2 o:n 0.50 0.020 2.0 l.-+2 0.95
U:Jconscrained
Constrained 0.160 1.2 0.77 0.50 O.Q1S 2.0 UO 0.90
I
Ty?:. C
.. 1 !.S O.Q1S I.S t.10 0.50
U nconscrained 0.100 0.80 0.50
Constrained I 0.100 2.0 (JoS5 0.50 0.013 l.6 l.00 0.50

Source: iRE (I'lf3 5)

 SO. OF LA;-,"CS REQ'D FOR UNCONSTRAI"EO :.lAX.;-iO.Of
TYPE OF
CONFIGURATION OPERATION, Nw WEAVING LANES, N_ (max)
1,4

Type B

Type C
J, All Y30ablcs :ire as dcttncd in Table:
b For !.sided weaving areas. ell freeway I~n<s may "0<:used as weavIng lan<'S.
:-iOTE: When ,yo s; :-1 (m.x). op<:raCion IS unconsualOed.
0
-
When :Y. > ,yo (m.x), operation is constrOlned.

~AXl:'iU~ ~!AXI~U~! ~,fAXI~U).l

TYPE OF WEAVlt'G CAPACITY, VOL. RATIO. WEAVI;-iG WEAVlNG
CONFIGURATION ~fAXIMUM '0 - VR ,-- RA no, R--- LENGn!, L
1,900
~phpl
3 0.45
4 0.35
5 0.22

-1,900
pcphpl

1.900
pcphpl

=_========13:S;;:Coo:==========:::AA
Level of
Service
Minimum Average Speed kmlh
Weaving Non-Weaving
C
~~---~-~ 1

1)~----~B
88 96
80 86
72 77
64- 67
48/56 48/56
< 48/56 <48/56

_ ~o"~~ .EAV!~G ,LO" (500 ~ 300 - seCI

70Th~ FLC" !N 7~~ "ZAV!NG A?~A (500+300~1500+~OO 1700)

.,'it _ VOW!':!: ?.ATIO ~ ~. ./7 (aOO/2iOO ," 0.296)

a • "£nV!~G ~~~!O • v~2/v~ (300/200 • 0.j75)

A ramp may be described as a length of roadway providing an exclusive connection between
two highway facilities.

The capacity of a ramp is the least of those of the following that are relevant in any particular
case, namely:-

(i) the capacity of the ramp proper;

(ii) the capacity of the cross-road terminal;
(iii) the capacity of the freeway entrance; and
(iv) the capacity of the freeway exit.

Under ideal conditions, a single-lane ramp roadway may accommodate up to 1700 passenger
cars per hour, although in general when the ramp volume exceeds 1500 passenger cars per
hour a two-lane ramp and a two-lane freeway ramp terminal should be provided. Approximate
service flow rates for single-lane ramp roadways are set out in Table 2.20.

In those cases where a single-lane ramp and ramp terminal are adequate from a capacity point
of view, a two-lane ramp may be provided if :-

(i) the ramp is longer than 300m, in order to provide opportunity to overtake stalled or
slow-moving vehicles;

(ii) additional storage is required on the ramp to accommodate queues in advance of a

controlled cross-road terminal; and

(iii) the ramp is located on a steep grade or has minimal geometries.

If a two-lane ramp is provided for any of these reasons, it is usually tapered to a single lane at
the freeway terminal.

Ramp cross-road terminals should be treated as normal at-grade intersections, and their
capacity analysed accordingly.

At freeway entrance and exit terminals, the merging and diverging manoeuvres, which occur in
the freeway lane adjacent to the ramp, should take place at the speed of the freeway traffic
stream and, without interference with it. The principal factor governing the capacity of freeway
terminals is the volume and nature of traffic in the adjacent freeway lane immediately upstream
of the entrance or exit ramp. This volume depends on factors such as ;-

(i) the type of ramp and the volume of traffic using it;
(ii) the proximity of nearby entrance and exit ramps and the volume of traffic using them;
(iii) the total volume of traffic on the through carriageway upstream of the ramp terminal;
and
(iv) the total number of through lanes.
A step-by step computational procedure for the analysis of ramp terminals is given as follows :-

(a) establish ramp geometry and volumes;

(b) compute lane 1 volume - lane 1 volume is computed using either one of 13 equations
included in Table 2.19 or the approximation procedure described by Table 2.21 and
only where equations are not available for the particular configuration being considered;

(c) convert all volumes to passenger cars per hour - all lane 1 volumes, ramp volumes and
freeway volumes must be converted to equivalent volume in passenger cars per hour.
Before converting lane 1 volume to pcph, it is necessary to determine truck presence in
this lane. Figure 2.15 or local data are used to estimate the percentage of total freeway
trucks in lane 1;

(d) compute checkpoint volumes - for each ramp analysis, there are up to three checkpoint
volumes for each ramp or pair of ramps :-

(i) Merge volume,Vrn

(ii) Diverge volume, Vd
(iii) Total freeway volume, Vf

Figure 2.16 illustrates the computation of checkpoint volumes for the case of an on-
ramp followed by an off-ramp;

(e) Convert checkpoint volume to peak flow rates (Table 2.18); and

(f) Find relevant levels of service - the level of service for a given analysis is found by
comparing the checkpoint flow rates for merging, diverging, and total freeway volume
with the criteria given in Table 2.18.
 2.12 Consider the following on-ramp, which has no adjacent ramps
within 1830m, and may be considered to operate in an isolated
manner :-

LEvel l"GRRAltJ
? H F =oc::\ 0 v Ph
i.v-vc.J<s
1\0 Kn--JL-r 1)~SI~h S?«.J

2.13 Consider the following ramp configuration. There are no other

ramps within the influence area of the ramps shown.

PHF = o' 95
1><:>I')h s~~J =\\0 t""lc-

7ijRAt-AP2 IY/RAMPI
500 vpk, '300 vph
5 0/0 -truc.kS 5 IJ/o 'truc.l< S

At what level of service would the two off-ramps be expected to

operate?
 2.14 Consider the following configuration. No other ramps influence
the behaviour of those shown.

..\

-=----=- -=== ~=-=~.

L~vGL T~Re-A\ r.J I
'5500 vpl...
\\0 ~-"}k De.Si,~SPe..c{==--=- ..-.- \ 0 ~I ~ i:.rlJt-ks
1'~q:::::= c-qo _

/Y/ ~"i"o

5
VPk
10 L,-..,e:.-l::.s
boo vpk 0

to "1~ ty-v~

2.15 The following off-ramp occurs on a five-lane urban freeway

segment. It is not within the operational influence of any adjacent
ramps :-

7200 VP!."
10 °/ •• Trvvo--
D~'>1'7'" ~ruJ= 110 lC""fL--
r(bl\(",.J \e«o..,:"'"
 ~. 1 6 Consider the left-side on-ramp shown below, which is far enough
away from other ramps to be considered as isolated :-

_____ ~;l.SO pc Ph
'PHF- =- c' '10
LtivG~ TEgRA\T·J
J>e.-s \.'7 "" ~ fc- 01 ::; 110 \C.W\ f \...

U sing the index provided in Table 2.19, it is seen that Equation 1 (and
Figure 1) are chosen as the appropriate equation for this case. Thus, the
lane 1 volume immediately upstream of the on-ramp is computed as :-

vf =
2500 vph;
Vr = 550 vph;

VI = 136 + 0.345 (2500) - 0.115 (550) = 935 vph

From Figure 2.14, about 67% of all trucks on the freeway will be in lane
1 immediately upstream of the ramp. Therefore

Total trucks on freeway =" 2500 x 0.10 = 250 trucks

Trucks in lane 1 = 250 x 0.67 = 168 trucks
168
Proportion of trucks in lane 1, volume = 935 = 0.18 or 18%
At this point, the lane 1 ramp and freeway volumes must be converted
to passenger cars per hour. Values of ET are selected from Table 2.3
and values of fHv are computed as :-

1
[1 + PT (ET -1)]

Item Volume ET Proportion fHV vol. (pcph) =

(vph) of trucks vol. (vph)/fH V
VI 935 1.7 0.18 0.89 1,051
Vr 550 1.7 0.05 0.97 567
Vf 2500 1.7 0.10 0.93 2,688

vm = Vr + VI = 567 + 1051 = 1618 pcph

Vf (after merge) = Vf (before merge) + Vr
= 2688 + 567 =
3255 pcph

These values are now expanded to peak flow rates by dividing by the
peak hour factor. The level of service is then found by comparing the
merge and freeway checkpoint flow rates to the criteria of Table 2.18.

1618
Vm = 0.90 = 1798 pcph (LOS E, Table 2.18)

3255
Vf = 0.90 = 3617 pcph (LOS D, Table 2.18)

In this case, the merge area is the controlling feature (an undesirable
condition), and the prevailing LOS is E.

As indicated in Table 2.19, note 2 must be consulted when analysing the

first ramp. Note 2 specifies the use of Figure 7 for this ramp, but
instructs that Vr be taken as equal to the total off-ramp volume on both
ramps. Figure 7 is also used for the second ramp.
Because there is no upstream on-ramp involved, the value "65.5 ~:" =
2 is used, as directed by item 2 under "Conditions for use on Figure 7"
(Table 2.19). As noted above, Vr will be taken as :-

The proportion of trucks with respect to lane 1 volume IS now

computed.

% total trucks in lane 1 = 56% (Figure 2.14)

Total trucks on freeway = 4500 x 0.05 = 225 trucks
Trucks in lane 1 = 225 x 0.56 = 126 trucks
Proportion of trucks in lane 1 volume = 126/1514 = 0.083, say 8%

% trucks in lane 1 = 53% (Figure 2.14)

Total trucks in lane 1 = 4200 x 0.05 = 210 trucks
Trucks in lane 1 = 210 x 0.53 = 111 trucks
111
Proportion of trucks in lane 1 volume = 1303 = 0.085, say 9%
 Item Volume ET Proportion fHV vol. (pcph) =
(vph) (a) of trucks (b) vol. (vph)/fHV
Vf 4500 4 0.05 0.87 5172
Vr (1) 300 4 0.05 0.87 345
Vr (2) 500 4 0.05 0.87 575
VI (1) 1514 4 0.08 0.81 1869
VI (2) 1303 4 0.09 0.79 1649

1. the freeway volume at the maximum point, before the two off-
ramps;

3. each checkpoint volume must be converted to a peak flow rate

and compared with the criteria of Table 2.18.

Vf = 5172
0.95 = 5444 pcph (LOS D, Table 2.18)

1869
Vd (1) = VI (1) = 0.95 = 1967 pcph (LOS E, Table 2.18)

1649
Vd (2) = VI (2) = 0.95 = 1736 pcph (LOS D, Table 2.18)

In this situation, the diverge at ramp 1 is clearly the critical restrictive

element on operations, and causes the overall LOS to be E.

The high lane 1 volume at this point, however, is greatly influenced by

the presence of a second, more heavily used, off-ramp within 230 m.
The diverge volume at ramp 1 is not really the problem per se, but the
total lane 1 volume at that point is. This would not be an easy situation
to remedy, although consideration to modify the location of the ramps
might be given, particularly if greater separation could be provided.
The addition of a freeway lane in the vicinity of these ramps might be
considered to separate off-ramp vehicles from the through volume in
lane 1. This lane could be dropped at the first or second off-ramp.
 I FREEWAY FLOW RATES (pCPH)e. VI

MERGE FLOW DIVERGE FLOW Ite k'l<1/J.l-DESIGN SPEED 100 KM Ifl DESIGN SPEED 80 !<11 J iI DESIGN SPEED
LEVEL OF RATE (PCPH)' RATE (PCPH)b
SERVICE v_ v, 4-LA:-'c 6-LANE 8-LANE 4-LANE 6-LANE 8-LA."IE I
r
4-LA:-'c 6-LA.'c 8-LA1'.c

A :s: 600 :s: 650 -" 1.400 :s: 2.100 :s: 2,800 d d d d d d

B :s: 1,000 :s: 1,050 :s: 2,200 :s: 3,300 :s: 4,400 :s: 2.000 :s: 3.000 :s: 4,000 d d d

C :s: 1,450 :s: l.500 :s: 3.100 :s: 4.650 :s: 6,200 :s: 2,800 ~ 4.200 :s: 5.c·Y:) :s: 2,600 :s: 3.900 :s: 5,200

D :s: 1,750 :s: 1,800 :s: 3,700 :s: 5.550 :s: 7.400 :s: 3.400 ~ 5,100 :s: 6.800 :s: 3,200 :s: 4,800 :s: 6.400

E :s: 2.000 :s: 2.000 :s: 4,000 :s: 6.000 :s: 8,000 :s: 4.000 :s: 6.000 :s: 8.000 :s: 3,800 :s: 5,700 :s: 7,600

a Lan~.l now rate Pl~ r~p fto~ rate for one· lane. righc.side on-ramps.
b Lane-I :low rate immediately upstream of olf-ramp for one-lane, right-side ramps.
e Total freeway flow rate in one direction upstream of olf-ramp and! or downstream of on-ramp.
d Level of service not attainable due to design speed restrictions.

w
Z
<!
-'
Z
(/)
~
U
:::J
0::
l-
-I
«
l-
a
l-
l.L
\
a
l-
z
W
I
U
0::
I---,-~_-
40 50 60 70
W
Q VOLUME (100 VPH)
 · ~LA."<: fR..EEWAY 6-LAI'.<: FREEWAY 8-LA.NE FREEWAY
(2 LANES EACH D[UCTION) (3 LANES EACH D[RECTION) (4 LA:'<"E.$ EACH DtRECTlON)

Isolated. One Lane

On-Ramo

Isolated. One Lane Approximate I

01T-Ramp Usillg Table~'21
and Fig. ,.It

7
Adjacent One-Lane Approximate Appro.u.;nate
On-Ramps Using Table.l·2.1 Using Tabk
••• and Fig.2.·I\" 2·il and Fig ...N\-

I
I
A.djacent One-Lane Aooroximate I Approximate
Off-Ramps U;~g Table2'~i Using Table
I ~');I and Fig..l.l(

77
On-Ramp ronowed
'od F;g. ,."

Approximate
by Olr-Ramp Using T:>.ole

7
1:Z.1and fig.2·n'

Off-R:!..mp Followed

\-;
by On-Ramp

-Approximate
Using Table
.:7.~I and ::'"ig.2·1('
( )

~'
----- .......••.. I
Dropping a L:me to I
----I
the Off· at OcT-Ramp

~
 ~-:"A;-';'E Fr..£.EWAY 6-LANE FREEWAY S-LANE FREEWA Y
(2 LAKES EACH DIRECTION) (3 LAKES EACH DIRECTION) (<: LAKES EACH D£RECTION)

Yrajor !unctions ~ Assume that lane B curies an amount of traffic equal to the merge cacck;x>intvolume in TableZ· 18 for the assumed
___ .___-1. level of senice. Ramp lane A then curies the remaining ramp tral1ic. Compute lane I volume using figure'
lane freeway), Figure 6
(4.-
(6-lane freeway), or Figure '1 (8-lane freeway), entering with ramp volume = lane .'.
----,-~. .••••••-A"':'"
volume. Find checkpoint levels of senicc. O:Jntinue computations until assumed LOS agrus with results.

-~ --------- ---
Yfajor Diverges

....::-
..- ,,""'----
- "c

NOTES:
t. U:s.e :=is\U'~ ;.. to f.nd V, i."\ ;l~V~ce;;:c" b.e om
•.. r~p. :;'ut cater vi~b ;l r. w!l1ch is <:qu~ to ~"'e tow voll.:C"l.C co Cocil rz.:np:s..~ ;cchnique is vl1id wilerc ;hc dis::.a.cc.cbct""ecn e.lmp5 is lc::s.s
:.b.Ul.2...,~ M.w;,c(<: :hc disUJ:le.:: ~rwea runp-s U. ix:w~45.1.n.d 11.:20".., 'J.S.c: Ta.olcl,-11a.nd Figu::l.cS\.:) J.ppro.timatc the Situllcion. [{ the: du(.U\ce
l>:cw~ (3.G1p$ is .;ra{.« th-.o11.10 ho\ ~ons;1('r

r:1m~ ~o be i50l;lCed and conside:r s.ep.r;lcdy.

!. Us.: fi~\Jr<: 7 ~o find VI in ;ldV2JIcc o( th.e: fme e<l.mp, but enter ~;ta. •. V•.••.bien '1$ equ.3.1 to tbc wtal volume: on both ocT·e.1rtlps. This tc:enn.ique is valid ~..!:l.e:re the: distance: between n.r~01o
:cs.s th.lln::2.t+5J1l1\_Fo( ocher. duane<:. $e<: noce: l.
J. Treat a.s lWO su~ive on-r2.tO!>5 j,(:pan.=od byIU"'; divide ramp .volume equally o.:to,'ccn (Wo rz.mp lu\es.
-I.. Treat a.s two suCC<::S,)i"e: orT-rampS s.<:p.a.n:to:!: byl1lJr& : divide: otT.r.unp volume: eqw.ily beto,'oc:n rwo n.mp Ia.t\cs.

" Si~jl"-l ,,<:.

•.. 0 ••.• rc.,.....p 01"\ 4- Ic.",•. F,..•.•. ....,O'7,wi;l...
or wijh"",T (>..,-"-~(~•..••.
T"o",, Io."e..
up.sr,. .•."'..•...
2, Fe'- use "''"'
I
7.-::: h.•.~ >",co
/'
"" •.•.)"'-'-
-.T
o •...-"."' •.••..
p e.)(;s/s hJ;Th, .•.• 610 •..•....•
·
V /i Ie 3,+-00 VPh
No'"""'....I rC>o~e of U>e.
I •
3' ~"-'-oo
V :: 50 To 14-00 vpl
r
S ,,,,,Ie -10."'<:' "FF -Y"an-p a •.... c:o.
/v;-r/-,. 0 r "",7t.."vT IX dec.. ••.fe r~ '1,.0•...•
o>?fr wJ.,.e.", -{A.ere /j
no ••. JJo..c.e.",7
::z., Foo-' uSe
I/f ;1"<:11...-.
0" _ .,-",,..,.,p NI7'),,:'" q'8o..,...,.
" " ::-400 to l./2.00 vfL.
3. No'''''''''''' / ~c."'y: t!Jf C/S~ f
=~ C:> t'"o 1500 VPl..
Vr
Si..•.
,lc-lo."" oFf!->""'''''P 0 •..• 0.. Lf-("''''e!r •• w'''7,
'wori... 0 •••. w,-rf...o"T Do d€c ••..(-e•..••..
7,-o •..•. I", •.•• e., Hif/.. """
~di(;.'-q.hl ufS/I''''''''''''' 0/'1- ,..",,.,..,p ",;-11.,;' '16
0
h'l.

Nay ..•...
""\ ro."'Je c::> ~ use.: Vp ~ 70 4-.2.00 vpt.,. 't'o

"r = 50 1;.(bOO vf/...

V ':. ? c> 10 goe::> vfk
U
D := uS' ~. 180 m
tA

I' Si"'5k _l ••."e, loop -T'jre oY\-rGl""f """' '"

l+ - I"",e f-y~t 100./0../ WiTk. 0"" w,-rJ..,ocJT "' •.....
",-Cc.e./ €..,.. •.. 7,'"", 1" •.•.«..

~ = boo 7'0 2000 vpL

y = 600 "to /2.00 vpL.a)
I-
o To '00 IIPh (a.)
I· 5,',..,le -Jo>--'t. o>---y""",P "'''"' L+ -I ••. -7
",~ (-.•...•....•.
w 'l'l... 0. o{j o.c.. •. ""T tJP'ST .•.",,- "'" 0 l.,. -Y""l"""'f> WI'-rj'\l1':,

1;),0 -to 610 -. ,

2, NoT"-'-<-toryc..T.. (...S~'(..
Vl.l ,?}oc> •• vpl.,:

\ = Soe. 'Co 3 '0 0> VP L..

V = 100 -t~ 1')-00 VeJ..
r
V = 100 To 1000 vpl...
l.L
D :: 12
I.A.
(;> 7c 6(0;'"
!'f'V4;,.
NJZ.

6 VI ::-Il.I*o.~'t'-lV(0'03\-"t.l.1-1'l0 "J/DJ I· S,· •.•,Ie. -I"'''e 0 11 - •• """'r5 0'" 6-''''''e Fyt~"""'7 WITh
01'" ""i11.0" T ups r•. ~....
",",e-J /0'- ..l.o•.•.•...
sr,..., """
eF~ -,.. •......,..,f'S { N'rt. 0"" W'1J...O<JT o.c.c •. I-t..•.•..l,o., J ••. l'11!. •

2.. 1~ ~-<- ,'" hOUPs'TYe.""",", <:>?F - ya.""'r wiT/'"k gOD",",;

Use ilIA.;:: "50
3, Lf- -tl----(. /j no do...,,,", ~ Ire""", (!>ff - rr,"""'f ..",-(1.,:"
/7'-fc,>?1, _J 1If. .(.5 vI'/" IIse 000 /90::J... =5
I n
lr ••.•...
l.f-. No ••.•.••.•.. 'e ••f- 1.>11:: d
V :: '<+00 10 '2.00 'J~/... j)d=2~,? to 17lh> •••.•
f
Vo.::: 50 10 1/00 VPh b = 2;SO To 800..,.,
V
d
=
5 C> "To 1'300 vI'/... u..
Vr:: 100 "To liDCI 1./1' h. '
:
\. "5i"'~le ~ oFP-ro..>"l"I.p c>'" b -I"-,,,~ r-.,..•..• "" •.• j
-l ••.•..• !
With. 0"'" WirhouTupST •.e.",,,,,,, t:>\.~-"r.,.....pW;1t-o'-;
lJJii}.o~T de.c.-e.(e. •..••..7,o •.•. tAY\-e '

2.. Jr -rI--o'V~ IS no VPSI"'U'"""" OYl.- •....•• ,...,..,p N,'T/";"

174-0 •..•..•, use t.!:·r"Vu.!r:>

I..
=2-
3· Nl>Y.....-.A! Yc.""'~e err- (;Se: V -= 1100 To 6200 vp L,
V
l=-
= 2.0 "To l goo vp/.....
.; :: oS (> '/0 12.0 •• Vr't..
v..
1)
IA
= "2..80 •.....•To 11'-+°,"""
I. S,'''"'.,(e. -I e- ••. ~ 0"'" -.,.."..,...p 0"'- b-I"' •..e f..,.•. -<: ••••••.• r
w;if..... o...J.J- "..c.••.•••.
T ups 7y to •••.""" 01'\ - Y""o.""'f' I ~r1L.-
ov" w'Tl-.ouT o.c-ce./e.Y' ••. r,o,,", i~Y\e.s, I
2.. NoY"",,",,,"\ r"'''''je. "t- vS"e; Vr- .:::0 1'5'00 Tt> ~4-oC\lPh. i
V ::: \ 00 to 1'500 vp 1.- i
Y t.I.
:= ItlC (:b 1400 \If h II
't\.. ::: /50 TO S 00 J:>-1 i
I, S ••...
jle 10..., e. 01-'\ -y-o.,..,....p 0"," ? -/."'
•... 71
~ ~•.~•......
"'-I,'Th 0'" c..c..~e-I~•...•.i,·o •.• I•..k.-e..
V-Ji1/...ouT I
2,. NoT roV" lire If 14v~ 11 ~Jc--Le.J
dow'"'ST""'U>.. •......•.Opr;-ya,,,,,,,p wi/hI':" q{j-,.,.., I
3· Non· •..."'/ r~"'t': of- u~e: v~:: S"'C>O To/700\l1'L.. I
Y'l' = 300 il> 13 0'" \lPL.. !
I, "::>,'",,l-e -lo\l.,\e oV\ -rtL>?'\p "Y\ &-l~>'\.~ F..,..•. e~A71
Ilk ~c.Ll~"'"" "1'·0.... &-..tj "'Le~
\.U
do.., •...•.
ST.,..-e.Cl'o>'VI
to\lo'I.e.
1
c:>fF - .•..••.>')tp wITI-..\;" 41:>0 -Co 9fr
~I'TL.
M'I
2.. N Ov,.,....~I \ ,',.,...il<. o~ £Xc: y = $000 To 7100 v,:'J... I
f .
'1/
r
=? 00 i:e> II 0 0 Ii P l... I
.
~:= 100 te 800 vfk
D :: 4-"0 +0 'fIr M
d
VI == S t.f -r o· 070 oJ J- -+ 0 • 04 "l VI- I· l'-<JO -I o,l"\e. 0"",, yo. •..•..•• ps 01"\. b -1a.l'I.'t. r-')-HW',,"I'
v..riTh. c>-C.~-e.,..••..T"o •..... 1C'.l'\e oF- ,.,. II!.D.tT~4t{-H-">
V :;. -20\ +0 ..287 V +0- S7rV t.•.•..,7i.. •
It,., f r
I J.-

No",,,,,,,, I """"5<: ot use: V ::: ~= -to '30 00 vph

f
" :: \l 00 'to "3000 vpJ..,.
r
-r-~'
C It",,,,
et D-7t"""
tJ£.
V
Ct>",dih"'" ; Fo"- 05e

12- 0.) V ::: - I r8 -+ 0 '03c \I + O' fG>7 vr \. "Two -I o.",,~ off!- -l'""C<.1'YI P O"'l 6 -letO"\e.
ITA f win...
~,..t.tll>J"''1 S d e.c.e-/~ ....,..1Io"" Ie>. 1"'\ l2.,s
b) VI = 18 + 'If -to· 072.. V r
o' ObO ~ 1-e••.••T 2./ f" """ '''' I B-y..'57J.
2· No""""""",! Y'"ClV'\je of u)-e:
V :: 2100 To '000 vpk
i-
Yr =: \\ 00 To 6000 vpk

13 0.) Vc.. -= b '-f + 0·2- 8 i5' II -j- (>, 1'1\ vr \- p-1o....)o>- d Ive."'je ju"'c:...T;-o •••..s on. 0.- b - r••." e.
F
(.b)
", .::. (73 t o·.2q.r V
. l-
-0'320 V
f
t yct.--"o-7

Tc> -tI.-.rO
I ~i rt-..
.2-IA •..•
e
""fi...l'""~o(. I••••.•. es
r o.-.ol.w#..'j J
c£tv,d""J

2- No •."" •...' r"b.V\)e or- t>5e : V

l-
= 12.00 To 4foovf'l.-

vr =3 00 r;, 2'S-OYfl
I I
vI ;:: vF'sr,...~lJ.~ Lo..l'\e YollJ'r'Y\e C.Vpl..)

YJ- = UPSTy~\o'Y'\ ~~ r,.r()...1 Vclvl'Y\t c.ypJ-..)

Vr == on -,ecvmf VolvW'l e C VPt-.)

" ().. ;;;

Volum €

l)(srp..V'o.c~ -rc:::.
of UPS/orf:"'~

UpSTlo-"t.~"",
OY\ -

0
,Qo-rr'lP c..vPhJ
f2.o..vn.p (yY\)
D= V\ -

u
'Dr5T~'r\c..-e.. l~ r-..... To DowV\sTre't~ crf~-RtJ.~
Ve>luW\. e. <!:>{- AJ-j o..c."Yl,.T UPSTr-eOlYY' -f! - Rt:tJ?'l.p (VPh)

t'vp~r'Je. Volu~e (,L.o,V\.e. 1 + r<o..I"\'I.P Let>'\-e.., A) Y Ph..

 TableZ.2.oApproximate Service Flow Ratesfor Single
Lane Ramps (passenger cars per hou.r)

Level of Ramp Design Speed km/h

Service ~32 33 - 48 49 - 64 65 - 80 > 80
I
A a I

B
a a a 600 I
c
a
a
a
a 1,100
a 600
1,250
900
1,300
I
D a 1,200 1,350 1,550 1,600
E 1,250 1,450 1,600 1,650 1,700
F b b b b b
a. Level of service not anainable due to restricted design I

speed.

TABLE1.2f ApPROXI:v'.ATE PERCENTAGE OF THROUGH TRAF-

FIC'.REMAIl'iiNG IN LANE 1 IN THE VICINITY OF RAMP TER-

MINALS .

THROUGH VOLUME R.=:1-'.A1NING

rNLAi'l""E 1 (%) Percentage of On-Ramp Traiiic Present in Lane 1 at
TOTAL 7HROtiGH
VOl.m1E,OKE S-LA:-'C: O-LA~-:=' Various Distancas Irom Ramp
!)tREcrroN (V"PH) :=:::..EEWAY FREEWAY

?: 6500
6CCO - 0499
5500 - 5999
5CCO - 5~99
~5oo - ~999 Percentage 01 Off-Ramp Traffic Present in Lane i at
:;{XX) - ~99 Various Distances trom Ramp
3500 - 3999
3CCO - 3499 Ii the percentage found in chis figure is less
2500 - 2999 than the pe,cenc of through voiuoe in lane
2CCO - 2499
I from Tabid.· 2/. use the percentage given
1500 - 1999
< 1<\99 for through volume in Table

.1 Throus:b. tr;lmc nae invoive:d tn any ramp within 12"Zomof t:he subject lOCatiOn. Figure ~·IS. Percentage of ramp yehicles ir. lane 1.
. 5ou,,"Ce: IRS C.I<J~S)
lo.J./~A 2-21 Conveys,,,,,, F<>-iAo •••. F:~ _'::::::":'-=:':-':':~-=-"":':=-~ II
eo"$;Je ,..••..7lt>11 o~ R•••.•.•
p t> i"I F"vc _ I
-
I ••.'" e se'wo •.••."7$ (j) ® CD I
fl...FlMP 5-LlQNf Ff?f£WAy teo",rfRflDV
Ty!'£
! I! I
VOl..v",E- ("PH) ~Ilcrote ! I I i
-- --~- ~~--~--~V.,
r---~\s I- ~ ....
vrll!
>- i
cn-R,.".,P All "olutl'1~~ 0'78 I

~ • l~ '''I:::::'':::::j
oi-'F-ROMf ~4ooo \'00 I
4001 - Soo ••. 0''10

5501-7000 o-8S
0-80
~1ooJ
, i
I
I
'I
I i31 ~' •• "QY C,'te<'l.QOIt'lt VOlurT'l~ tuostreorn of oft-tam9,

d~l'(~am of on-romp, t.t ....••n ~.'\.(c:mo .•) ar ?~lnt

I
LI _

Figure 2.16 CompucGcion of checkpoinc yolumes for an on-ramp

followed by an off-ramp. $Ollrce: -rRB (1"18~)
Multi-lane highways are those highways that are below freeway standard because they do not
have full control of access and, in some instances opposing directional lane groups are not
divided by a median.

The total service flow rate in one direction for the prevailing roadway and traffic conditions, for
any selected level of service, can then be calculated by using the following equation :-

SFi = MSFi x N x fw x fHV x fe x fp

Where :-

SFi = total service flow rate in one direction under prevailing roadway and traffic conditions
for level of service i;
Cj = capacity per lane for a multi-lane highway with a design speed j, i.e. 2000 pclh/lane for j
= 100 kmJh OT 110 k:m/h
N =number of lanes in one direction;
=maximum
~-)i volume/capacity ratio which can be accommodated at level of service i,
c
obtained from Table 2.22
fw = adjustment factor for lane width and/or lateral clearance restrictions, obtained from Table
2.23
fHV = adjustment factor for heavy vehicles
1
= [1 + PT (ET -1) + PR (ER - 1) + PB (EB -1)]
PT, PR and PB = the proportion of trucks, recreational vehicles and buses respectively in the
traffic stream, expressed as a decimal;
ET, ER and EB = the average passenger car equivalents for trucks, recreati9nal vehicles and
buses, obtained from Table 2.24, 2.25, 2.26, 2.27, 2.28 and 2.29.
fe = adjustment factor for the development environment and type of multi-lane road, obtained
from Table 2.30
fp = adjustment factor for driver population, obtained from Table 2.32.
2.4.2. Procedures for Application

The methodology ma.~ be applied in three ways :-

(i) Operational Analysis - this is intended to predict the operating characteristics of an

existing or planned roadway when subjected to a present or future demand.

The general approach taken in operational analysis is to use Equation 2.9 or Equation
2.10 to solve for the effective value of MSF or ~ ratio. This is then used to find the
level of service in Table 2.22 and to enter Figure 2.17 and 2.18 to find the likely
density and speed of the traffic.

(ii) Design - the objective of a design analysis is straightforward: the detennination of the
number of lanes needed in each direction on a multi-lane highway.

The general approach to design uses Equation 2.9 or Equation 2.10 to solve for N, the
number of required lanes. It should be noted that for significant grades, the upgrade
and downgrade must be considered separately.

(ill) Planning - the objectives of a planning analysis are similar to those in design. The
primary difference between design and planning is the detail of available information.
In the planning stage, details of horizontal and vertical alignments, and even of final
location, are not yet known.

The general computational approach in planning analysis is to convert the design year
AADT to a DDHV and apply a general estimate of service flow rate per lane to find N.
 2.17 A suburban multi-lane highway, four lanes (two lanes each
direction) is undivided with telephone poles 204m from the
pavement edge, 33m lanes, 100kmlh design speed and a driving
population of commuters. The highway is on rolling terrain and
has peak-hour directional volume of 1600 vehicles (10% trucks,
5% buses and 1% recreational vehicles) and a peak-hour factor of
0.90. Determine the level of service.

2.18 The highway described in problem 2.17 is improved to a six lane

divided multi-lane facility (three lanes each direction). However,
because of right-of-way restrictions, the lanes are only 2.7m wide,
obstructions are on both sides 0.6m from the pavement edge and
the design speed had to be reduced to 80 kmlh. If all traffic-
related conditions remain the same as those given in Problem
2.17, determine the new level of service.

2.19 Consider the multi-lane highway segment illustrated in Figure 2.6,

which shows an undivided, suburban multi-lane highway with
light standards located 0.6m from the travelled way at both
roadsides and bridge aboutments located in the centre of the
roadway at frequent intervals. The facility has 33m lanes which
narrow to 3.0m at bridge abutments. The design speed of the
segment is 100 kmlh, and the driver population consists
primarily of commuters. If the segment carries a peak hour
demand of 2000 vph in one direction, with 15 percent trucks and
a PHF of 0.91, what LOS can be expected in this segment?

L G=v6l- TeR~;'\1 N o
:=,ubuy v •.rolo'\-..J
bo.."" G •.•.
U-e<;i')\.o.. S~~ .:::IbO~/L--

:7
7
1°
06
'" ~

.
o
li~hj S1~~ol.",yol..>
 2.20 Consider the multi-lane highway segment illustrated in Figure 2.7.
It depicts a rural, divided, multi-lane highway with an ideal cross
section. An analysis of the existing level of service and operations
is desired for a segment of this highway on a significant grade of 3
percent, 1525m long.

The directional demand on this segment is 2200 vph in peak

periods, with 10 percent trucks, 5 percent RV's, and a PHF of 0.85.
The segment serves primarily recreational traffic. Determine the
level of service at which the multi-lane highway segment operates
during peak period (upgrade and downgrade).

RuYo..l El-\vryo",,~
\\ 0 K-"",,- { \.... 'Doe '> ,'j \,0, S p-e.<.. ol

2.21 A rural multi-lane highway segment on a long, steep grade must

be designed to accommodate a DDHV of 1000 vph, with 20 percent
trucks, and a PHF of 0.85.

The driver population is composed of regular users. The grade in

question is 6 percent, and is 1600m long. Because of the terrain,
the design speed on the segment is limited to 100 kmlh.

2.22 A planner must determine the most probable size of a multi-lane

highway to be built through a rural area of rolling terrain. The
AADT is expected to be 15,000 vpd, with 10 percent trucks. The
PHF in the region is generally 0.92, and the desired LOS is B. It
will be assumed that the highway will be divided.
The level of service can be estimated by computing a volume-to-
capacity ratio

vlc = ------------- SF
c j x N x f w x fH V x fE x fp

v 1600
SF == PHF == 0.9 == 1777.78 vph
Cj == 2000 pcphpl (100 km/h design speed)
N == 2 (given)
fw = 0.95 (303m lanes, no obstruction closer than 1.8m, Table 2.23 - used
N == not applicable use factor for one-side obstruction)
fE == 0.8 (suburban, undivided, Table 2.30).
fp == 1.0 (communters, Table 2.32).
ET == 4.0, EB == 3.0, ER == (rolling terrain, Table 2.24)

1
:. fHV == 1 + O. 10 (4 - 1) + 0.05 (3 - 1) + O. 0 1 (3 - 1)

1777.78
vlc == 2000 x 2 x 0.95 x 0.704 x 0.80 x 1. 0

which gives LOS E (Table 2.22) for 100 km/h design speed (Le. 0.8 <
0.83 < 1)

The primary judgement in this problem is the selection of the fw factor.

The major constriction occurs at bridge abutments, where lane width is
3.0m, with a roadside obstruction at 0.6m and a center obstruction at
Om. Other sections of the roadway, however, have 3.3m lanes, with a
roadside obstruction at 0.6m, and no center obstruction. Because the
most conservative analysis would consider the abutments, and because
these abutments are "frequent", the minimum condition will be used.
Note also that fw factors for center obstructions are specifically intended
for periodic isolated obstructions, which condition applies to the high
segment described.

vlc = ------------- SF
[Cj x N x fw x fHV x fE x fp]

Where :-
2000
SF = 0.91 = 2198 vph (given)
Cj = 2000 pcphpl (Table 2.22)
N = 2 (given)
fw = 0.80 (Table 2.23, undivided highway, 3.0m lane, obstructions both
sides at an average 0.3m - for 0.6m undivided, fw = 0.86, for Om
undivided, fw = 0.74, and average =
0.86 ; 0.74 = 0.80);

ET = 1.7 (Table 2.24, level terrain);

1
fHV = [1 + O. 15 (1. 7 _ 1)] = 0.90
fE = 0.80 (Table 2.30, suburban undivided highway);
fp = 1.00 (Table 2.32, commuters); and
2198
vlc = [2000 x 2 x 0.80 x 0.90 x 0.80 x 1.00]
= 0.95
Entering Table 2.22 with a vlc of 0.95 for a highway with 100 km/h
design speed, the LOS is found to be E. Entering Figure 2.18 with a vlc =
0.95, the expected speed of the traffic stream is 53 km/h. Entering
Figure 2.17 with this value, the expected density is 36 pc/km/h.

Where K varies from 0.15 to 0.20 for rural area (use 0.175) and D is
approximately 0.65 for most rural roads.
From Table 2.33, the per lane service volume for multi-lane highways IS
rolling terrain, with 10% trucks, at LOS B is :-

DDHV
N-- [SFL x fE x PHF]

1706
- [850 x 1.00 x 0.92] = 2.18 lanes

Therefore, to maintain a minimum of LOS B, three lanes would be

needed in each direction.
 Level of Density Design Speed Design Speed Design Speed
Service pc/km!lane 110 krnlh 100 kmIh 80 krn/h I
Speeda . v/eb MSF Speed vie MSF Speed vie MSF
krnlh kmlh kmIh
A ~7.5 ::::91 0.36 700 ::::80 0.33 650 - - -
B :; 12.5 :2:85 0.54 1,100 :2:77 0.50 1,000 :2:67 0.45 850
C ~ 18.8 ::::80 0.71 1,400 ::::70 0.65 1,300 :2:62 0.60 1,150
D ~26.3 :2:64 0.87 1,750 :2:64 0.80 1,600 :2:56 0.76 1,450
E ~ 41.9 ::::48 1.00 2,000 :2:48 1.00 2,000 :2:45 1.00 1,900
F > 41.9 <48 d d· < 48 d d < 45 d d
II
a. Average travel speed I
b. Volume/capacity ratio
c. Maximum rate of service flow per lane under ideal conditiC'ns, rounded to the nearest 50 pe/hllane.
d.' highly variable ._ ..

-
Soc.I>--c.e.: 7[(8 (1985)
Table 2.23' Adjustment factors for restrj.CJed lane width and latewl clearance

Distance from edge Adjustment factor

of travelled way to
obstruction Obstruction one Obstruction on both
m (ft)" side of roadwayb sides of roadway'

Lane width m (ft)

3.7 (12) 3.3 (11) 3.0 (10) 2.7 (9) 3.7 (12) 3.3 (11) 3.0 (10) 2.7 (9)

Divided multi-lane road - two lanes in each direction

:2:1.8 (6) 1.00 0.97 0.91 0.81 1.00 0.97 0.91 0.81
1.2 (4) 0.99 0.96 0.90 0.80 0.98 0.95 0.89 0.79
0.6 (2) 0.97 0.94 0.88 0.79 0.94 0.91 0.86 0.76
0 0.90 0.87 0.82 0.73 0.81 0.79
I 0.74 0.66

Divided multi-lane road - three lanes in each direction

:2:1.8 (6) 1.00 0.96 0.89 0.78 1.00 0.96 0.89 0.78
1.2 (4) 0.99 0.95 0.88 0.77 0.98 0.94
! 0.87 0.77
0.6 (2) 0.97 0.93 0.87 0.76 0.96 0.92 0.85 0.75
0.70
0 0.94 0.91 0.85 0.74 0.91 0.87
I 0.81

Undivided multi-lane road - two lanes in each direction

:2:1.8 (6) 1.00 0.95 0.89 0.77 N N N N

1.2 (4) 0.98 I 0.94 0.88 0.76 N N N N
0.6 (2) 0.95 0.92 0.86 0.75 0.94 0.91 0.86' N
0 0.88 0.85 0.80 0.70 0.81 0.79 0.74 0.66
I
Undivided multi-lane road - three lanes in each direction

~ 1.8 (6) 1.00 0.95 0.89 0.77 N N N N

1.2 (4) 0.99 0.94 0.88 0.76 N N N N
0.6 (2) 0.97 0.93 0.86 0.75 0.96 0.92 0.85 N
0 0.94 0.90 0.83 0.72 0.91 0.87
I 0.81 0.70

a. Use the average distance to obstruction on both sides when the distances to obstructions on the left and right sides
differ.
b. Factors allow for the effect of opposing flow.
c. For divided roads, this provides for an obstruction in the median as well as on the outside.
N '" not applicable. Use factor for one-sided obstruction.

Table.:?,< 4 Passenger car e~uivalents for trucks. bl~ses and

recreanonal vehicles on multi-lane roaas"

Vehicle type

Trucks, ET
Buses, E-s
Recreational vehicles, ER
GRADE LENGTH I
(%) I (~M) i
I
All I 2 2

I
~.e~;~bl
2: \'6
2 2
3
3
2 2
3

I ~:~-.2.41
=~:il
b'8-\'2
\.'2.
,~
5
4
4-
I 2: 2''f 6 5 4-

I~:~
~:~I
0'6 - I'b
5
6
6
4
5
5
4-
5
5 -
5
6
6
4-
5
5
I \·6 -:1.'4-1 i 6 5 6 5
! 2: ,2'4 i 6 5 6 5
O·{)-o·/.t-
o·Lt-o·6 5 9
0'8 -I'b 6 10
2:I'" 7 11
0'0-0'4 5 8
e·4-o·8 6 8
I O'B -1'6 8 12
I > \. 6 8 12
I 0'0- 0,4 6 I 9
0,4-,:>'5 7 I 11
1
0·6 ..,..1·2- 7 I 11
: :<: \.2 9 I 13

?ASSENGER·CAR EQUIVALENT, Er
GRADE LENGT:-{
(%) (\(.M) 4-LA,'-'" HIGHWA YS 6-LANE HIGHWAYS

All i
0'0-0 .41
0.4-- 0·8 i
0·8-1·2. i
\.2-1'61
> l·b
10.0"-0,'1!
'I 0<4- 0·8 I
o.g -I'b:
! > \·6 i
i 0·0-(>-«+1
!C.tt-\·61
I
i
>
0'0-
I'"o·q.
!
!
(>'4- I'b 'I
1 > l'b
 PASSENGER·CAR EQUlVALEI'-', Er
GRADE
(%) 4-lAN'E HIGHWAYS 6-LA.'""E HIGHWAYS

I All I 2
I 0·0 -0'41 2 - - -
I O·'T- 0'8 3 3
.,,
3
.
3 3 3 3 3 3 3 3 3 3 3 3 3
0'8 - 1·2 4 ~ ., 3 3 3 3 ~ ~ ~ 3 3 3 3 3
\.2.-I'b 5 ~ i ~ 3 3 3 3 5 4 ~ J.
- 3 3 3
.,,
1'0-:2'''1-
> .2 .~
6
7
5
5
5
5
5
5 ~
:1
:1 .
:1
.,
3
3
6
7
5 5
5
J.
5
J. J.
4
3
3
3

2 0'0- 0',+ ~ ~ .
., 3 3
J.
3 3 3 4
5
4 ~ 3
4

3 3 3
.
3
3
0'4-0'8 7 6 6 5 4 4 4 7 5 5 5 4 ., ., 4-
0·8 - \.2. S 6 6 5 5 :1 4 :1 S 6 6 6 5 5 4 .,,
\.2- -I-b 8 6 6 6 5 5 5 5 S 6 6 6 5 5 5 5
\'b ~)..'t 9 7 7 7 6 6 5 5 9 7 7 6 5 5 5 5
i > 2.·4- -+ 10 7 . 7 7 6 6 5 5 lO 7 7 6 5 5 5 5
3 o· 0 - 0'4 6 5 5 5 4 4 4 3 I 6 5 5 5 4 4 ~ 3
!>'4-Q'8 9 7 7 6 5 5 5 5 8 7 7 6 5 5 5 5
0'8 -\ ..2 12 8 S 7 6 6 6 6 10 8 7 6 5 5 5 5
\.2. - \'b 13 9 9 8 7 7 7 7 11 8 8 7 6 6 6 6
I > l'b 14 lO 10 9 8 8 7 7 I 12 9 9 8 7 7
..
7 7
4 I 0·0-0',+ 7 5 5 5 4 ~ 4 ~ 7 5 5 5 4 ., 3 3
o''i-o'8 12 8 8 7 6 6 6 6 10 8 7 6 5 5 5 5
,0,8-1'2- 13 9 9 8 7 7 7 7 II 9 9 8 7 6 6 6
I\.2. -1'6 15 10 10 9 8 8 8 8 12 10 10 9 8 j 7 7
> 1'6 I 17 12 12 10 9 9 9 9 13 10 10 9 8 8 8 S
'0·0-0·1f 8
o·'t- 0'8 13
0'8 - ,.'2. 20
> 1·2. 22

I ~~~O;~
I I~
I >" -6 I 28

GRADE I LENGTH I
(%) tKI'ol) I
I
All I
0'0-0'31 2 2
~ 0 ·8 3

O'o-o''t
t>.!+ -1·2-
I 2
3
2
3
~ \.2-
0'0-041
a.tot-I·z..1
~ \ .2-
0'0 -""''11
C>'4 - \.2.
~ \.'2. 1

:"iOTE; If il leng"..h ot g:-ac.:> {ill.s..on-a tx>uo.duy conG.ition. tbc equivalent troeo. t.1C longer gr:lde c:ia.s.! t:l uSc:d •• o\..'~grade. ~ee!='C( than the perc.::1C s~ow:'\ ffit:st :.IS-:the: nc;;;C his;"'er grade.
:.a.tcgory.

Tablez.,z9 Passenger Car Equivalents for Buses on

Upgrades on Mu.lti-Lane Roads Table2·30 Adjustment Factor for Development
Environment and Type of Multi-Lane Road
Passenger Car Equivalent, EB
1.6 Development Road Type
0-3
4 1.6 I Environment Divided Undivided
5 3.0 Rural 1.00 0.95
6 5.5 Suburban 0.90 0.80
 DENSITY SPEED
(PC/KWLS) (kwW
110 Klo"I/H ELEMEi'7S

0.30 600 A 6.!; q'2..

iI·S
0.36" 700 A 92.-
0.40 800 B E'f
0.50 1.000 B \1.,,- qo
0.54" 1.100 B 12.'5 ~7
85
0.60 1.200 C
13- -;:
0.70 1,4OJ C 8<+
Ii· L.f
0.71" 1,4OJ C 80
18'6 eo
1.600 D 21.1 76
\00 Kt-1/H ELEME~

82-
0.30
0".J) "
600
650
A
A
I iL I
1"5 &0
0040 800 B q.f:,
0.50" 1.000 B
7q
\2.. 5
77
0.60 1.200 C
0.65" 1.300 c 1'5'5 72-
18'6 71
0.70 1,4OJ D l'1·q
'0.80" ;J. L{ • .,
68
1.600 D 64
60 \( ¥I I H . ELEMESTS
0.30 550 B 8·, 69
0.40
0.45" I 750
850
B
B
to."t
\2·5
68
~8
C.50 [ 950 c \'t."! 56
0.60" 1,150 c
0.70
I 1.350 D
\6·6
2.3·{,
63
0.76" 1,~50 D be
26,) Sl;
I
I 1.500 E 2(:·Q 55
~ Desitn may be wilhL"1 a LOS.
" ~luimum ',hr<shold 1/< ior LOS shown. $cu..-c.e : -rF<8
c:Ro~nd~d :0 the :"lares: 50 pcphpl. (r9 85)
Table 2'3:> Service Flow Rate per Lane on Multi-Lane
Roads for Planning Applications for a Design Speed of
110 kmlh (Vehicles/how)

Level of Percent Trucks

Service
0 5 10 15 20
Level Terrain
Driver Population
Commuter, or other regular user 1.00
Factor

Recreational, or other non-regular user 0.75 - 0.90 *

Engineering judgement and/or local data should be used

l
I
A
B
C
D
E
700
1,100
1,400
1,750
2,000
700
1,050
1,350
1,700
1,950
650
1,000
1.300
1,650
1,850
650
1,000
1,250
1,600
1,800
600
1,000
1,250
1,550
1,750
in selecting an appropriate value
Rolling Terrain
Source: TRB (1985)
A 700 600 550 500 500
B 1,100 950 850 800 700
C '1,400' 1,200 1,100 1,000 900
D 1,750 1,500 1,350 1,250 1,100
E 2,000 1,750 1,550 1,450 1,250
Mountainous Terrain
A 700 500 *00 350 300
B 1,100 800 650 550 450
C 1,400 1.050 850 700 600
D 1,750 1,300 1,050 850 750
E 2,000 1,500 1,200 1,000 850

Source: TRB (1985)

., ~3.
c

~'" 37.2.
x
....
~ 3J·o
)-
I-
Ui
Z
1LI
o

capacity
reflects t< "'/H
speed limit
vIe ratio based on capacity 01 2000 pcphpl. applies only to 100 and 110 Kl'l/l/deslgn speeds

o
w
1LI
a..
CJl
..J

~ ~
<i
--_. --- --- ---------
II:
I-
1LI
Cl
<
a:
1LI
>
<i

246 8 10 12 14 16 18 20 VOLUME/LANE (pcphpl)

(0.10) (0.20) (0.30) (0.40) (0.50) (0.60) (0.70) (0.80) (0.90) (1.00) vIe RATIO •••

capacity
reflects 55 MPH sp"ed limit
vie ratio based on capacity 01 2000 pcphpl. applies only 10100 andllO Kt'l/I-l design speeds
Two lane rural roads have one lane for use by traffic travelling in each direction. Overtaking of
slower vehicles requires the use of the opposing traffic lane when sight distance and gaps in the
opposing traffic stream permit. As traffic volumes and/or geometric restrictions increase, the
ability to overtake decreases and platoons or bunches form in the traffic stream. Vehicles in
these platoons or bunches are subjected to delay because of the inability to overtake slower
moving vehicles.

Ideal conditions for two-lane rural roads occur when there are no restrictive geometric, traffic
or environmental conditions. More specifically, they occur when all of the following apply :-
(i) The design speed is equal to or greater than 100 km/h;
(ii) Traffic lane widths are equal to or greater than 3.7m;
(iii) Clear shoulder widths are equal to or greater than 3.7m;
(iv) There are no lengths of road with a sight distance of less than 450m;
(v) The traffic stream consists of passenger cars only;
(vi) There is a 50/50 directional split of traffic;
(vii) There are no restrictions due to traffic control or turning vehicles; and
(viii) The terrain is level.

Under these conditions, the capacity of a two-lane rural road is 2800 passenger cars per hour
total for both directions offlow.

Percent time delay is defined as the average percent of time that all vehicles are delayed while
travelling in platoons due to the inability to pass. "Percent time delay" is difficult to measure
directly in the field. The percent of vehicles travelling at headways less than 5 seconds can be
used as a surrogate measure in field studies.

Level of service criteria for the three general terrain classifications are given in Table 2.34. For
each level of service, the percent time delay and the average travel speed of all vehicles are
shown. The volume/capacity ratio for each level of service and the prevailing terrain and
percentage of the road with sight distance less than 450m are also shown.'

The total service flow rate for both directions of flow for the prevailing roadway and traffic
conditions, for any selected level of service, can then be calculated by using the following
equation :-

SFi = 2800 x et"J x fd x fHY x fw

SFi = total service flow rate in both directions under prevailing roadway and traffic conditions
for level of service i;
I..Y..) = maximum volume/capacity ratio which can be accommodated at level of service i for a
c
given terrain and percent of length with no overtaking, obtained from Table 2.34;
fd = adjustment factor for directional distribution of traffic, obtained from Table 2.37;
fw = adjustment factor for narrow lanes and shoulders, obtained from Table 2.38;
fHY = adjustment factor for heavy vehicles
1
=[1 + PT (ET -1) + PR (ER - 1) + PE (EE - 1)]
PT, PR and PE = the proportion of trucks, recreational vehicles and buses respectively in the
traffic stream, expressed as a decimal;
ET, ER and EE = the average passenger car equivalents for trucks, recreational vehicles and
buses, obtained from Table 2.39.

The above procedure for the analysis of general terrain classifications is usually applied to road
sections of at least 3km in length. Procedures are also available for analysing specific grades.

The service flow rate for any given average upgrade speed is given by the following
relationship :-

(v/c)i = vie ratio for level-of-service i or speed i, obtained from table 2.40;
fd and fw = obtained from table 2.451 and 2.38 respectively;
fg = adjustment factor for the operational effects of grades on passenger cars;

1
fg = [1 + (Pp Ip)]

Pp = Proportion of passenger cars in the upgrade traffic stream, expressed as a decimal;

Ip = impedance factor for passenger cars, computed as :-

E = base passenger-car equivalent, for a given grade, and speed, selected .from table 2.42
Eo = base passenger-car equivalent for 0 percent grade and a given speed, selected from table
2.42.
fHY = adjustment factor for the presence of heavy vehicles in the upgrade traffic stream;

f _ 1
HY - [ 1 + PRV (ERV - 1) ]

PHY = total proportion of heavy vehicles (truck + RV's + buses) in the upgrade traffic stream;
EHY = passenger-car equivalent for specific mix of heavy vehicles present in the upgrade traffic
stream, computed as :-
PTIHV= proportion of trucks among heavy vehicles, i.e., the proportion of trucks in the traffic
stream divided by the total proportion of heavy vehicles in the traffic stream; and
E= base passenger-car equivalent - selected from table 2.42.

Operational Analysis of General Terrain Segments - the objective in operational analysis

is to determine the level of service for a given segment or segments of roadway for a
known existing set of conditions, or for future set of conditions which are hypothesized
and/or forecast.

The general approach will be to compute service flow rates for each level of service and
compute service flow rates for each level of service and compare these values with the
exiting flow rate on the facility.

Operational Analysis of Specific Grades - The operational analysis of specific grades is

similar to the procedure for general terrain segments. The level of service for the
upgrade direction is sought, and is found by comparing an actual two-way flow rate to
the service flow rates for the various levels of service.

Planning - Table 2.43 may be entered with a known or forecast AADT to determine
expected level of service during the peak 15 minutes of flow, or with a known LOS to
fmd the maximum allowable AADT.
2.23 A rural two-lane highway is on level terrain with 33m lanes, 0.6m paved shoulders
and 80 percent no passing zones. The directional split is 80120 and there are 5 percent
trucks, 2 percent buses and 5 percent recreational vehicles. Determine the service flow
of the roadway at capacity.

2.24 Consider the conditions described in problem 2.23. If the peak-hour vehicles count is
522 with a peak-hour factor of 0.90, determine the level of service.

2.25 A segment of rural two-lane highway is expected to have thefollowing characteristics:-

(a) Roadway characteristics -100 km/h design speed, 307m lanes; 3.Ompaved
shoulders; level train; 0 percent no passing zones; length = 8km.

(b) Traffic characteristics - 70130 directional slit; 10 percent trucks; 5 percent

recreational vehicles; 1 percent buses; 84 percent passenger cars.

What is the capacity of the section? What is the maximum flow rate which can be
accommodated at level-of-service C?

2.26 A two-lane rural highway carries a peak hour volume of 180 vph and has the following
characteristics: -

(a) Roadway characteristics - 100 km/h design speed; 33m lanes; 0.6m shoulders;
mountainous terrain; 80 percent no passing zones; length 16km.=
(b) Traffic characteristics - 60140 directional split; 5 percent trucks; 10 percent
recreational vehicles; no buses; 85 percent passenger cars. At what level of
service will the highway operate during peak periods?

2.27 A rural two-lane highway in mountainous terrain has a 6 percent grade of 3.2 km.
Other relevant characteristics include :-

(a) Road characteristics - 307m lanes; 204m shoulders; 60 percent no passing

zones.

(b) Traffic characteristics - 70130directional split; 12 percent trucks; 7 percent

recreational vehicles; 1percent buses; 80 percent passenger cars; PHF = 0.85.

What is the maximum volume that can be accommodated on the grade at a speed of 64
km/h (LOS D, Table 235)?

2.28 A rural two-lane highway has a 4 percent upgrade of204km and has thefollowing other
characteristics :-

(a) Roadway characteristics -level terrain approach; 3.7m lanes; 204mshoulders;

40 percent no passing zones.

(b) Traffic characteristics - DHV = 400 vph; 15percent trucks; 5 percent

recreational vehicles; 1 percent buses; 79 percent passenger cars; 60140
directional split; PHF = 0.85,

Is the addition of a climbing lane justified at this location?

2.29 A rural two-lane highway in mountainous terrain is located in an area where the design
hour factor, k, is 0.14. What is the maximum AADT which can be accommodated
without the LOS falling below D during the peak 15 minute flow period?

2.30 A rural two-lane highway is located in rolling terrain in an area where the design hour
factor, K is 0.12. It's current AADT is 1,000 vpd. What is the likely LOS during the peak 15
minutes of flow?

The roadway reaches capacity at LOS E which, from Table 2.34, gives a vie ratio of 1 on level
terrain with 80% no-passing zones. Thus, the service flow at capacity can be computed using
Equation 2.11 :-

VicE = 1.0 (Table 2.34)

fd = 0.83 (80120 directional split, Table 2.37)
fw = 0.88 (303m lanes, 0.6m shoulders, LOS E, Table)
ET = 2.0, ER = 1.6, EB = 1.6 (level terrain, LOS E, Table 2.34)

1
:. fHv= 1 + 0.05 (2 - 1) + 0.05 (1.6 - 1) + 0.02 (1.6 - 1)

= 0.916

SFE = 2800 x 1.0 x 0.83 x 0.88 x 0.916

= 1 873.33 veh/hr

The solution to this problem is found by computing the service flow rates for levels-of-service
C and E (capacity) using Equation 2.11 :-

SFi = 2800 (V/C)ix fd x fw x fHY

f _ 1
HY-[l + PT (ET - 1) + PR (ER - 1) + PB (EB - 1)]
(v/c)c = 0.43 (Table 2.34, level terrain, 0% no passing zones, LOS C);
(V/C)E = 1.00 (Table 2.34, level terrain, 0% no passing zones, LOS E);
fd = 0.89 (Table 2.37, 70/30 split);
fw = 1.00 (Table 2.38, 3.7m lanes, > 0.6m shoulders);
ET = 2.2 for LOS C, 2.0 for LOS E (Table 2.39, level terrain);
ER = 2.5 for LOS C, 1.6 for LOS E (Table 2.39, level terrain);
EB = 2.0 for LOS C, 1.6 for LOS E (Table 2.39, level terrain).
PT = 0.10 (given);
PR = 0.05 (given); and
PB = 0.01 (given).

1
fHV (LOS C) = [l + 0.10 (2.2 _ 1) + 0.05 (2.5 _ 1) + 0.01 (2.0 _ 1)] 0.83

1
fHV (LOS E) = [1 + 0.10 (2.0 _ 1) + 0.05 (1.6 _ 1) + 0.01 (1.6 _ 1)] = 0.88
and :-

SFc = 2800 x 0.43 x 0.89 x 1.00 x 0.83 = 889 vph

SFE = 2800 x 1.00 x 0.89 x 1.00 x 0.88 = 2193 vph

Thus, the highway will have an expected capacity of 2193 vph, total in both directions and can
accommodate a flow rate of up to 889 vph at level-of-service C.

It is assumed that a climbing lane on a two-lane highway is generally justified when the
following conditions are met:-

1. Upgrade flow rate is greater than 200 vph.

2. Upgrade truck flow rate is greater than 20 vph.

3. One of the following occurs :-

(a) The grade operates at LOS E or F.
(b) The typical heavy truck reduces its speed by more than 16 kmIh on the grade.
(c) The LOS on the grade is two or more levels poorer than on the approach to the
grade.

Each of these conditions should be checked to justify the construction of the climbing lane:-

400 x 0.60
Upgrade flow rate = 0.85 = 285 vph > 200 vph okay
 400 x 0.15 x 0.60
Upgrade trucks = 0.85 = 42 vph > 20 vph okay

To justify a climbing lane, only one of the conditions specified in item 3 must be demonstrated.
The LOS will be E or worse if the actual flow rate exceeds the service flow rate for LOS D.
This value is computed using Equation 2.12 :-

1
fg = [1 + P p Ip (2.13)
Ip = 0.02 (E - Eo) (2.14)

1
fHV = [1 + PRV (ERV - 1)] (2.15)

EHV = 1 + (0.25 + PTIHV) (E - 1) (2.16)

(V/C)D= 1.00 (Table 2.40, 4% grade, 64 km/h, 40% no passing zones);

fd = 0.87 (Table 2.41, 60/40 directional split);
fw = 1.00 (Table 2.38);
E = 3.8 (Table 2.42, 4%, 2.4 km grade, 64 km/h);
Eo = 1.3 (Table 2.42, 0%, 64f km/h);
PHV = 0.15 + 0.05 + 0.01 = 0.21; and
PTIHV = g:~i= 0.71

Ip = 0.02 =
(3.8 - 1.3) 0.05
1
fg = [1 + (0.79 x 0.05)] = 0.96
ERV = 1 + (0.25 + 0.71) (3.8 - 1) = 3.69
fRV = \f(l,[1 + 0.21 (3.69 - 1)] = 0.64
SFD = 2800 x 1.00 x 0.87 x 1.00 x 0.96 x 0.64 = 1497 vph

The actual flow rate is the DHV divided by the PHF, or ri.~ = 471 vph. As this is clearly less
than the service flow rate for LOS D, the existing LOS is not E, and this condition is not met.

The next condition to investigate is whether a 16 km/h speed reduction of heavy trucks would
exist on the grade described. Based on the assumption that the typical truck on this grade has a
weight/horsepower ratio of 200 lb/hp, Figure 2.19 is used to estimate the speed reduction
experienced as shown below :-

It can be seen that the speed reduction will be well in excess of 32 km/h, which is greater than
16 km/h, fulfIlling the last required condition for justifying a climbing lane. Note that because
only one of the conditions in item 3 needs to be satisfied, it is not necessary to investigate the
third condition.

The solution is simply found by entering Table 2.43 with mountainous terrain, LOS D, and a
k-factor of 0.14. The maximum permissible AADT is found to be 2700 vph.

Again, the solution is straightforward using Table 2.34. The maximum AADT's for the
various levels of service are found for rolling terrain and a k-factor of 0.12. The 5000 AADT
is seen to fall between the maximum values for LOS C (4400 vpd) and LOS D (6600 vpd).
The LOS is therefore expected to be D during the peak 15 minutes of flow.
 Table)..· 34-
Level of Service Criteria for Two-Lane Two- Way Rural Roads for General Terrain Classification

Level of Percent Time Average - Volume/Capaci[)' Ratio' for Level Terrain-

Service Delayed Speedb Percent of length with sight distance less than 450m

0 20 40 60 80 100
-- A < 30 > 93 0.15 0.12 0.09 0.07 .. 0.05 0.04
B <45 > 88 0.27 . 0:24 0.21 0.19 0.17 0.16
C <60 > 83 0.43 0.39 0.36 0.34 0.33 0.32
D <75 > 80 0.64 0.62 0.60 0.59 0.58 0.57
·E >75 >72 1.00 1.00 1.00 1.00 1.00 1.00
F 100 <72 - - - - - -
Level or Percent Time Average Volume/Capacity Ratio' for Rolling Terrain
Service Delayed Speedb Percent of length with sight distance less than 450m

0 20 40 60 80 100
A <30 > 91 0.15 . 0.10 0.07 0.05 0.04 0.03
B <45 > 86 0.26 0.23 0.19 0.17 0.15 0.13 I
C <60 > 82 0.42 0.39 0.35 0.32 0.30 0.28
D <75 > 78 0.62 0.57 0.52 0.48 0.46 0.43
E >75 >64 0.97 0.94 0.92 0.91 0.90 - 0.90
F 100 <64 - - - - - -
Level of Percent Time Average Volume/Capacity Ratio' for Mountainous Terrain
Service Delayed Speedb Percent of length with sight distance less than 450m

0 20 40 60 80 100
I
A <30 > 90 0.14 0.09 0.07 0.04 0.02 0.01
B <45 > 86 0.25 0.20 0.16 0.13 0.12 0.10
C <60 >78 0.39 0.33 0.28 0.23 0.20 0.16
D < 75 >72 0.58 0.50 0,45 0040 0.37 0.33
E >75 > 56 0.91 0.87 0.84 0.82 0.80 0.78
F 100 < 56 - - - - - -
I
a. Ratio of the flow rate to an ideal capacity of 2,800 pe/h.
b. Average speed of an vehicles in kmJh for roads with a design speed equal to or greater than 100 kmfh.
For roads with lesser design speed. reduce the speed by 6 kmJh for each 16 km/h reduction in design speed.
Sou~ce : .TRB(l985)

Table )'.3JLevel of Service Criteria for Specific Grades

on Two-Lane Two-Way Roads

Level of A verage Upgrade I

Service Speed (km/h)
I
A > 88
B > 80
C >72
D >64
E > 40-64'
F < 40-64'

a exact speed at which capacity occurs varies with percent

and length of grade, traffic composition and volume.
SoV'....ce: TRB (1985)
 TOTAL 2·WAY PEAK HOUR TOTAL 2·WAY PEAK HOUR
HOURLY VOLUME FACTOR HOUR!.Y vOLUME FACTOR
(VPH) (PHF) (¥PH) (PHF)

100 0.83 I.CXXJ 0.93

200 0.87 1.100 0.94
300 0.90 1,200 0.94
.iOQ 0.9l [.300 0.94
500 0.91 1.400 0.94
600 0.92 1.500 0.95
700 0.92 1,600 0.95
800 0.93 [.700 0.95
900 0.93 1.800 0.95
~ [.900 0.96

Leve[ of Service ABC D E

Peak Hour Factor .• 0.91 0.92 0.94 0.95 1.00

Table '),,3.7 Adjustrr.ent Factors for Directional Distribution of Traffic

on General Te'!.ain Segments

Directional Distribution 100/0 90/10 80/20 70(30 60/40 SO/50

AdjusunentFactorfd 0.71 0.75 0.83 0.89 0.94 1.00

Usable 3.7trn)ldMS 3.3(rn) lo.",es 3.0(rn) !o..•..•e5 2.7(m) Io..~e.s

shoulder
WIdth' LOse LOSb LOS LOSh LOS LOSh LOS LOSh
rn AtoD E AtoD E A to D E A to D E

2: 1.8 1.00 1.00 0.93 0.94 0.84 0.87 0.70 0.76

1.2 0.92 0.97 0.85 0.92 0.77 0.85 0.65 0.74
0.6 0.81 0.93 0.75 0.88 0.68 0.81 0.57 0.70
0 0.70 0.88 0.65 0.82 0.58 0.75 0.49 0.66

a. When the shoulder width is different on each side of the road use the average shoulder width.
b. This factor applies for all speeds less than 72 kmlh
c. LOS is level of service.

Table;(.3 t3 Average passenger car equivalent for trucks. recreational vehicles

and buses on two-lane rural roads

Type of terrain

Vehicle type Level of service Level Rolling Mountainous

Trucks, ET A 2.0 4.0 7.0

Band C 2.2 5.0 10.0
D and E 2.0 5.0 12.0
Recreational vehicles. ER A ~7
.".- 5.0
2.2
Band C 2.5 3.9 5.2
D and E 1.6 3.3 5.2
Buses, Es A 1.8 3.0 5.7
Band C 2.0 3.4 6.0
D and E 1.6 2.9 6.5

Source: TRB (1985)

 Table2.4oVolumelCapaciry Ratiosfor Varying Percent Upgrades. Average Upgrade
Speeds and Percent Length with Sight Distance less than 450m

Percent LOS Average Upgrade Volume/Capacity Ratios'

Upgrade Speed (km/h) Percent of Length with Sight distance
less than 450m
0 20 40 60 80 100
3 A 88 0.27 0.23 0.19 0.17 0.14 0.12
B 80 0.64 0.59 0.55 0.52 0049 0.47
C 72 1.00 0.95 0.91 0.88 0.86 0.84
D 1.00 1.00 1.00 1.00 1.00 1.00
I 64
4 A 88 0.25 0.21 0.18 0.16 0.13 0.11
B 80 0.61 0.56 0.52 0.49 0.47 0.45
C 72 0.97 0.92 0.88 0.85 0.83 0.81
D 64 1.00 1.00 1.00 1.00 1.00 1.00

5· A 88 0.21 0.17 0.14 0.12 0.10 0.08

B 80 0.57 0.49 0.45 0.41 0.39 0.37 I
C 72 0.93 0.84 0.79 0.75 0.72 0.70
D 64 0.98 0.96 0.95 0.94 0.93 0.92
E-F 56 1.00 1.00 1.00 1.00 1.00 1.00

6 A 88 0.12 0.10 0.08 0.06 0.05 0.04

B 80 0048 0040 0.35 0.31 0.28 0.26
C 72 0.85 0.76 0.68 0.63 0.59 0.55
D 64 0.97 0.91 0.87 0.83 0.81 0.78
E-F 56 1.00 0.96 0.95 0.93 0.91 0.90
E-F 48 1.00 0.99 0.99 0.98 0.98 0.98
7 A 88 0.00 0.00 0.00 0.00 0.00 0.00
B 80 0.34 0.27 0.22 0.18 0.15 0.12
I C 72 0.77 0.65 0.55 0.46 0040 0.35
D 64 0.93 0.82 0.75 0.69 0.64 0.59
E-F 56 1.00 0.91 0.87 0.82 0.79 0.76
E-F 48 1.00 0.95 0.92 0.90 0.88 0.86
a Ratio of flow rate to ideal capacity of 2,80OpcIh, assuming passenger car operation
is unaffected by grades.

Table :l.£+IAdjustment Factors for Directional

Distribution of Traffic on Specific Grades if)

Percent of Adjustment
Traffic on Factor
Upgrade

100 0.58
90 0.64
80 0.70
70 0.78
60 0.87
50 1.00
40 1.20
30 1.50
Source: TRB (1985)
 Table ~.Lj2Passenger Car Equivalent for Specific
Grades on Two-Lane Two-Way Rural Roads

Percent Length of Average Upgrade Speed km/h

Grade Grade (km) 88 80 72 64 48
0 all 2.1 1.6 1.4 1.3 1.3

3 0.5 3.1 2.1 1.8 1.6 1.5

1.0 4.2 2.6 2.1 1.9 1.8
1.5 6.0 3.3 2.5 2.2 2.1 Table;;.43Maximum AADTsfor Various Levels of
2.0 8.0 4.3 3.0 2.6 2.3 Service and Types of Terrain on Two-Lane Two-Way
3.0 17.0 6.3 4.3 3.5 2.8 Rural Roads
4.0 36.0 8.8 5.9 4.6 3.3
a
6.0 a 18.0 10.0 7.2 4.6 K factor Level of Service

4 0.5 3.5 2.4 1.9 1.8 1.7 A B C D E

1.0 5.3 3.1 2.5 2.2 2.0 Level Terrain
1.5 8.5 4.2 3.0 2.5 2.3
0.10 2,400 4,800 7,900 13,500 22,900
2.0 13.0 5.9 4.0 3.2 2.8
6.4 0.11 2,200 4,400 7,200 12,200 20,800
3.0 35.0 10.0 4.9 3.6
0.12 2,000 4,000 6,600 11,200 19,000
4.0 a 15.0 10.0 7.1 4.7
0.13 1,900 3,700 6,100 10,400 17,600
6.0 a 43.0 20.0 12.0 6.9
0.14 1,700 3,400 5,700 9,600 16,300
0.15 1,600 3,200 5,300 9,000 15,200
5 0.5 4.0 2.5 2.1 1.9 1.8 I
1.0 6.9 3.8 2.8 2.4 2.2
1.5 5.4 Rolling Terrain
12.0 3.7 3.1 2.7
2.0 24.0 7.8 5.1 4.1 3.3 0.10 1,100 2,800 5,200 8,000 14,800
I
i
3.0 70.0 16.0 9.2 6.9 4.6 0.11 1,000 2,500 4,700 7 ;200 13,500 I
4.0 a 27.0 16.0 11.0 6.3 0.12 900 2;300 4,400 6,600 12,300 I
6.0 a a 47.0 22.0 10.0 0.13 900 2,100 4,000 6,100 11,400
0.14 800 2,000 3,700 5,700 10,600 II
6 0.5 4.6 2.8 ?~
-.::> 2.0 1.9 0.15 700 1,800 3,500 5,300 9,900 I

1.0 8.7 4.5 3.3 2.8 2.5

1.5
2.0
17.0
36.0
7.0
12.0
4.5
6.7
3.7
5.2
3.1
4.0
Mountainous Terrain
II
0.10 500 1,300 2,400 3,700 8,100 I
3.0 a 25.0 14.0 9.6 5.9\ 0.11 400 1,200 2,200 3,400 7,300 I
4.0 a 50.0 27.0 17.0 12.0 0.12 400 1,100 2,000 3,100 6,700 I
6.0 a a 77.0 40.0 16.0 0.13 400 1,000 1,800 2,900 6,200 I
0.14 300 900 1,700 2,700 5,800
7 0.5
1.0
5.3
11.0
3.1
5.2
2.4
3.7
2.2
3.1
2.0
2.7
0.15 300 900 1,600 2,500 5,400 I
I
I
1.5 26.0 8.8 5.5 4.4 3.6 I a. K is the rario of the design hour volume to the annual I
2.0 average daily traffic.
60.0 16.0 8.8 6.6 4.8
3.0 a . 40.0 20.0 14.0 7.6 i
I
b. All values rounded to the nearest 100 vehicles per day. I
I Assumed conditions include 60140 directional split. 14
4.0 a a 44.0 27.0 12.0 I
percent trucks. 4 percent buses. Percentages of length
6.0 a a a a 26.0 ! with no overtaking for level. rolling and mountainous I
I i terrain of 20, 40 and 60 have been assumed. !
I a. Speed not attainable on the grade. I

5o(./)-ce : Source: TRB (1985) .

TRB (1985)
 10 SPEED REDUCTiON BELOW AVERAGE SPEED REDUCTION BELOW
9
RUNNING SPEED OF ALL TRAFFIC (Kmln) AVERAGE "RUNNING
9 SPEED OF ALL TRAFFIC (I<"yh)
8
8
7 7

~ 6 ~
0
6
0

w
w 5 0
5
0 <l:
<l: 4 0:::
0:::
<:>
Q..
<:> 4
Q..

::> 3 ::>
3
'2
2

I
0 ~'+s 305 <;"57

LENGTH OF GRADE (rn)

Improvement and widening of shoulders;

Removal of lateral obstructions closer to 2.0 m to the pavement edge, e.g. by widening
narrow bridges and culverts;

Improvement (and proper maintenance) of the pavement surface; and

Provision of truck climbing lanes - these may virtually remove slow-moving vehicles
from the uphill traffic stream on long grades. The additional lane should be carried far
enough over the crest to permit trucks to regain sufficient speed to merge with the
passenger car traffic without undue interference.
 SIGN:A.T"IZE:D
INTER.SECTION"S

Most road intersections are not signalized due to low traffic volumes and adequate sight distances.
Furthermore, the economic expense of installing traffic signals and the possible increase in vehicle
delay that signals may cause argue against their use. However, at some point, traffic volumes (and
other factors) reach a level that make signalization economically feasible.

Following are a number of definitions of commonly used intersection-related terminology.

Approach A lane or group of lanes through which traffic enters an intersection.

Cycle One complete sequence (for all approaches) of signal indications (greens,
yellows, reds).

Cycle Length (c) The total time for the signal to complete one cycle.

Traffic Signal Phase The part of the cycle length allocated to a traffic movement that has the right
of way, or any combination of traffic movements that receive the right of
way simultaneously. The sum of the phase lengths (in seconds) is the
cycle length.

Indication The illumination of one or more signal lenses (greens, yellows, reds)
remain the same for all approaches.

Green Time (GT) The time within a cycle in which an approach has the green indication.

Yellow Time (IT) The time within a cycle in which an approach has the yellow indication.

All-Red Time (AR) The time within a cycle in which all approaches have a red indication.

Change Interval The yellow time plus all-red time (the short period of time in which all
approaches have a red signal) that provides for clearance of the intersection
before conflicting traffic movements are given a signal indication that
allows them to enter the intersection (expressed in seconds).
Lost Time (LT) Time during which the intersection is not effectively used by any approach.
These times occur during the change interval (when the intersection is
cleared) and at the beginning of each green indication as the first few
vehicles in a standing queue experience start-up delays.

Effective Green (g) The time that is effectively used by the approach for traffic movement.
This is generally taken to be the green time plus the change interval minus
the lost time for the approach.

Effective Red (r) The time that is effectively not used by the approach for traffic movement.
Stated in seconds, it is the cycle length minus the effective green time.

Saturation Flow (s) The maximum flow that could pass through an intersection from a given
approach, if that approach were allocated all of the cycle time as effective
green with no lost time.

Approach Capacity The maximum flow that can pass through an intersection under prevailing
highway and traffic conditions, given the effective green time allocated to
the approach. It is equal to the saturation flow multiplied by the ratio of
effective green to cycle length (C = s x glc).

Major Street The street at an intersection that has the higher traffic-volume approaches.

Minor Street The street at an intersection that has the lower traffic-volume approaches.

Protected Turn A turning movement made without the conflict of opposing traffic or
pedestrians.

Pennitted Turn A turning movement that is made through opposing traffic flow or through
conflicting pedestrian movement.

Signal Timing The operating characteristics of the signal with the parameters being the
signal cycle length, green time, red time, yellow time and all-red time.

Pretimed Signal A signal whose timing is fixed over specified time periods and does not
change in response to changes in traffic flow at the intersection.

Semi-Actuated Signal A signal whose timing is affected when vehicles are detected (eg. by
magnetic-loop detectors in the pavement) on some, but not all, approaches.

Fully Actuated Signal A signal whose timing is completely influenced by the traffic volumes,
when detected, on all of the approaches.

ANAL YSIS OF SIGNALIZED INTERSECTIONS WITH DIDIl

QUEUING
The assumption of DID/l queuing (as discussed in chapter 1) provides a strong intuitive appeal that
helps understanding of the analytic fundamental underlying traffic analysis at signalized
intersections. To begin applying DID/l queuing to signalized intersections, we consider the case
where the approach capacity exceeds the approach arrivals. Under these conditions the assumption
of DID/l queuing will result in a queuing system as shown in Fig. 1, where

A = the arrival rate (typically in vehicles per second)

11 = the departure rate (in vehicles per second)
g = the effective green (in seconds)
r =the effective red (in seconds)
t is the total transpired time (in seconds)
"arrivals At" line gives the total number of vehicle arrivals at time t,
"departures J1t" line gives the slope of vehicle departures (number of vehicles that depart) during
effective greens
to is the time from the start of the effective green until queue dissipation (in seconds), and
c = cycle length (in seconds).
k = per-cycle approach arrivals
Ilg = per-cycle approach capacity departure

Figure 1. DID/1 signalized intersection queuing with approach

capacity (J1g) exceeding arrivals (k) for all cycles.

1. For the time to queue dissipation after the start of the effective green, to' note that ACr + to) = J1to
and, with traffic intensity p = Yll (as in Chapter 1),

pr
t =---
o (1- p)
2. The proportion of the cycle with a queue, Pq, is

P = r+to
q c

P = A.-(r+to) = r+to = P
S A.-(r+g) c q

P =A.-(r+to)=llto=~
A(r+ g)
S Ac pc

A.-r2
D=--
t 2(1- p)

6. The average delay per vehicle, d, is

A.-r2 1
d=---x-
2(1- p) Ac

ANALYSIS OF SIGNALIZED INTERSECTIONS WITH

PROBABILISTIC ARRIVALS
The assumption of nonuniform traffic arrival, although often more realistic than the uniform
(deterministic) arrival assumption, adds considerable complexity to the analysis of delay at
signalized intersections. The Webster formula for approach delay at a pretimed signalized
intersection is

2
X_ __
d' = d+ __ 0.65(_c )1I3X2+S(g/C)
2A.-(1- x) A.-2
where
d'is the average vehicle delay
d is the average vehicle delay computed by assuming DIDIl queuing
x is the ratio of approach arrivals to approach capacity (also referred to as the volume-to-capacity
ratio and equal to Ac/J.1g)
c is the cycle length
g is the effective green, and
A is the average vehicle arrival rate.

Allsop proposed that average vehicle delay, based on the assumption of random vehicle arrivals, be
computed as

d' ~[d+
=
10
x
2

2A(l-X)
]

Allocating effective green times to competing approaches in some optimal fashion has been a goal
of road engineers since traffic signals were first used. However, a drive on virtually any road with
signalized intersections shows that optimal signal timing has not yet been universally achieved.
The problem of optimal timing is complicated by a number of factors. For example, although the
distribution of traffic can be approximated by some arrival assumptions (deterministic or Poisson,
as explained in Chapter 1), to properly optimize signal timing the arrival pattern must be known
with certainty. Traffic sensors embedded near intersections have resulted in traffic signal systems
that respond to variations in traffic flow and thus have gone a long way to satisfy the traffic arrival
information needed.

To demonstrate one possible signal optimization strategy (putting aside the optimization issues just
discussed), assume that the sole objective of signal timing is to minimize vehicle delay and that
traffic can be represented by a simple DID/l queue

Traffic signals are costly to instal, with even some basic pretimed signals costing in excess of
$150,000. Following is a summary of the 11 warrants:

1 : Minimum Vehicle Volume

When traffic volume is the principal consideration for signal installation, this warrant gives
minimum volumes required on the major street and on the minor street.

2 : Interruption of Continuous Traffic

This warrant applies when traffic on the major street is so heavy that traffic on the minor street
suffers excessive delay or hazard when entering or crossing the major street.
3 : Minimum Pedestrian Volume
This warrant gives pedestrian volumes required to warrant signal installation. This warrant is used
when pedestrian safety and delay are a principal consideration for signal installation.

4 : School Crossing
Traffic signals are warranted at school crossings when traffic studies show that the frequency and
adequacy of the time headways between vehicles does not meet specified requirements in relation to
the number of children crossing.

5 : Progressive Movement
The idea is that installation of a traffic signal can maintain proper grouping of vehicles on a street
and thus result in a reduction in vehicle delay.

6 : Accident Experience
This warrant establishes criteria relating to the record of accident types and accident severities
needed to justify signal installation.

7 : System Warrants
In some instances, traffic signal installation may be warranted to encourage concentration and
progression of traffic flow on the traffic network.

8 : Combination of Warrants
In some cases, signals may be justified when no single warrant is met but where two or more of
warrants 1,2 and 3 are satisfied to 80% or more of their stated values.

9 : Four-Hour Volumes
This warrant applies when the traffic volume in each of any four hours in an average day exceeds
specified values.

10 : Peak-Hour Delay
The peak-hour delay warrant applies when traffic conditions are such that for one hour of the day
the minor-street traffic suffers undue delay when entering or crossing the major street.

11 : Peak-Hour Volume
This warrant is similar to warrant lOin that it applies when traffic conditions are such that for one
hour of the day the minor-street traffic suffers undue delay when entering or crossing the major
street.

The development of a traffic signal timing plan can be complex, particularly if the intersection has
multi-lane approaches and requires protected turning movements (eg. right-turn arrow, with
associated phasing). However, the timing plan analysis can be simplified by dealing with each
approach separately. As timing plans become more complex, they simply build upon these
fundamental principals.

There are eight steps in the development of a traffic signal timing plan:

• Select signal phasing

• Calculate equivalent straight-through passenger cars
• Select critical lane volumes
• Calculate change interval
• Calculate minimum cycle length
• Allocate green time
• Check pedestrian crossing time
• Prepare a signal indication summary

A cycle is the sum of individual phases. The most basic traffic signal cycle will be made up of two
phases as shown in Fig. 3. It is important, however, to note that each time the phase changes there
will be lost time associated with the traffic movement.

Each time a phase is added (such as the protected right-turn phase added in the three-phase
operation diagram shown in Fig. 2) the lost time increases. As a rule, a separate right-turning
phase should be considered for the signal timing plan when

1. The product of right-turning vehicles and opposing traffic volume exceeds 50,000 during
the peak hour on a two-lane highway, or 150,000 on a four-lane highway.

2. Two or more vehicles are still waiting to turn right at the end of the phase.

3. There are more than 50 vehicles turning right during the peak hour and the approach speeds
are greater than 72 km/h.

4. There are five or more accidents associated with turning movements during a 12-month
period.

CALCULATE EQUIVALENT STRAIGHT-THROUGH PASSENGER

CARS
Traffic arrives at an intersection in mixed-vehicle form with respect to vehicle type (ie. cars, buses,
and trucks) and turning movement (left-turning, right-turning and straight-through). Heavy
vehicles are slower to accelerate and decelerate and they have larger turning radii than passenger
cars. Also, all types of vehicles require additional time to execute both left and right turns.
Because turning movements and heavy vehicles require additional green time to complete their
manoeuvres, a method has been developed to account for this. Mixed-traffic volumes and non-
through movements are converted to equivalent straight -through passenger cars for the signal
timing analysis, using the adjustment factors in Table 1.
 Vehicle Type and Movement Adjustment Factor
. Passenger car (straight-through) 1.0
Heavy vehicle 1.5
Right-turning 1.6
Left-turning 1.4

TVJ 0 _ p~o..se
0'P~"""C:O-'/I~O'"

""11.-. .••."- ~ - ,?~e.

°P~""o.'Tl·C ""

l _
After the equivalent straight-through passenger car volumes have been calculated, the critical-lane
volumes for each approach can be determined. The right-turn volumes are considered a separate
movement if a right-turn bay is provided. The equivalent straight-through and left-turn movements
are combined and distributed over the number of approach lanes. The distribution of traffic over
the lanes is best done by a traffic survey that determines actual lane volumes by observing vehicles.
If such a survey is not available, the amount of traffic in the most heavily used lane in the approach
(ie. critical lane volume) can be approximated as the total approach volume multiplied by a critical-
lane factor. For through lanes or shared through and turning lanes, the factor is 0.525 and 0.367
for two- and three-lane approaches, respectively. For two-lane exclusive right-turn approaches, the
factor is 0.515, and for two-lane exclusive left-turn approaches, the factor is 0.565. For through
or shared-lane approaches with more than three lanes, and exclusive-turn approaches with more
than two lanes, a traffic survey is recommended.

This interval provides for clearance of the intersection before conflicting traffic movements are
given a signal indication that allows them to enter the intersection (expressed in seconds). In the
past, a yellow indication was used to warn the driver of a signal change and provide clearance time.
Today, however, there is routine red-indication abuse and frequent running of red indications after
the yellow time. As a result, the all-red indication has become commonly used.

Accepted formulas for calculating yellow and all-red times are

V
IT=t +----
p 2a+2grG

AR= w+Z
V

where
IT is the yellow time (usually rounded to the nearest 0.5 second)
$>, is the driver perception/reaction time taken as 1.0 second
v is the speed of the vehicle in m/s
a is the deceleration rate for the vehicle taken as 3.05 m/s2
G is the percent grade divided by 100
gr is the acceleration due to gravity (ie. 9.807 m/s2)
AR is the all-red time
w is the width of the cross street in metres, and
Z is the length of the vehicle (taken as a conservative 6 m).

A dilemma zone is created for a driver if a safe stop before the intersection cannot be accomplished
but continuing through the intersection at a constant speed (ie. without accelerating) will result in a
red indication. Figure 3 shows a sketch of the dilemma zone. Referring to this figure, suppose a
vehicle travelling at a constant speed requires distance Xs to stop. If the vehicle is closer to the
intersection than distance dd' then it can enter before the all-red indication. If the vehicle is in the
shaded area (ie. Xs - dd from the intersection) when the yellow light is displayed, the driver is in the
dilemma zone and can neither stop in time nor continue through the intersection at a constant speed
without passing through a red indication. In order to avoid a dilemma zone the total change interval
(yellow plus all-red times) should always be equal to or greater than the sum of Eqns 10 and 11.

-----,.-
Appro(i-e-t.'j '"C.,.v-PI'~

The cycle length is simply the summation of the individual phases. In practice, cycle lengths are
kept as short as possible, typically within the 40- to 60-second range. However, complex
intersections with five or more phases can have cycle lengths of 120 seconds or more. It should be
noted that cycle lengths greater than 120 seconds are unusual and should be used only in
exceptional circumstances.

A practical equation for the calculation of optimum cycle length was developed by Webster (1958),
who assumed that vehicles arrive at an intersection on a random basis. He developed a practical
expression for cycle length that sought to minimize vehicle delay. Webster's optimum cycle-length
formula is

1.5LT+5
c=-----
1.0- .k.../=l
~~ Yr·

where
c is the cycle length (usually rounded up to the nearest 5 second increment)
LT is the lost time approximated as the total yellow and all-red times per cycle, and
Yi is the ratio of the critical-lane volume to the per-lane saturation flow for signal phase i.
After a cycle length has been calculated, the next step in the traffic signal timing process is to
determine how much green time should be allocated to each phase. The cycle length is the sum of
all green times plus all lost time. In other words, the cycle length minus the sum of yellow and all-
red times (ie. the approximated lost time) will leave the total green time available for all phases.
This total green time is diYided among the phases that comprise the cycle. The green time is usually
distributed to each phase in proportion to the ratio of critical-lane volume of the phase to total
critical-lane traffic (su~ation ofthe critical-lane volumes of all phases).

In urban areas and other locations where pedestrians are present, the signal timing plan should be
checked for its ability to provide adequate pedestrian crossing time.

To calculate the minimum green time required for pedestrian movements, the time required for a
pedestrian to walk across the street must be determined. For signal timing purposes, it is assumed
that pedestrians walk at a rate of 1.2 m/s. In areas where there are a large number of elderly
pedestrians, a walking speed of 0.9 m/s is appropriate. In addition to the crossing time, a
pedestrian reaction time of 7.0 seconds is added to the required green time. In equation form the
minimum pedestrian green time is

w
PGT = 7 +-- - YT - AR
PWS
where
PGT is the pedestrian green time in seconds
w is the width of the street in metres
PWS is the pedestrian-walking speed in m/s
YT is the yellow time in seconds, and
AR is the all-red time in seconds.

After the pedestrian green time is evaluated, the calculation of interval times for each phase is
complete, and a signal indication summary can be prepared. The signal indication summary shows
the phases for each movement and, at this time the phases and cycle length can be finalized and
conflict checks can be made. The signal indication summary for the Vine and Maple streets
intersection is given in Table 2. Note that the red times (RT) presented in this table are simply
equal to the summation of green time (GT), yellow time (Y1), and all-red time (AR) for the
particular phase.
 Table 2. Signal-Indication Summary for the Intersection of Vine and Maple Streets
(Total cycle length = 70 seconds).

Street Phase 1 Phase 2 Phase 3

Vine Street GT= 20 s
(through) YT = 3.5 s RT= 20.5 s RT= 24 s
AR = 2.0 s
Vine Street GT= 15 s
(right turn) RT= 25.5 s YT = 3.5 s RT= 24 s
AR = 2.0 s
Maple Street GT= 19 s
(through) RT= 25.5 s RT= 20.5 s YT= 3.0 s
AR = 2.0 s

EXAMPLE! (Signalised intersection with DID!! queuing)

An approach at a pretimed signalised intersection has a saturation flow of 2400 veh/h and is
allocated 24 seconds of effective green in an 80-second signal cycle. If the flow at the approach is
500 veh/h, provide an analysis of the intersection assuming D/D/1 queuing.

A = 500 veh/h
3600 s/h

2400 veh/h
J.! = 3600 s/h

0.139 veh/s
p=---
0.667 veh/s
Checking to make certain that capacity exceeds arrivals, we find that the capacity (J1g) is 16
veh/cycle (0.667 x 24) which is greater than (permitting fractions of vehicles for the sake of clarity)
the 11.12 arrivals (Ae = 0.139 x 80). Therefore, Eqns 1 to 7 are valid. By definition r = c - g = 80
- 24 = 56 s. Thus for

0.208(56)
t =----
o (1- 0.208)

= 14.71 s

p = 56+14.71
q 0.884

p = 14.71
s 0.208(80)

= 0.884

Qm = 0.139(56)

= 7.78

D = 0.139(56)2
t 2(1- 0.208)

562
d=-----
2(80)(1- 0.208)
7. Maximum delay of any vehicle (Eqn 7),

dm =r

EXAMPLE 2 (DID!! - Approach arrivals exceed capacity)

An approach to a pretimed signalized intersection has a saturation flow of 1700 veh/h. The signal's
cycle length is 60 seconds and the approach's effective red is 40 seconds. During three consecutive
cycles, 15,8 and 4 vehicles arrive. Determine the total vehicle delay over the three cycles assuming
DID/l queuing.

Solution:

1700 veh/h
J.l = 3600 s/h

During the first cycle, the number of vehicles that will depart from the signal is (permitting fractions
for the sake of clarity)

Therefore, 5.56 vehicles (15 - 9.44) will not be able to pass through the intersection on the first
cycle even though they arrive during the first cycle. At the end of the second cycle, 23 vehicles (15
+ 8) will have arrived but only 18.88 (2 J.lg) will have departed, leaving 4.12 vehicles waiting at the
beginning of the third cycle. At the end of the third cycle, a total of 27 vehicles will have arrived
and as many as 28.32 (3 J.lg) could have departed, so the queue that formed during the first cycle
will dissipate at some time during the third cycle. This process is shown graphically in Fig. 4.
From this figure, the total vehicle delay of the first cycle (the area between arrival and departure
curves) is

D1 = .!.(60)(15) - .!.(20)(9.44)
2 2

D2 = .!.(60)(15 + 23) - (40)(9.44 + 18.8)

2
To determine the delay in the third cycle, it is necessary first to know exactly when, in this cycle,
the queue dissipates. The time to queue dissipation after the start of the effective green, to' is (with
\ being the arrival rate during the third cycle, and n3 being the number of vehicles in the queue at
the start of the third cycle)

A = 4 veh
3 60s
100

~o

%0 -

'Q?
.) 7°
~
~
),. 60

"'-
Q
'So

L 40
~•
t
~ ~o
<.
~o

Cy d£' 2.. < c."t (I-e 3

,'2..0

which gives to = 16.8 seconds. Thus the queue will clear 56.8 seconds (40 + 16.8) after the start
of the third cycle, at which time a total of 26.8 vehicles (0.067 x 56.8 + 15 + 8) will have arrived
and departed from the intersection. The vehicle delay for the third cycle is
 D3 = 2-(56.8)(23 + 26.8) - (40)(18.88) - 2-(16.8)(18.88 + 26.8)
2 2

EXAMPLE 3
(Intersection arrivals and/or departures are deterministic but varying
in a fashion similar to that shown earlier in this chapter)
The saturation flow of an approach to a pretimed signal is 12,000 veh/h. The signal has a 60-
second cycle with 20 seconds of effective red allocated to the approach. At the beginning of an
effective red (with no vehicles remaining in the queue from a previous cycle) vehicles start arriving
=
at a rate A(t) 0.88 + 0.02t + 0.00057r (where A(t) is in vehicles per second and t is the number
of seconds from the beginning of the cycle). Thirty seconds into the cycle the arrival rate remains
constant at its 30-second level and stays at that rate until the end of the cycle. What is the total
vehicle delay over the cycle (in vehicle-seconds) assuming DID/l queuing?

Solution:
Vehicle arrivals for the first 30 seconds (again allowing fractions of vehicles for the sake of clarity)
are

r30
= JoO.88 + 0.02t + 0.00057t3

30
= 0.88t + 0.01t2 + 0.00019t3
o

= 40.53 veh

The arrival rate after 30 seconds is 1.993 veh/s [0.88+0.02(30)+0.0000057(30)2] and is no longer
time varying. During the effective green, the departure rate is 3.333 veh/s (12,000/3600). To
determine when the queue will dissipate, let t' be the time after 30 seconds (ie. the time after which
the arrival rate is no longer time varying), so,

which gives t' = 5.37 seconds. Thus the queue clears at 35.37 seconds after the beginning of the
cycle. For delay, the area under the arrival curve
 ~ 1
=
l 0.88t + 0.01t
o
2
+ 0.00019t3 + -{ 40.53 + [1.993(5.37) + 40.53]}(5.37)
2

30
= 0.44t2 + 0.0033t2 + 0.0000475t2 + 246.38
o

1
= -(51.23 x 15.37)
2

EXAMPLE 4 (Probabilities arrivals)

Compute the average approach per cycle using Webster's and Allsop's formulas given the
conditions described in Example 1.

Solution:
As computed in Example 1, the delay, assuming uniform arrivals (D/D1l queuing), d is 24.75
s/veh. The ratio of approach arrivals to approach capacity is
A.c
X=-
f.Lg

0.139(80)
=----
0.667(24)

= 0.695

2
d' = 24.75 + 0.695 0.65( 80 )1/30.6952 + 5(24/80)
2(0.139)(1- 0.695) 0.1392

= 24.72 + 5.697 - 2.921

 2
d'=~[24.75+ 0.695 ]
10 2(0.139)(1- 0.695)

EXAMPLE 5 (Optimal traffic signal timing)

A pretimed signal controls a four-way intersection with no turning permitted and zero lost time.
The eastbound (eb) and westbound (wb) traffic volumes are 700 and 800 veh/h, respectively, and
both movements share the same effective green and effective red portions of the cycle. The
northbound (nb) and soundbound (sb) directions also share cycle times with volumes of 400 and
250 veh/h, respectively. If the saturation flow of all approaches is 1800 veh/h, the cycle length is
60 seconds, and DIDIl queuing applies, determine the effective red and effective green times that
must be allocated to each directional combination (ie. north-south, east-west) to minimize total
vehicle delay, and compute the delay per cycle.

A = 700 veh/h = 0.194vehls

eb 3600 s/h

A = 800 veh/h = 0.222 vehls

wb 3600s/h

400veh/h
A.nb = ----= 0.111 vehls
3600s/h

A = 250 veh/h = 0.069 vehls

sb 3600s/h

f.1 = 1800 veh/h = 0.5 vehls

3600s/h

Because the departure rate is the same for all approaches, the traffic intensities are Peb = 0.388, Pwb
= 0.444, Pnb = 0.222, and Psb = 0.138. If it is assumed that approach capacity exceeds approach
arrivals for all approaches (an assumption that will be tested later), the total vehicle delay at the
intersection (using Eqn 5) is
The problem states that east and west effective reds are equal and north and south effective reds are
equal. So let rew be the effective red of eastbound and westbound directions (ie. rew = reb = rwb) and
let rns be the effective red of northbound and southbound directions (ie. rns = rnb = rsb). So, by
definition, with a 60-second signal cycle, rns = 60 - rew• Substituting this into the total delay
expression gives

Dr = 0.1585rew + 0.1996rew + 0.07115 (60 - rewl + 0.04 (60 - rew)2

= 0.46925rew - 13.338rew + 400.14

dDr = 0.9357rew -13.338 =0

drew

This gives r ew = 14.2 s (gew = 45.8 s), and rns = 45.8 s (gns = 14.2 s). For total delay,

Dr = 0.1585 (14.2)2 + 0.1996 (14.2)2 + 0.07115 (45.8)2 + 0.04 (45.8)2

= 305.36 veh/s

Finally, a check of the earlier assumption that approach capacity exceeds approach arrivals for all
approaches must be undertaken. In the 60-second cycle, eastbound and westbound arrivals are
11.64 (0.194 x 60) and 13.32 (0.222 x 60), respectively. With 45.8 seconds of effective green,
the capacity for both approaches is 22.9 (0.5 x 45.8), so the assumption is satisfied. The
northbound and southbound arrivals are 6.67 (0.111 x 60) and 4.14 (0.069 x 60), respectively,
and, with 14.2 seconds of effective green, the capacity for both approaches is 7.1 (0.5 x 14.2).
Again, the assumption is satisfied and the method used in this example is valid.

EXAMPLE 6 (Signal phasing selection)

The intersection shown in Fig. 5 does not satisfy right-turn phase requirements for waiting
vehicles, approach speeds greater than 72 krn/h, or turning-movement accidents. Determine if the
product of right-turning vehicles and opposing vehicles suggests the use of a right-turn phase.
(Note in Fig. 5 that the traffic volume numbers in parentheses are heavy vehicles, such as buses
and large trucks).

There are 180 northbound vehicles that turn right during the peak hour. The product of the
northbound right-turning vehicles and the opposing traffic (left-turn and straight-through vehicles)
is 126,000 (180 x [648 + 52]). Because the product is greater than 100,000 (the requirement for a
four-lane highway), a right-turn phase is suggested. The other approaches do not require a right-
turn phase using this criterion because the products for all other approaches are less than 100,000

100
(6)

~ >- 5~c
(So)
4~
E:.o (4)
Cb) S68
(bb)
550
(so) (40)

c q) \"
(be
(2.0

t>-eo...K - h ••••.••...t:,... ••.•~LC \/blu.-.....s

(h.e..--.J7 y~h~(..\C' "oluvv..c.S
\...... Yo..Ye'tl."escs)
M o-pl e S f,..-c-e.\
cd.. ':> p1. •...J
A P\>T o•••.
(40 ~"'" , h)

I Mqple sr.•.
~.••..
"T
APPye.o-(.. k s~c.••.J
I (40 '<::--""-)
I
I
I
I
Figure 5. Intersection geometry and peak traffic volumes for example problems.

Eastbound right-turnJopposing traffic product is 33,264, westbound right-turnJopposing traffic

product is 52,480, southbound right-turnJopposing traffic product is 64,900. Therefore, a three-
phase traffic signal plan should be used.
 Example 6 (Change interval)
Determine the yellow and all-red times for vehicles traveling on Vine and Maple Streets as shown in
Fig. 4.

IT = 1.0+ (55 x 1(00/3600)

2(3.05)

AR= 18.3+6
55 x 1000/3600

= 1.6 s (rounded to the nearest 0.1)

For the Maple Street phasing (applying Eqn 11),

IT = 1.0 + (40 x 1000/3600)

2(3.05)

AR = 15.4+6
40 x 1000/3600

= 1.9 s (rounded to the nearest 0.1)

The yellow time for Maple Street (2.8 s) is rounded to 3.0 because many traffic signal controllers
handle timing only to the nearest 0.5 s. The yellow time for the right-turn phase on Vine Street is
calculated by using the same values as the Vine Street phasing. Thus a yellow time of 3.5 s is
obtained.

For the all-red times, Maple's 1.9 s and Vine's 1.6 s are both rounded up to 2.0 s (the nearest 0.5 s
interval, which can be handled by most signal controllers). The all-red time on Vine Street for the
right -turn phase will be the same as for the through movements. A turning vehicle will travel a
circular path to clear the approaching lanes. The travel distance for the turning vehicle will be
approximately 10 m, which is shorter than the width of Maple Street; however, because a turning
maneuver requires more time, the all-red time is also set as a conservative 2.0 s.

EXAMPLE 7 (Equivalent Straight-Through Passenger Cars)

Calculate the equivalent straight-through passenger cars for the northbound, southbound,
eastbound and westbound traffic streams shown in Fig. 4. (Note in Fig. 4 that the traffic volume
numbers in parentheses are heavy vehicles, such as buses and large trucks).
Using Table 1, calculations for the total northbound equivalent straight-through passenger cars for
each movement are (rounding up to the nearest 1.0 vehicle)

Right turns = 160 (1.6) + 20 (1.6) (1.5) = 304 veh/h

Left turns = 50 (1.4) + 9 (1.4) (1.5) = 89 veh/h

= 550 (1.0) + 40 (1.0) (1.5) = 610 veh/h

This summation gives 1003 veh/h total equivalent straight-through passenger cars. The equivalent
straight -through passenger car traffic for the other directions is calculated in the same manner. The
results give 923 veh/h, 839 veh/h and 707 veh/h for southbound, eastbound and westbound,
respectively.

EXAMPLE 8 (Select critical-lane volumes)

A traffic survey shows that the through volumes of all approaches are distributed equally among the

street intersection.

______ >_ J£.4 I b~

Ii
through lanes. Using the equivalent straight-through passenger cars calculated in Example 7,
determine the critical-lane volumes for east-west and north-south approaches to the Vine and Maple

-----":\-- •••3 b40 J32.~~ 3r+-

~ 1\ \ 2 ~8~-------

351f304 ~~<

In Example 7, the northbound right-turning equivalent straight-through passenger car volume was
determined to be 304 vehicles. The straight-through equivalent was 610 vehicles with 89 left-turn
equivalents. Combining the left-turn and straight-through movements gives 699 vehicles (610 +
89). As shown by the traffic survey (and stated in the example), the straight-through traffic is
distributed equally over the lanes as 350 vehicles per lane (699/2, and rounded up to the nearest
vehicles for practical reasons). The other approaches are handled in the same way with the straight-
through passenger cars distributed as shown in Fig. 6.

The critical-lane volumes are then determined for each phase of the traffic-signal timing plan. As
discussed earlier in example 6, a three-phase timing plan is to be used. Consequently, critical lanes
must be selected for the north-south movement, the east-west movement, and the right-turning
phase. Observing the east-west traffic volumes shown in Fig. 5, it can be seen that the eastbound
movement of 364 vehicles per hour per lane is greater than the 288 vehicles per hour per lane for
the westbound movement. Therefore, the critical-lane volume for the east-west phase is 364
vehicles per hour per lane. Likewise, the 377 vehicles per lane for the southbound approach
represents the critical volumes for the north-south phase. The right-turn phase will use 304
vehicles per hour as the critical-lane volume.

EXAMPLE 9 (Minimum cycle length)

Using Webster's delay formula, calculate the optimum cycle length using the information provided
in the preceding examples for the Maple Street-Vine Street intersection.

In Example 9, it was determined that the yellow times for Vine Street and Maple Street phases are
3.5 seconds for the Vine Street phase, 3.5 seconds for the Vine Street right-turn phase, and 3.0
seconds for the Maple Street phase. The summation of these yellow times gives a total yellow time
of 10 seconds. Also, in Example 9, it was determined that the all-red time for each phase was 2.0
seconds, giving a total all-red time of 6 seconds. Thus the total lost time is approximated as 16
seconds (10 + 6). From Example 8, it was determined that the critical lane volumes are 377 veh/h
per lane for the Vine Street phase, 304 veh/h for the Vine Street right-turn phase, and 364 veh/h per
lane for the Maple Street phase. Using these values in Eqn 12 gives

1.5(16)+ 5

c= 1.0-[(~)+(~)+(~)]
1800 1800 1800

!EXAMPLE 10 (Allocate Green Time)

Determine the green time allocations for the 70-second cycle length found in Example 10.

Solution:

Because there are 16 seconds oflost time, the total available green time is 54 seconds (70 - 16).
Summing the critical-lane volumes gives a total critical-lane volume of 1045 veh/h (377 + 304 +
364). Using the ratio of critical-lane volumes to total critical-lane volume gives 19.5 seconds
[(377/1045) x 54] of green time (GT) allocated to the Vine Street phase, 15.7 seconds [(304/1045)
x 54] for the Vine Street left-turn phase, and 18.8 seconds [(36411045) x 54] for the Maple Street
phase. Rounding values (which always involves some compromise), we allocate 20 seconds, 15
seconds, and 19 seconds to the Vine Street phase, Vinet Street left-turn phase, and Maple Street
phase, respectively.

IEXAMPLE 11 (Pedestrian Crossing Time)

Determine the minimum amount of pedestrian green time required for the intersection of Vine and
Maple streets. (Assume pedestrian walking speed is 1.2 m/s).

As shown in Fig. 4, a pedestrian who crosses Maple Street will cross while Vine Street has a green
interval. The minimum green time needed on Vine Street is (using Eqn 13),

PGT = 7 + 18.3 - 3.5 - 2.0

1.2

= 16.75 s

From Example 11, Vine Street is assigned 20 seconds of green time, thus there is enough green
time for pedestrians to cross the street. The minimum green time needed on Maple Street (using
Eqn 13) is

PGT=7+ 15.4 -3.0-2.0

1.2

= 14.83 s

From Example 11, Maple Street is assigned 19 seconds of green time, so the minimum pedestrian
green time is again met.
Faculty of Engineering & Computing
Department of Civil Engineering

Transportation Engineering 461

Chapter 9

Unsignalised Intersections

Professor Hamid Nikraz

At unsignalised intersections carrying relatively low volumes of cross-traffic and
turning traffic it is unnecessary to carry out capacity analyses. Rather, the important
considerations are safety and traffic operations. Table 1 sets out details of intersection
traffic volumes below which capacity is not a consideration, although separate lanes
for left - and right - turning vehicles may sometimes be added for safety reasons.

Maximum design hour volumes

Road type (two way, vehicles per hour)

Two-lane through road 400 500 650

Cross-road 250 200 100
Four-lane through road 1000 1500 2000
Cross-road 100 50 25

When the volumes in table 1 are exceeded, or when channelisation is being

considered, the capacity of the intersection in the unsignalised state should be
analysed. At an unsignalled intersection, the major traffic streatn will usually have
priority and minor cross-road traffic will be required to give way and to select
appropriate gaps in the major stream. The fIrst vehicle in the queue in the minor
stream requires a critical acceptance gap of T seconds in its major stream, while
successive vehicles require a shorter follow- up headway of T seconds in its major
stream, while successive vehicles require a shorter follow-up of To seconds. Typical
critical gaps and follow up headways for various conditions are shown in Table 2.

The capacity of an unsignalised intersection may be expressed in terms of the volume

of traffic that can enter the major stream from a minor stream. This volume is called
the absorption capacity (C) of the major stream. When a minor stream volume is at
or close to the absorption capacity of the major stream, long queues develop and long
delays result. For this reason, a practical absorption capacity, ( Cp), which is taken as
80 percent of the absorption capacity, is often adopted for design and operational
purposes.

Having selected the appropriate' critical gap and follow-up headway for table 2, Figure
1 can be used to give an estimate of the practical absorption capacity (ie.the amount of
cross-road traffic that can cross or merge with the major stream). Figure 1 is based on
the assumption that the arrival of vehicles in the conflicting traffic streams is
randomly distributed. If this is not the case, it will probably underestimate the
practical absorption capacity.
 Critical acceptance Follow-up headway,
gap, T seconds To seconds

Crossing manoeuvres
2 lanes 2 way
4 lanes 2 way
2 lanes I way
3 lanes 1 way
4 lanes I way
Simple merging manoeuvre
Right-turning manoeuvre
across a I-lane flow 4-5 2-3
across a 2-lane flow 5 3
across a 3-lane flow 6 4

The average delay to a minor stream vehicles entering the major stream is then given
by the following

Where
Wm= average delay to minor stream vehicles, per vehicle, in secs
qp= major stream volume in veh/sec
qm=minor stream volume per lane in veh/sec
ta= critical acceptance gap in veh/sec
tr= follow up headway in veh/sec

Figures 2 a and 2b give solutions to the equation for ta=4 seconds, tF2 seconds and
for ta =5 seconds, and tr =3 seconds respectively.
1500

1000
900
aoo
~
ci. 700
.i
~ 600

~ 500
~
u
0
400
~
~
.Q
300
<
~
~
...
0: 200

Figure 1 Unsignalised Intersections - Practical absorption Capacity (source:

NAASRA, 1988)
 100
go
80
~
- ." .".,.
-
~ ..
- - ;.; "
.""':-
10
80
~I ~
'" '" I,;::
"
~ ~ ~r.
50
/I)
"/ 1/ 1/ "I ,,/1//
~ ::Sf / "I If

I I I
"/ j
II

:-. ••
"::
:/ /
••••
,.
/ /
II
1/
7//
~ ~/ / I
Cl
./ /
i3 3Jl
I
"::

f5:.. z~~ ~ /0
;/
"'
II'

::. ID
~ / ~
••••
I

•
J

5
.
7
'/
II
I

<

,
ta = 5 see
ta = 4 see tf = 3 see
,
tf = 2 see 1
100

Figure 2 Average Delay to Minor Stream Vehicles at an Unsignalised Intersection

(source: NAASRA, 1988)

At an unsignalised intersection, the simplest measure of the quality of traffic service

or traffic performance is the ratio of the actual minor stream volume to the absorption
capacity of the major stream. Thus, the quality of the traffic service t
A c 7 liP-I !'1~no r- STy~""'-4">-~ Vo /1./ ,'''of?

ob SO I'P 7, ;>r~ C"" pc.- ~T'j

_ I+c7t/",,1 M I n.,.- 50 T,'€-".i'Y! Vo1vou;:
;.2..S"" P rIA-Vii' t: .•, J D bSor P It'D "'I (4PC;,';;Ty
Alternatively, it is possible to calculate mean delays and queue lengths for minor
stream vehicles.

The storage requirements , such as for example the storage required for an
unsignalised right turn movement may be calculated as follows:
• Determine minor stream movement lane volume qrn.
• Determine the maximum service at rate qs, for the minor stream movement. Usually
this is taken as the absorption capacity C for the movement.
• Calculate the utilisation rate p (ratio of the arrival rate to the service rate)ie, p=qrn/qs.
• Decide on the probability that the mean queue length will not be exceeded. The 95
or 98 percent level is typically selected.
• Use figure 3 to determine the maximum queue length appropriate to the selected
probability level.
• Allow six metres for each passenger car and 12m for each heavy vehicle in the
maximum queue length.
 I
I
@
cr:
:5
8

I :
a 7
w
cr:
(J) 6
I
w

~
I I

(J)
w 98%
<.? 4
-<
cr:
:2
I 95.%
I

(J)
3

I :
I :
I

Figure 3 Unsignalised Intersection Vehicle Storage Requirements (source: NAASRA,

1988)

A roundabout is a particular form ofunsignalised intersection. Australian practice for

determining the capacity of roundabouts is based on gap acceptance theory because
operational experience has indicated that, for the sizes of roundabouts normally used,
traffic entering the roundabout gives way to, and accepts gaps in, the circulating
traffic stream. For capacity analysis, the circulating flow on the roundabout is taken as
the major flow, and the entering flow is taken as the minor flow, as in figure 4, where
Ile is the number of entry lanes. Typical critical acceptance gaps and follow up
headways at roundabouts are shown in table 3.

Figure 5 indicates the flow per entry lane on any given approach that is possible (i.e.
the capacity per entry lane) for various circulating flows. The required number of
entry lanes can then be determined by dividing the actual or estimated flow by the
capacity per entry lane.

The quality of traffic flow at an entrance to a roundabout may be taken as the actual
traffic flow per lane at the entrance divided by the capacity per entry lane. This value
should preferably not exceed 0.8, although this cannot always be achieved.

Number of lanes on Critical acceptance Follow-up headway,

circulating roadway gap, T seconds To seconds

Single-lane
Multi-lane
 ===} Circulating flow = qp

Total entry flow = n x q

Where n. = n~mbe~ of entry lanes
=
qm entry flow per lane

Figure 4 Major and Minor Flows at a Roundabout Figure 5 Roundabout Capacity

(source: NAASRA, 1988)

Figure 6 shows the relationship between the circulating flow (Qp vehicles per hour)
and the capacity per entry lane (Ce vehicles per hour) for single lane and multi lane
circulating roadways. When using Figure 6, the following should be noted:
• When the proportion of trucks in the traffic stream exceeds 5 percent vehicular
flows should be converted to passenger car units. For this purpose, a single unit truck
should be taken as 2 passenger car units, and on articulated truck as 3 passenger car
units .
• Separate calculations are required for each entrance to the roundabout.
The degree of saturation at an entrance to a roundabout is given by the actual flow per
lane at the entrance divided by the capacity per lane. Desirably the degree of
saturation at the various entrances to the roundabout should not exceed 0.8, although
this cannot always be achieved.
11300

I I I
'l
, I
I
i
cJ
120a
l\\- I
I I
w
Z tOOO
I I
:5 I M"'IlI-lanec~ ••1lIIin91'kl'" I
I I I'-.. I-
~ soo
I
~ I
ffi soo
,.
0. Sin,...lanecon:;ulallnQrIow'
I
>- .00
u
iI' f-
'" ,00

"

Capacity per Entry Lane:

c .3600 q,(l-q, <)exp(-'l.,('.-<))

• 1- exp(-q, ',l
where:
q~ .•• circulating (priority) flow in vdl/so:::
t. :Ill Critical Accept3Ilce Gap ill seconds (from Table 8.3)
'J "" raUow-up headway o( the entry Oow in soconds (from Table 8.3)
't' '"" minimum headway (or circulating traffic in seconds (from Table 8.3)

There are two components of the delay experienced at roundabouts, namely

• Queuing delays to vehicles waiting to accept a gap in the circulating traffic.
• Geometric delays to vehicles slowing down to proceed through the roundabout, and
then accelerating back to normal speed.
 The average queuing delay to vehicles entering roundabouts with single lane and
multi-lane circulating road ways can be read directly form figure 7.

The average geometric delay to vehicles passing through roundabouts depends on

whether they must stop or not. The proportion of entering vehicles that stop at
roundabouts can be read form Figure 8.

The geometric delay to vehicles that stop at roundabouts can be read from Table 4 and
the geometric delay to those that do not stop can be read from Table 5.

The distance D in tables 4 and 5, which is the distance travelled at the roundabout
negotiating speed, is defmed as in figure 9, depending on the movement under
consideration.

50 50

.0 .0
I I
30
I I
30

20 20
I I I I 11/ /

U
~
e
·u
~ / I I 1/ li/ /1/ /
5:
:5 10
~
;;: 10
/ / 1/ '1//
W
0
9 -'
w
9
(9 8 0 8
Z (9
u:; Z
::l W
w ::J
::l w
0 ::l
W a
~ w
a:
w
~
a:
;:;: w
~

4 see ta '" 4 see

2 see tf '" 2 see
2 see 't '" 0 see
I I

(a) (b)
Figure 7 Average Queuing Delay to Vehicles Entering Roundabout with Single and
Multi-lane Circulating Roadways (source NAASRA, 1988)
 Figure 8 Proportion of Entering Figure 9 Defmition of the Negotiation Distance
at Roundabouts at Roundabouts
(source: NAASRA, 1988)

Approach Speed

Negotiation 40kmih 60 kmih 80 kmih l00kmih

Speed through D(m) D (m) D (m) D (m)
Roundabouts
(kmih) 20 60 100 140 180 20 60 100 140 180 20 60 100 140 180 20 60 100 140 180
. Delay (seconds)
15 10 19 13 23 17 26 20 30
20 8 15 22 11 18 26 15 22 29 18 25 33
25 7 12 17 10 15 21 13 19 25 17 22 28
30 7 9 13 18 10 13 18 22 13 17 21 26 17 20 25 30
35 7 7 10 14 18 10 10 15 19 23 13 14 19 _ 23 27 17 18 22 26 30
~.," .......
40 10 10 12 , 15 19 13 13 16 19 23 17 17 20 23 27
45 10 10 10 12 15 13 13 13 16 19 17 17 17 20 24
50 . 10 10 10 10 10 13 13 13 13 16 17 17 17 17 20

Approach Speed

Negotiation
Speed through
Roundabouts
(km/h) 20
40kmih
D (m)

60 100 140 180

60 km/h
D(m)

20 60 100 140 180

I 20
80 km/h
D (m)

60 100 140 180

100 kmih
D (m)

20 60 100 140 180

I Delay (seconds)
15 7 17 11 20 14 24 1827
20 4 11 19 8 15 22 11 19 26 1522 29
I 25 2 7 13 5 11 17 9 15 20 1218 24
30 1 4 8 13 4 8 13 17 7 11 16 21 1015 20 25
35 0 0 4 8 12 3 4 9 13 17 6 8 13 17 21 9 12 16 20 25
40 2 2 5 8 12 5 5 9 13 16 8 9 13 17 20
45 1 1 1 4 7 4 4 5 9 12 7 7 10 13 16
50 1 1 1 1 2 3 3 3 3 7 6 6 6 12
 650
A one way street carry 65V'vehlh crosses a two way street carrying 800 vehlh in the
same period. There are two traffic lanes on the'$L way street and traffic on it has right
of the way. The intersection is unsignalised. Determine the practical absorption
capacity for the two way stream, and the number of traffic lanes required on a one way
street approach.

The circulating flow on a roundabout immediately upstream of a singly lane entrance

to it is 1000 vehlh. Ifthe entering volume is 200 vehlh what is the degree of saturation
at the entrance.

A one way street carrying 600 vehicles per hour crosses a two way collector street
carrying 800 vehicles per hour in the same period. Assume there are two traffic lanes
on the collector street which has the right of way. The intersection is unsignalised.
Determine:
• number of lanes required on the one way street.
• average delay to one way street traffic.
• average queue length on the one way street approach.
• queue length on the one way street approach that is unlikely to be exceeded 98% of
the time.
Problem 4
For the unsignalisation intersection shown below, operating under give way to right
rule, determine the design queue length for the right turn movement 3.
For the intersection with traffic movements shown below, determine storage length
for unsignalised right turning movements.
For the intersection with traffic movements shown below, it is proposed to install a
single lane roundabout. The proportion of trucks is negligible, and the area is flat.
Check the degree of saturation and extent of delays at the proposed roundabout.
DIVERGING

I"'" Elemental -I

~
RIGHT
L
LEFT
-< MUTUAL
~
MULTIPLE

MERGING
I~ Elemental
"'1

7
RIGHT
•.. ~

LEFT
.. >-- MUTUAL
~
MULTIPLE

CROSSING
1- Elemental -- ,. 1

-t- ~ 7~ ~
DIRECT OPPOSE D OBLI QUE MULTIPLE

WEAVING

~
ELEMENTAL MU LTIPLE
 L

~L
THROUGH FLOW CROSSING
CONFLICT

II
 1~
80 km Ih

(f) a b = RELATIVE SPEED BETWEEN

VEHICLES A~B

RELATiVE SPEE 0
VEHiCLES A' C

~---------~
---------------l!OF-

~--------
-----=======J)