Design Manual Ver 14设计守则

THE REPUBLIC OF UGANDA
MINISTRY OF WORKS, HOUSING AND COMMUNICATIONS
ROAD DEVELOPMENT PROGRAM PHASE 2 PROJECT
CONSULTANCY SERVICES FOR THE PRODUCTION OF

ENGINEERING MANUALS AND SPECIFICATIONS AS WELL
AS PROVISION OF INSTITUTIONAL SUPPORT
Contract RDP/GN/S008
Road Design Manual

(Road safety revision)
(Draft)
Client: Lead Consultant:

Ministry of Works, Housing & TRL Limited
Communications Old Wokingham Road
PO Box 10 Crowthorne
ENTEBBE Berkshire
RG45 6AU
30 May 2004 United Kingdom
Table of Contents
0 INTRODUCTION 9
0.1 Main Requirements 9
0.2 Safety Considerations 9
0.3 Environmental considerations 11
0.4 Economic considerations 12
1 GENERAL (NO REVISIONS) 14
2 ROAD SYSTEM (NO REVISIONS) 14
3 DESIGN STANDARD 15
3.1 Design Procedure 15
3.2 Basic Prerequisites 15
3.3 Design Criteria 19
3.4 Main Design Features 33
4 CROSS-SECTION 35
4.1 Cross-section elements 35
4.2 Headroom and Lateral Clearance 36
4.3 Selection of Cross-section 37
4.4 Detailed Design of Cross-section 41
4.5 Roadside Area Design 46
5 ALIGNMENT 51
5.1 General 51
5.2 Alignment choice and terrain adaptation 51
5.3 Sight distances 62
5.4 Horizontal alignment 68
5.5 Vertical alignment 82
5.6 Climbing and overtaking lanes 85
6 INTERSECTIONS 90
6.1 Introduction 90
6.2 Location 90
6.3 Distance between Intersections 91
6.4 Intersection Types 92
6.5 Selection of Intersection Type 94
6.6 Design of Priority Intersections 99
6.7 Design of Roundabouts 107
6.8 Design of Signalised Intersection 116
7 SPEED MANAGEMENT 126
7.1 Introduction 126
7.2 Speed Management Principles 126
7.3 Speed Control Measures 127
8 OTHER ROAD FACILITIES 134
8.1 Pedestrian Facilities 134
8.2 Bus Lay-bys 137
8.3 Cycle Facilities 139
9 ROAD FURNITURE AND OTHER COMPONENTS 141
9.1 Traffic signs and road markings 141
9.2 Safety Barriers 141
9.3 Kerbs 147
9.4 Bridge parapets 149
9.5 Traffic islands 150
9.6 Pedestrian barrier 151
9.7 Street Lighting 152
10 PAVEMENT DESIGN (NO REVISION) 153
11 HYDROLOGY AND DRAINAGE (NO REVISION) 153
Government of Uganda
Ministry of Works, Housing and Communications
List of figures
Figure 0.1 The risk of being killed in traffic accidents depending on collision speed 10
Figure 3.1 Road environments 19
Figure 3.2 Speed power law 20
Figure 3.3 Swept path for single unit truck/bus with turning radius 10 metres 27
Figure 3.6 Swept path for semi-trailer combination with turning radius 10 metres 30
Figure 4.1 Single carriageway cross-section elements 35
Figure 4.2 Dual carriageway cross-section elements 35
Figure 4.3 Headroom and lateral clearance 36
Figure 4.4 Headrooms and clearances for traffic signs 37
Figure 4.5 Typical single carriageway rural road cross-section 39
Figure 4.6 Typical dual carriageway rural road cross-section 40
Figure 4.7 Typical cross-section for single carriageway roads in built-up areas 40
Figure 4.8 Typical cross-section for dual carriageway roads in built-up areas 41
Figure 4.9 Separate foot and cycleway on rural roads 43
Figure 4.10 Raised, kerbed footway in urban areas 43
Figure 4.11 Footway/cycleway on physically separated shoulders 43
Figure 4.12 Service roads 44
Figure 4.13 Median designs at speed limit 80 and 100 km/h 45
Figure 4.14 Segregated footway on bridges 46
Figure 4.15 Empirical data on the final position of vehicles that have run off the road 47
Figure 4.16 Example how to calculate clear zones 47
Figure 4.17 Clear zone correction factor for bends 48
Figure 4.18 Principles of smooth road side area design on fills and in cuts 48
Figure 4.19 Design of area in front of obstructions 49
Figure 4.20 Access road design 49
Figure 4.21 Culvert end design 50
Figure 4.22 Safe side drains 50
Figure 5.1 Description of the scale concept 52
Figure 5.2 Examples of landscape types 53
Figure 5.3 Drivers space or room concept 53
Figure 5.4 Some examples on rhythmical landscape adaptation 54
Figure 5.5 Example of adaptation to the landscape 54
Figure 5.6 Route planning by tangent method 55
Figure 5.7 Route planning by arc method 56
Figure 5.8 Basic space elements 57
Figure 5.9 Comparison of eye impression at straight arc and straight clothoid - arc 58
Figure 5.10 Road with and without optical kink 58
Figure 5.11 Comparison of alternative transitions from 2 % to + 3 % 59
Figure 5.12 A short straight between two curves should be avoided 59
Figure 5.13 Illustration of combination of horizontal curves and vertical grades 60
Figure 5.14 Coordination of vertical and horizontal elements in turning area 60
Figure 5.15 Examples of adaptation of bridge alignment 62
Figure 5.16 Bridge alignment without and with a bridge view 62
Figure 5.17 Measurement of stopping sight distance 63
Figure 5.18 Measurement of full overtaking sight distance 65
Figure 5.19 Offset and sight distance 65
Figure 5.20 Required offset at various horizontal radius and stopping sight distances 66
Figure 5.21 Required offset at various horizontal radii and overtaking sight distances 66
Figure 5.22 Terrain height for stopping sight distance 67
Figure 5.23 Terrain height (h) for overtaking sight distances 67
Figure 5.24 Available offset (a2) 68
Figure 5.25 Dynamic force system in curve 69
Page 3
Road Design Manual (Safety Draft May 2004
Revision)
Figure 5.26 Clothoid elements 72

Figure 5.27 Minimum length due to rate of change of centripetal acceleration 73
Figure 5.28 Simple clothoid 73
Figure 5.29 Compound curve 73
Figure 5.30 S-curve 74
Figure 5.31 Transition due to drainage and aesthetics for tangent-clothoid-radius 75
Figure 5.32 Rotation width calculation 76
Figure 5.33 Superelevation run-off length U speed limit 50 76
Figure 5.34 Super elevation run-off length U speed limit 80 77
Figure 5.35 Superelevation run-off length U speed limit 100 77
Figure 5.36 Superelevation transition design on tangent clothoid - circle 78
Figure 5.37 Alternative S-curve superelevation transition design 78
Figure 5.38 Drainage tracks at 9 m rotated width with propeller and displaced design 79
Figure 5.39 Super elevation run-off on compound curve. 79
Figure 5.40 Superelevation run-off on tangent - circle. 80
Figure 5.41 Superelevation run-off between two circles 80
Figure 5.42 Widening 81
Figure 5.43 Extra width needed to allow two semi-trailers to pass on a curve 81
Figure 5.44 Extra width needed to allow two 12 m buses to pass on a curve 82
Figure 5.45 Road widening design for tangent - circle 82
Figure 5.46 Vertical radius with R = 500 m 83
Figure 5.47 Example of effects of vertical alignment adjustment 83
Figure 5.48 Heavy truck speed behaviour in grades 85
Figure 5.49 Climbing lane outside the ordinary lane 86
Figure 5.50 Example profile 88
Figure 5.51 Example to assess truck speed profile 89
Figure 5.52 Resulting heavy truck speed profile 89
Figure 6.1 Example of planting at an intersection 91
Figure 6.2 Minimum distances between minor roads in a staggered intersection 92
Figure 6.3 Typical design of T-intersections 93
Figure 6.4 Typical designs for control intersections 94
Figure 6.5 Selection of intersection category as to safety 96
Figure 6.6 Selection of intersection category as to capacity 97
Figure 6.7 Selection of priority intersection type as to safety 98
Figure 6.8 Selection of control intersection type 99
Figure 6.9 Realignment of skew intersection 100
Figure 6.10 Visibility splays for Stop and Give way regulation 100
Figure 6.11 Traffic island in the minor road in partly channelised intersection 102
Figure 6.12 Traffic island in the minor road in a channelised intersection 103
Figure 6.13 Main design features in right turn lane design 104
Figure 6.14 Major road cross section 104
Figure 6.15 Design of left turn deceleration lanes 106
Figure 6.16 Design of left turn acceleration lanes 107
Figure 6.17 Three and five arm roundabout 108
Figure 6.18 Required visibility for entering a roundabout 109
Figure 6.19 Required visibility for drivers within a roundabout 109
Figure 6.20 Roundabout radii and widths 110
Figure 6.21 Radius of central island and circulating carriageway radius in normal roundabouts 111
Figure 6.22 Roundabout radii in small roundabouts 112
Figure 6.23 Number of entry lanes 112
Figure 6.24 Alternative design to increase the capacity for an entry 113
Figure 6.25 Design of approach deflection 113
Figure 6.26 Driving paths for passenger cars 115
Figure 6.27 Alignment between entry and exit 115
Figure 6.28 Location of pedestrian crossings 115
Figure 6.29 Primary motor vehicle conflicts 116
Figure 6.30 Protected right turn stage sequence 117
Figure 6.31 Right turn lanes with protected right turn 118
Page 4
Revision)
Figure 6.32 Right turn lanes with permissive right turn 118
Figure 6.33 Visibility requirements on intersection approach 118
Figure 6.34 Intervisibility zone without pedestrian crossing 119
Figure 6.35 Right turn lane design 120
Figure 6.36 Ghost island layout 120
Figure 6.37 Sight shadow design problem at permissive right turns 120
Figure 6.38 Lane drop design principles 121
Figure 6.39 Left turn slip lane with taper to facilitate large vehicles 121
Figure 6.40 Combinations of tapers and corner radii 122
Figure 6.41 Examples of swept path checks 122
Figure 6.42 Signal location advice 123
Figure 6.43 Primary signal location advice 123
Figure 6.44 Alternative signal locations for right turn lanes 124
Figure 6.45 Traffic signal island and pedestrian refuge 125
Figure 6.46 Example of a signal-controlled intersection with a staggered pedestrian crossing 125
Figure 7.1 Speed management principles on 50km/h through road 126
Figure 7.2 The design of gates 127
Figure 7.3 Alternative design of humps 128
Figure 7.4 Detailed design of hump transition 129
Figure 7.5 Example of hump drainage design at kerbed footway 130
Figure 7.6 Markings and signing of humps 131
Figure 7.7 Design of rumble strips 132
Figure 7.8 Design of road narrowings 132
Figure 7.9 Example of road narrowing with central island 133
Figure 8.1 Dropped kerb 135
Figure 8.2 Pedestrian crossing facilities 136
Figure 8.3 Standard bus lay-by 138
Figure 8.4 Recommended location for bus lay-bys 138
Figure 8.5 Cyclist dimensions 139
Figure 9.1 Guardrail determination of need 143
Figure 9.2 Determining the length of need 143
Figure 9.3 Steel beam strong post guardrail - typical details 144
Figure 9.4 Flaring of the end of guardrail to reduce the risk of impact 146
Figure 9.5 Typical transition (W-beam guardrail to rigid object) 146
Figure 9.6 Median barriers 147
Figure 9.7 Types of kerbs 148
Figure 9.8 Vehicle / pedestrian parapet - typical details 150
Page 5
Revision)
List of tables
Table 3.1 Functional road categories 16
Table 3.2 Vehicle traffic volume measures 17
Table 3.3 Design Hour Volumes percentage 17
Table 3.4 Classification of pedestrian volume 18
Table 3.5 Types of terrain 18
Table 3.6 Normal speed limits 21
Table 3.7 Definition of levels of service 22
Table 3.8 Level of service criteria for two-lane roads 23
Table 3.9 Selection of level of service 24
Table 3.10 Lane width factor 24
Table 3.11 Average passenger car equivalents for trucks and buses on two-lane roads 25
Table 3.12 Directional factor for two-lane roads 25
Table 3.13 Maximum AADT for two lane rural roads 25
Table 3.14 Design vehicles 26
Table 3.15 Normal use of standard road types for rural roads 33
Table 3.16 Normal use of standard road types for main roads through built-up areas 33
Table 3.17 Sight distances 34
Table 3.18 Curve radii 34
Table 3.19 Classification of at-grade intersections 34
Table 4.1 Additions to headroom at crests 36
Table 4.2 Minimum lateral clearances for traffic lanes, foot- and cycleways 37
Table 4.3 Recommended crossfalls 41
Table 4.4 Clear zone widths 46
Table 5.1 Desired and minimum stopping sight distances on level ground 63
Table 5.2 Stopping sight distances on grades 64
Table 5.3 Overtaking sight distances 64
Table 5.4 Rotation axle alternatives 68
Table 5.5 Desirable and minimum horizontal radii for safe driving 70
Table 5.6 Minimum radii required to maintain overtaking sight distance on right-hand flat curves 70
Table 5.7 Minimum radii required to maintain overtaking sight distance on left-hand flat curves 71
Table 5.8 Minimum values for resulting radius Rr without transition curve 71
Table 5.9 Minimum radius for convex vertical curves (crests) 84
Table 5.10 Minimum radii for concave vertical curves (sags) 84
Table 5.11 Maximum grades 85
Table 5.12 Threshold criteria for climbing lanes 86
Table 5.13 AADT-volumes to justify crawling lanes 87
Table 5.14 Entry and exit taper widths 87
Table 6.1 Classification of at-grade intersections 92
Table 6.2 Details of compound curve 101
Table 6.3 Normal length of right turn lanes 105
Table 7.1 Detailed design of circular humps 129
Table 7.2 Detailed design of plateau humps 130
Table 7.3 Road width for different speeds and meeting situations 133
Table 8.1 Criteria for provision of footways 134
Table 8.2 Minimum subway dimensions 137
Table 8.3 Recommended widths for cycle facilities 139
Table 9.1 Safety barrier characteristics 142
Page 6
Revision)
FOREWORD
Our Terms of Reference were to revise sections of the Ministrys Road Design Manual so that
the advice given reflects good safety engineering practice. It was also specified that the road
safety standards recommended should be in conformity with those in the neighbouring countries.
All sections of the Manual except the last two (Pavement Design, Hydrology and Drainage) have
been checked, and revised as necessary. No changes are proposed for sections 1 and 2.
The Road Design Manual needs to be updated and amended from several perspectives, not just
road safety. Because we were required to present complete sections of the Manual, we have had
to try and update all the technical advice. However, our principal focus has been on safety, and
the treatment of other non-safety issues may not be comprehensive or authoritative. We cannot
accept responsibility for any problems that may arise from following recommendations given
here regarding matters other than safety. Main sources for the updated safety recommendations
are:
SATCC 1998: Draft Code of Practice for the Geometric Design of Trunk Roads
Republic of Kenya 1979: Road Design Manual Part 1 - Geometric Design of Rural Roads
SNRA (Swedish National Road Administration) 2002: Road Design 94 S2 (in Swedish)
Danish Road Directorate: Road Recommendations (in Danish)
Strassenbau von A bis Z (the German design guidelines)
Austroads 1997: Rural Road Design - Guide to the Geometric Design of Rural Roads
Austroads 1993/95: Guide to Traffic Engineering Practice Part 6 Roundabouts and Part 5
Intersections at Grade
TRRL Research Report 114 1987: A review of some recent geometric road standards and
their application to developing countries
UK Highway Agency: Design Manual for Road and Bridges - Volume 6 Road Geometry
To focus on traffic safety issues and to make the manual easier to use it is suggested that the
section on geometric design in Road Design Manual is divided into two sections and that two
new sections are introduced. The table on the next page shows an overview of the proposed
changes.
Page 7
Revision)
Road Design Manual (RDM) Traffic Safety Revision

November 1994 Section Scope of revision
0. Introduction New section
1. General 1. General No revisions
2. The Road System 2. The Road System No revisions
3. Design Parameters and Control 3. Design Standard
4. Cross Section
4.Geometric Design
5. Alignment
5. Junctions 6. Intersections Major changes and additions
- 7. Speed Management
- 8. Other Road Facilities
6. Road Furniture 9. Road Furniture
7. Pavement Design Guide - Not revised
8. Hydrology and Drainage Not revised
Section 1 and 2 deal with the road network and road design in general, and contains no specific
traffic safety questions.
Section 3 is restructured, and a design procedure focused on the most important design and
safety criteria speed is presented.
Section 4 is divided into two sections and is rewritten to focus on traffic safety issues.
Section 5 is rewritten, and new designs of priority intersections (T-intersections) and control
intersections (roundabouts and signalised intersections) are presented.
Two new sections are included, covering speed management (essential for safer roads) and
miscellaneous road facilities.
Section 6 is shortened (parts moved to other sections) and rewritten.
Section 7 and 8 deal with technical questions that have no direct bearing on traffic safety.
Page 8
Revision)
0 INTRODUCTION
0.1 Main Requirements

Roads are designed and built to meet the demands of operational comfort and capacity of
transportation of people and goods. However, they must also be safe, environmentally friendly
and economical. The design manual describes how to design roads to achieve operational
comfort and capacity considering safety and economy and, to some extent environment. Thus, all
these considerations and features are built into the principles, criteria and values for the various
design elements given in the manual.
0.2 Safety Considerations

0.2.1 General Principles
Background
The view on traffic safety and traffic safety work has changed with the development of traffic
and the role of road traffic in the society. In the early years of motorization, cars were looked
upon as horse drawn carriages. Safety measures were mainly focused on vehicle requirements.
The development of technology, especially increased power and speed of motor vehicles, made
the comparison with horse-drawn carriages out-of-date. Safety measures were focused on
adapting people to this new traffic situation.
Today, the whole transportation system, of which the road traffic system is one part, is
contemplated. Safety measures are focused on reducing the exposure of risks, eliminating risk
factors and reducing the consequences of accidents. Typical measures are speed limits and
separation of motorized traffic from other types of traffic. With this approach, the purpose of
traffic safety work in road design is mainly to eliminate the risk factors and mitigate the
consequences of accidents. The long-term objective is that no one should be seriously injured or
killed when using the road traffic system.
Injury risks
The risk of being injured or killed in an accident increases considerably with increased speed. In
summary, many studies have shown that:
The number of injury accidents increases with the square of the average vehicle speed
The number of fatal accidents increases with the fourth power of the average vehicle speed.
Figure 0 .1 shows how the risk of being killed in a crash varies with collision speed. The graph
for pedestrians is well supported by research results, while the graphs for vehicle collisions are
partly based on expert assessments.
The graph shows that the risk of being killed increases rather slowly up to a speed where the risk
of being killed is around 10 percent - and then the risk increases rapidly. The conclusion is that a
road transport system should be designed to avoid conflicts at speeds where the risk to be killed
is higher than around 10 percent. This means that speeds should not exceed:
30 km/h in a pedestrian/vehicle collision
Page 9
Revision)
50 km/h in a side-on vehicle/vehicle or vehicle/object collision

70 km/h in a head-on vehicle/vehicle or vehicle/object collision.
From this, some basic planning and design rules can be derived, for example:
Vulnerable road users should be separated from motor vehicle traffic
At points of conflict between vulnerable road users and motor vehicles, speeds should be
low (preferably 30 km/h or lower)
Intersections should be designed to reduce collision speeds, especially for side-on collisions
(preferably to 50 km/h or lower)
The risk for head on collisions and collisions with rigid objects must be reduced to the
greatest possible extent.
However, measures to lower the speed, for example in intersections, are not sufficient. Measures
to reduce the risk of conflicts and the consequences of collisions must also be taken. Examples of
such measures are: the use of standard type intersections and reducing the number of potential
conflict points and the sizes of conflict areas.
R is k o f b e in g k ille d , %
100
60 P e d e s tr ia n
S id e c o llis io n
H e a d o n c o llis io n
20
40 80 120
C o llis io n s p e e d , k m /h
Figure 0.1 The risk of being killed in traffic accidents depending on collision speed
Safety responsibility
Research has shown that the human being is an unreliable operator in the road traffic system.
The most typical mistakes made by drivers and other road users are common to almost all drivers
and not limited to only a few. Consequently, all road users can be expected to make mistakes
which can lead to accidents. A road traffic system in which such common human mistakes leads
to fatal and serious injuries cannot be accepted. Common human mistakes should not lead to
catastrophes.
The responsibility for road safety must be shared between the road users and the road transport
system providers (mainly road authorities and vehicle manufacturers as well as legislative,
surveillance and enforcement bodies). The road users responsibility is to follow the system
requirements i.e., to obey the traffic laws and regulations, use available protection equipment
and behave with good judgement and responsibility. The responsibility of the system provider is
to provide a road system designed to minimize the risk of accidents and to only allow accidents
Page 10
Revision)
imposing forces to the human body that can be resisted without serious injuries. This
responsibility lies to a great extent with the road designer.
0.2.2 Safety considerations in design
Designing safety into roads is one of the main objectives of geometric design. It is important that
safety features are built into the road from the very start of the design. Safety considerations in
roads have the two objectives to provide design features to:
Prevent accidents, and
Reduce the seriousness of the accidents that occur.
Accident prevention
For the prevention of accidents the following points are especially important:
Provision of physical separation between motor vehicles in opposing directions and also
with other road users (especially pedestrians and cyclists)
Avoidance of surprise elements for the drivers, for example abrupt changes in standard,
insufficient visibility or poor phasing of horizontal and vertical alignment
Avoidance of situations where drivers must make more than one decision at the time
Provision of design features that reduce speed differentials between vehicles, for example
flat grades and speed change lanes
Proper location and design of intersections
Proper design, application and location of traffic signs, road markings and other traffic
control devices
Provision of design elements compatible with traffic volumes and type of traffic
Provision of proper drainage of the road surface.
Reducing the severity of accidents

A lot can be done to reduce the severity of accidents that we fail to prevent. The basic principles
are:
There should be a clear zone (safety zone) along each side of the road that is clear of
hazards such as lighting columns, other utility poles, rocks, drainage structures, etc.
Roadside slopes should be as flat as feasible (1:4 or flatter)
Large diameter sign posts and other supports which must be located within the clear zone
should be of a breakaway type
Safety barriers should be provided to protect vehicles from hitting dangerous obstacles that
cannot be removed or made breakaway and also to protect vehicles from falling of the road
down embankments.
0.3 Environmental considerations

No road project is without both positive and negative effects on the environment. The location
and design of the road should aim at maximising the positive effects and minimising the negative
effects. A positive effect could be to remove traffic from environmental vulnerable areas. The
negative effects are related to both the road as a physical feature and to the traffic on the road.
Page 11
Revision)
Effects related to the road as a physical feature

Effects related to the road as a physical feature are generally difficult to quantify. In many cases
it is necessary to seek the advice and services from other professions to reach a proper evaluation
of the problems and to find adequate remedial measures. The following factors should be
considered in the location and design of roads:
The preservation of the natural beauty and scale of the countryside and the adaptation to the
conditions and architecture of the city
The preservation of areas and land use of particular value, including
- national parks and other recreational areas
- wildlife and bird sanctuaries
- forests and other important natural resources
- land of high agricultural value or potential
- other land use of great economic or employment importance
- historic sites and other man-made features of outstanding value
The prevention of soil erosion and sedimentation
The prevention of health hazards by ponding of water leading to the formation of swamps
The avoidance or reduction of visual intrusion.
Effects related to the traffic

Negative effects related to the traffic can often be quantified, for example noise levels and air
quality. The effects which should be considered are:
Noise pollution
Air pollution
Ground water pollution
Vibrations
Severance of areas (barrier effect).
Countermeasures to these environmental problems are outside the scope of this safety revision.
In short, however, the best way to avoid the problems is to locate the road outside trading centres
and towns. If this is not possible the best way to reduce the problems is to lower the speed and
provide safe crossings for local traffic, pedestrians and cyclists.
0.4 Economic considerations

0.4.1 Introduction
The total costs for the society for roads and road traffic are to a large extent depending on the
design of each individual road project. The most economic design will often not involve the
shortest route or the use of minimum standards. Savings in road maintenance costs, vehicle
operating costs, travel time costs and accident costs etc may offset the extra construction costs
for a road with higher design standard. The economic outcome of the design will be decided both
in the route selection and in the geometric design of the chosen route.
Page 12
Revision)
0.4.2 Route selection

To find the optimum alternative for each project several different routes should be tried. Some of
these should be chosen and compared in a comprehensive study including the following types of
effects:
Land acquisition and intrusion
Construction and operation costs
Road user costs including travel time, accidents, vehicle operating costs
Environmental effects
Some effects, for example, road construction and maintenance costs and also travel time and
accidents, can be described in monetary values. Other effects, like traffic noise and air pollution,
can be quantified but are difficult to value in money (though valued in most European countries),
while other effects, for example impacts on wild life or other valuable resources, generally only
can be described verbally.
A cost-benefit and objective analysis including as many effects as possible should be made for
all studied alternatives.
0.4.3 Detailed design

The geometric standards of individual elements of the road will usually vary with the terrain.
Elements for which the chosen geometric standards are difficult to obtain should be identified.
The economic consequences of different standards for these elements should be considered and a
cost-benefit analysis including construction and road user costs should be made. In general, the
higher the class of road, and hence volume of traffic, the more likely will benefits from road user
savings lead to the justification of a higher road standard.
The standards described in this manual are intended to provide guidance for designers rather than
to be considered as rigid minima. For some projects or elements relaxations of standards may be
essential in order to achieve an acceptable rate of return of investment.
Page 13
Revision)
1 GENERAL (NO REVISIONS)
2 ROAD SYSTEM (NO REVISIONS)
Page 14
Revision)
3 DESIGN STANDARD
3.1 Design Procedure
Basic prerequisites
Every road is unique as to function, traffic volumes and composition, terrain, geotechnical
conditions and environment. Thus the design should be based on some basic prerequisites
concerning objectives and restrictions which should be identified by the designer for each section
of the road according to section 3.3. The objectives are determined by the following
prerequisites:
road category
expected traffic volumes and composition.
Restrictions are mainly imposed by:

terrain (topography) and geotechnical conditions
road environment (built up areas etc).
Design criteria
For each section of the road the design standard should be based on a number of design criteria,
normally selected in an iterative process from the feasibility study phase to the preliminary
design phase and on to the final design. The most important design criteria to be selected by the
designer are:
Speed limit and design speed
Separation of vulnerable road users from other traffic
Level of service
Design vehicles.
Design components
Having identified the basic prerequisites and selected the design criteria the highway designer
can proceed to assemble the basic components. These include:
Cross-section, including the roadside area
Alignment (horizontal and vertical)
Intersections (location and type)
Facilities for vulnerable road users
Facilities for public transport
Route guidance and other signing.
3.2 Basic Prerequisites

The basic prerequisites are related to the desired function and the use of the road defined by:
Road category
Page 15
Revision)
Expected traffic volumes.

and the restrictions imposed by:
Terrain (topography) and geotechnical conditions
Road environment.
The basic prerequisites and restrictions should be identified and defined as follows.
3.2.1 Road category

The functional road category has major importance for the selection of speed limit, separation of
vulnerable road users, cross-section with roadside area, and alignment standard. The road
category must be defined for different parts of the project. The Road Design Manual defines the
following road categories with desired speed limits:
Road Desired rural

Description Function
category speed limit
Category A Principal arterial Connection between the national road 100 km/h
system system and those of neighbouring countries
(Primary roads) Linkage between the provincial capitals, main
centres of population and production centres
Category B Minor arterial system Connection between local centres of 80 km/h
(Secondary roads) population
Linkage between districts, local centres of
population and development areas with the
principal arterial system
Category C Collectors Linkage between locally important traffic 80 km/h
(Tertiary roads) generators and their rural hinterland
Provision of service to smaller communities
Category D Local road system Provision of access to land adjacent to the 80 km/h
(Feeder roads) collector system
Table 3.1 Functional road categories
3.2.2 Expected traffic volumes

The traffic volumes are used to decide on level-of-service for motor vehicles on links and in
intersections, see section 3.3.3 for links and section 6.5 for intersections, and the need of
separation for pedestrians and cyclists.
The design of the road, cross-section, intersection types, alignment etc, should be based upon a
forecast of the traffic volumes of motor vehicles, cyclists and pedestrians to be expected in the
design year - normally chosen as year 10 after the year of opening to traffic.
Projects to upgrade existing roads should always include a survey of existing traffic conditions
for all types of traffic. Major generators of vehicular and pedestrian traffic must be identified,
including bus stops, interchange points, etc.
Page 16
Revision)
Motor vehicle traffic

The general measures of motor vehicle traffic volumes to be used on links are given in table 3.2.
Traffic volume should be expressed in passenger car units (pcu). Trucks and buses should be
converted to passenger car units using the equivalence factors in table 3.11.
Measure Definition
AADT Average Annual Daily the total traffic volume for both directions of flow during the
Traffic year divided by 365 in passenger car units per day (pcu/day)
ADT Average Daily Traffic total traffic volume for both directions of flow during a given
period (less than a year) divided by the number of days in
that period (pcu/day)
DHV Design Hour Volume for a two lane road: the total two-way full hour volume that is
selected for design purposes (pcu/h)
Table 3.2 Vehicle traffic volume measures
For low volume roads the design traffic is AADT in the design year. For routes with great
seasonal variations the design traffic is the ADT during the peak months of the design year.
The Design Hour Volume, DHV, is expressed as:
DHV = AADT * K or ADT * K
K is normally chosen as the traffic volume of the 30th highest hour expressed as percentage of the
AADT or ADT. Typical K-factors are given in Table 3 .3 below. The higher percentages in the
table refer to roads with relatively high concentration of traffic during rush hour or large seasonal
variations.
Traffic condition K (30th DHV)
heavily trafficked road under congested urban conditions 8%
normal urban conditions 10 15 %
roads catering for recreational or other traffic of seasonal nature 20 30 %
Table 3.3 Design Hour Volumes percentage
Traffic volumes by direction should also be estimated for major intersections to assist in the
selection of right turn lanes, roundabout or traffic signal control.
Volumes of vulnerable road users

The term vulnerable road users normally means pedestrians, pedal cyclists and motorcyclists.
However, for the purposes of this manual, which mostly concerns rural roads, we can treat
motorcyclists the same as other motorised vehicles. Surveys must be made of pedestrian and
cycle volumes - a general classification is given in table 3.4. The survey must identify major
Page 17
Revision)
pedestrian generators, such as bus stops, schools, informal markets, as well as the main routes
people take, and the places where they come into conflict with motorised traffic.
Class Volume Typical situations
(2-way flow
per day)
Very low <100 Rural very few houses
Low 100 500 Rural some houses
Moderate 500 1,000 Minor trading centre, or approach to
major trading centre or town
High 1,000 5,000 Major trading centre or small town
Very high >5,000 Major town
(adapted from SATCC 1998: Draft Code of Practice for the Geometric Design of Trunk Roads )
Table 3.4 Classification of pedestrian volume
3.2.3 Terrain (topography) and geotechnical conditions

The type of terrain and geotechnical conditions have major importance on investment and
maintenance costs for the selection of speed limit and consequently alignment standard. The type
of terrain must be defined for different parts of the project and geotechnical conditions
thoroughly surveyed.
The Road Design Manual defines the following types of terrain:
Type of terrain Description
Level terrain Level or gently rolling country which offers few obstacles to the construction
of road having continuously unrestricted horizontal and vertical alignment
Rolling terrain Rolling, hilly or foothill country where the slopes generally rise and fall
moderately gently and where occasional steep slopes may be encountered.
It will offer some restrictions in horizontal and vertical alignment
Mountainous terrain Rugged, hilly and mountainous country and river gorges. This class of
terrain imposes definite restrictions on the standard of alignment obtainable
and often involves long steep grades and limited sight distances
Table 3.5 Types of terrain
3.2.4 Road environment

The environment is of major importance for the selection of speed limit, alignment and speed
control devices and the need of separation. The environment should be defined for all parts of the
project as:
game or forest reserve
rural
minor trading centre
major trading centre
town
game or forest reserve.
Page 18
Revision)
Rural Town
Minor trading centre Major trading centre
Figure 3.2 Road environments
3.3 Design Criteria

The most important design criteria are:
Speed limit and design speed
Separation of vulnerable road users
Level of service
Design vehicles.
Design criteria must be chosen for each section of the road with possible revisions from the
feasibility stage to the final design. The following sections give an overview how to decide on
design criteria.
Page 19
Revision)
3.3.1 Speed limits and design speed

Speed and traffic safety
Actual vehicle speeds on a road section is of utmost importance for level-of-service as well as
for traffic safety. The speed at which a driver will choose to travel a section of road is generally
determined by what he considers safe based on the visual concept of the cross-section and road
alignment with terrain adaptation, surface quality and other conditions. Individual driver
decisions will vary significantly due to personality, experience and vehicle capacity. The
standard deviation of speeds is normally some 10 % in European countries but far larger in
Uganda.
Road and traffic history reveals that drivers tend to overestimate safe speeds. Speed limits
should be imposed where speeding is a problem, but compliance with signs is low, and
enforcement is difficult. Consequently it is crucially important to reinforce the speed limits
with physical speed control measures, especially on through roads in trading centres and
towns.
Modern traffic safety research indicates that the number of fatal accidents is proportional to the
average speed in power four, fatal and serious injury accidents in power three and fatal and all
injury accidents in power two as illustrated in Figure 3 .3.
150 Accident change (%)
fatal
fatal and severe injury
fatal and injury 100
50
Speed change (%)

0
-25 -20 -15 -10 -5 0 5 10 15 20 25
-50
-100
Figure 3.3 Speed power law

This speed-power law gives as an example for an increase of average speed from 80 km/h to 90
km/h, some 12 %, i.e. without compensation of improved drivers, vehicles or road design an
increase of:
the number of fatal accidents with some 60 %
the number of fatal and serious accidents with some 40 %
the total number of fatal and injury accidents with some 25 %.
Page 20
Revision)
Speed limits and design speed

Each project should include a speed management plan. This plan should show sections with the
normal speed limit and sections with higher and lower speed limits.
The normal speed limit on rural roads should be 80 km/h, see Table 3 .6.
Speed Limit Road Category Road Environment
80 km/h A-D Rural
50 km/h A-D Trading centres, towns, cities
Table 3.6 Normal speed limits

For high-standard category A roads in rural areas with low volumes of pedestrians and cyclists
the speed limit can be increased to 100 km/h. For roads of a lower standard (e.g., where the
cross-section and alignment are constrained by the terrain) the speed limit should be reduced.
The normal built-up area speed limit of 50 km/h should be reduced to 30 km/h on roads where
there is major conflict between vulnerable road users and motor vehicle traffic. This might occur
on narrow main streets through towns, especially where there are no footways. 30 km/h is the
appropriate speed limit for local access streets in built-up areas.
The speed limit is the maximum allowed speed on the road. The Traffic and Road Safety Act
1998 provides for there to be a maximum permitted speed for each vehicle type car, bus, truck,
etc. Regulations currently under preparation set the maximum speed limit for cars at 100 km/h,
and trucks and buses at 80 km/h. Hence, on a road with the speed limit 100 km/h cars are
allowed to go 100 km/h, but trucks and buses only 80 km/h.
Design speed is a measure of the cross-section and alignment quality of the road. It may be
defined as the maximum safe speed that can be maintained at a given section of the road where
conditions are so favourable that the design features of the road govern.
Alignment elements such as minimum vertical and horizontal curves, superelevation and
alignment combination rules and minimum sight distances are directly dependant on design
speed. Other features like carriageway and shoulder widths and surface smoothness also affect
speeds. Chicanes, narrowed cross-sections and humps are effective design measures to control
speeds in urban and village areas with many vulnerable road users.
The design speed for longer sections should not be lower than the speed limit and preferably 10
km/h higher than the speed limit. Short rural sections with design speeds lower than the speed
limit could be treated with warning signs and no overtaking road markings. Special alignment
combination rules should also apply, see section 5. Roads of category C and D must sometimes
be designed to much lower design speeds than the formal speed limit.
3.3.2 Separation of vulnerable road users from motorised traffic

Separating vulnerable road users from motorised traffic has great safety benefits. Whether and
how to do this requires careful assessment. The main ways of achieving separation in rural areas
are:
unpaved shoulder
paved shoulder separated from the traffic lane with an edge line marking
separate footway in some circumstances it may also be used by cyclists.
Page 21
Revision)
For roads through trading centres and towns the options are:
paved shoulders
a footway (or combined footway/cycleway) physically separated from the traffic lane by
a barrier kerb or similar
a raised, kerbed footway
a raised, kerbed footway with a service road beyond it.
The general recommendations on separation are as follows: (see also section 4.3 and 4.4)
A separate footway must be provided on Category A and B rural roads if they are to have a
speed limit of 100 km/h an exception can be made on sections where the volume of
pedestrians is very low, in which case 1.5m wide paved shoulders must be provided
Rural roads with a speed limit of 80 km/h shall be provided with 1.5m wide paved shoulders
for use by pedestrians and cyclists on sections where the volume of pedestrians is moderate
or high a separate footway shall be provided, in which case the paved shoulder can be
reduced in width to 1m.
A separate footway shall normally be provided alongside through roads in built-up areas.
3.3.3 Level-of-service
Definition of level of service
The maximum volume that can be carried at any selected level of service is referred to as the
service volume for that level. Service volumes take place under ideal conditions, i.e. no
restrictive geometric, traffic or environmental effects. The definitions of the various levels are
shown in Table 3 .7
Traffic volume
Level of Operating
(total for both Traffic conditions
service speed
directions)
Level A 95 km/h 420 pcu/h 75 % of desired passing manoeuvres
possible with little or no delay
Level B 80 km/h 760 pcu/h Beginning of stable flow; most drivers
affected by other vehicles (but not
unreasonably)
Level C 65 km/h 1200 pcu/h Stable flow
Level D 55 km/h 1880 pcu/h Unstable flow

Level E 50 km/h 2800 pcu/h Capacity
Table 3.7 Definition of levels of service
Page 22
Revision)
The following level of service criteria apply for two-lane rural roads - Table 3.8.
No Volume-capacity ratio (V/C) (2)

Level Percent overtaking
of time zone (% of Level Rolling Mountainous
service delay total length) terrain terrain terrain
(1)
0 0.15 0.15 0.14
A 30 20 0.12 0.10 0.09
40 0.09 0.07 0.07
0 0.27 0.26 0.25
B 45 20 0.24 0.23 0.20
40 0.21 0.19 0.16
60 0.19 0.17 0.13
0 0.43 0.42 0.39
20 0.39 0.39 0.33
C 60 40 0.36 0.35 0.28
60 0.34 0.32 0.23
80 0.33 0.30 0.20
0 0.64 0.62 0.58
20 0.62 0.57 0.50
D 75 40 0.60 0.52 0.45
60 0.59 0.48 0.40
80 0.58 0.46 0.37
E See US Highway Capacity Manual
Table 3.8 Level of service criteria for two-lane roads

Note 1 The percentage of road length along which sight distance is less than 500 m on 80 km/h
speed limit roads and 700 m on 100 m/h speed limit roads.
Note 2 Ratio of hourly volume to an ideal capacity of 2800 passenger cars in both directions.
Selection of level of service

Level of service for motor vehicles should be decided for each section of the road. The analysis
should be based on average terrain, geometric and traffic conditions, and is usually applied to
road sections of at least 5 km length. Analysis of specific grades on a two-lane road is more
complex, involving operation of upgrade vehicles. The procedure in the U.S. Highway Capacity
Manual could be used in those cases. The following level-of-service should normally be selected
for the design hour due to road class and terrain type, see Table 3 .9.
Page 23
Revision)
Level of service
Road class Level Rolling Mountainous
terrain terrain terrain
A and B Arterials C* C D
C Collectors C C D
D Local roads E E E
* B could be considered if speed limit 100 km/h
Table 3.9 Selection of level of service
Determination of level of service

The service volume is calculated as follows:
SV = 2800 * (V/C) * Fw * Ft * Fd
Where: SV = total service volume for selected level-of-service
V/C = volume-capacity ratio ...
Fw = lane width factor
Ft = truck factor
Fd = directional factor
Lane width factor
For road sections with lane width less than 3.5m, the reduction factor Fw, is given in the
following table.
Lane width Factor
m Fw
3.50 1.00
3.35 0.89
3.00 0.73
2.75 0.61
Table 3.10 Lane width factor

Truck factor
The truck factor, Ft, is dependent on the level of service, the type of vehicle and the type of
terrain. The truck factor is determined from the relationship:
Ft = 1 / (1 + Pt (Et 1) + Pb (Eb 1)
Where: Ft = truck factor
Pt = proportion of trucks in the traffic, expressed as a decimal
Pb = proportion of buses in the traffic, expressed as a decimal
Et = passenger car equivalent for trucks
Eb = passenger car equivalent for buses
Page 24
Revision)
Average passenger car equivalents for trucks and busses on two-lane roads are shown in Table
3 .11 below.
Type of terrain
Type of Level of
vehicle service Level Rolling Mountainous
terrain terrain terrain
A 2.0 4.0 7.0
Trucks Et B and C 2.2 5.0 10.0
D and E 2.0 5.0 12.0
A 1.8 3.0 5.7
Buses Et B and C 2.0 3.4 8.0
D and E 1.5 2.9 6.5
Table 3.11 Average passenger car equivalents for trucks and buses on two-lane roads
Directional factor
All of the V/C values given in the tables above for selection of level of service for rural roads are
for 50/50 directional distribution of traffic on a two-lane road. For other directional distributions,
the factor shown in Table 3 .12 shall be applied.
Directional distribution 50/50 60/40 70/30 80/20

Directional factor, Fd 1.00 0.94 0.89 0.83
Table 3.12 Directional factor for two-lane roads
Examples
The following maximum AADTs for different level of services due to terrain type could be used
for planning purposes. The basic assumptions for the calculations have been 25 % trucks, a K-
factor of 0.15 and 20, 40 and 60 percent no passing zones for level, rolling and mountainous
terrain.
Level of Maximum AADT

service Level terrain Rolling terrain Mountainous terrain
A 1 600 700 300
B 3 200 1 650 700
C 5 200 3 000 1 250
D 8 700 4 500 1 900
Page 25
Revision)
Table 3.13 Maximum AADT for two lane rural roads
Capacity calculations could be performed for alternative designs to analyse the impact of cross-
section and alignment on traffic operations.
3.3.4 Design vehicles

Introduction
The road system must be adapted to the various sizes of vehicles using the road system. The
principal vehicle properties affecting design are:
vehicle dimensions such as height, width and length with axle configuration and maximum
wheel angle
vehicle power/weight-ratio.
The principal design elements affected are:

cross-section of the road
road widening in horizontal curves
intersection lay-out
grade design and climbing lanes.
Design vehicles
Until detailed information on the different vehicle types using the roads in Uganda becomes
available, the following three design vehicles should be used as controls in geometric design:
passenger car
single unit truck
semi trailer combination.
Dimensions of the design vehicles are presented in Table 3 .14.
Front Rear
Wheel- Overall Overall
Design Vehicle over- over- Height
base length width
hang hang
Passenger car 2.9 m 0.9 m 1.2 m 5.0 m 1.9 m -
Single unit truck / bus 6.0 m 2.8 m 3.4 m 12.0 m 2.5 m 3.8 m
Semi trailer combination see figure below 16.0 m 2.5 m 3.8 m
Page 26
Revision)
1 .2 3 .2 1 .3 8 .0 1 .3 1 .0
Table 3.14 Design vehicles
Swept paths
Templates showing the swept paths for single unit trucks/buses and semi trailer combinations are
shown on the following six pages.
Page 27
Revision)
0 10m
(adapted from EU agreements on vehicle manoeuvrability)
Figure 3.4 Swept path for single unit truck/bus with turning radius 10 metres
Page 28
Revision)
0 10m
Page 29
Revision)
0 10m
Page 30
Revision)
0 10m
Figure 3.7 Swept path for semi-trailer combination with turning radius 10 metres
Page 31
Revision)
0 10m
Page 32
Revision)
0 10m
Page 33
Revision)
3.4 Main Design Features

The main design features described in the Manual are:
Cross-section including the roadside area
Alignment, with minimum vertical and horizontal elements, stopping and overtaking
sight distances, visual guidance, etc
Intersections (location and type).
These main design features must be developed from the feasibility stage towards to the final
design. An overview of these components is given in this section. Detailed descriptions of how to
select the design standard for each component are to be found in sections 4-9.
3.4.1 Cross-section road type

The Manual defines, see section 4, a number of standard road types, characterized by the cross-
section, for rural roads and for main roads through trading centres and towns.
The appropriate cross-section should be selected for each part of the road. It will be unusual for
one road to have the same cross-section throughout. Table 3 .15 and Table 3 .16 provide a
summary of the basic road types - see also section 4.3.
Rural road Functional Planned speed limit

Paved width
type road category Desirable Acceptable 1)
Class RI A 100 km/h 80 km/h 10.0 m
Class RII B 80 km/h 70 km/h 9.5 m
Class RIII C and D 80 2) km/h - 9.0 m
2)
Class RIV C and D 80 km/h - 4.0 m
1) Could be used if costs, environmental impact etc otherwise are heavy
2) often with relaxed design speed standard
Table 3.15 Normal use of standard road types for rural roads
Functional road Planned

Through road type Environment
category speed limit
Single carriageway Trading centres
A-D 50 km/h
through road Towns
Dual carriageway Major trading centres
A, B 50 / 70 km/h
through road Towns
Table 3.16 Normal use of standard road types for main roads through built-up areas
3.4.2 Roadside areas

The Manual describes (section 4.5) how to design safe roadside areas. It covers safety zones,
side slopes, side drains and roadside hazards.
Page 34
Revision)
3.4.3 Alignment
The alignment standard is determined broadly by the speed limit and design speed (see section
3.3.1.). The major speed-related alignment parameters are the stopping sight distances (SSD),
full overtaking sight distances (FOSD) and no overtaking sight distances. These are used to
determine the minimum safe radii for vertical and horizontal curves, given requirements for
comfort and whether or not overtaking is to be permitted. See section 5 for detailed advice.
SSD SSD No
Speed FOSD
desirable minimum overtaking
50 70 50 300 150
80 135 110 500 240
100 195 165 700 300
Table 3.17 Sight distances

(adapted to the speed limit and design concept from Kenyan, Swedish, German, Australian and UK design guidelines)
Vertical alignment Horizontal alignment

Sag curves Crest curves Grade FOSD Comfort
Speed SSD SSD
des min FOSD % Right Left des min
des min
80 1300 1100 32000 4500 3000 4-6 3000 5-9000 400 300
100 2000 1500 65000 10000 7000 3-5 5-6000 8-17000 600 500
Table 3.18 Curve radii

(adapted to the speed limit and design concept from Kenyan, Swedish, German, Australian, SATCC and UK design guidelines)
3.4.4 Intersections
Many accidents occur in intersections, even in lightly-trafficked at-grade intersections, and from
a safety point of view intersections require great attention in the design of a road. The selection
of intersection type and the detailed design of intersections is described in section 6.
The Manual covers only at-grade intersections. The classification of at-grade intersections is
shown in Table 3 .19.
Traffic regulation
Intersection
Intersection types
category
Major road Minor road
A. Unchannelised T-intersection
Stop or give way
Priority intersection Priority B. Partly Channelised T-intersection
control
C. Channelised T-intersection
D. Roundabout
Control intersection Stop or give way control E. Signalised intersection
Page 35
Revision)
Table 3.19 Classification of at-grade intersections
4 CROSS-SECTION
4.1 Cross-section elements
The major geometric design elements constituting the cross-section are the traffic lanes, median
(on a dual carriageway road) the shoulders, the roadside areas, and pedestrian and cycle
facilities, if any. Refer to Figure 4 .10 and 4.2. The term carriageway is used here to cover the
traffic lanes, any auxiliary lanes, and the shoulders.
The support strip, if needed is designed to give the pavement sufficient stability and bearing
capacity. The edge strip, the inner part of the shoulder, is assigned for edge markings.
The width and slope of traffic lanes and shoulders are based on traffic needs and drainage
requirements. The design of the embankments and cuttings depends on the type of material to be
excavated, environmental and traffic safety issues and the importance of the road. The road
reserve is also used as a legal term, see Roads Act, 1949.
c a rr ia g e w a y r ig h t-o f-w a y
b o u n d ary
su p p o rt su p p o rt
s tr ip tr a ff ic la n e tra ffic la n e s trip
s h o u ld e r s h o u ld e r cut
c a tc h d r a in
la te ra l
la te ra l c le a ra n c e
c le a ra n c e cam ber % cam ber % b a c k o r o u te r
s lo p e
f ill o r
em bankm ent
e d g e s tr ip f o r ro a d m a rk in g s
fo re o r in n e r ta b le d ra in
slo p e
n a tu r a l te rr a in v erg e
ro a d sid e a re a
s a fe ty z o n e
ro a d p rism
v e rg e
ro a d re se rv e
Figure 4.10 Single carriageway cross-section elements
d iv id e r
k erb ed edge s e p a ra te
fo o tw a y edge edge edge
s trip fo o tw a y /
s trip tra ffic la n e s s trip m e d ian s trip tra ffic la n e s c y c lew a y
o u ter in n e r
cam ber % cam ber %
in n e r h a rd o u ter h ard
o u te r h a rd s h o u ld e r
s h o u ld e r s h o u ld e r
Figure 4.11 Dual carriageway cross-section elements
Page 36
Revision)
4.2 Headroom and Lateral Clearance

Headroom is the required height to allow traffic to pass safely under objects restricting the
height. It should normally be provided over the full width of the carriageway.
Foot and h ea d ro o m
h ea d ro o m c y c le w a y tra ffic la n e s
la te ra l c le a ra n c e
Figure 4.12 Headroom and lateral clearance
The maximum legal height for a vehicle in Uganda is 4.0m. In determining the headroom
standard, allowance must be made for:
The road surface being raised during pavement overlay work
The possibility of an overbridge collapsing if hit by a vehicle
The need to allow for occasional oversized vehicles.
The recommended headroom under bridge structures should be 5.0 m on category A and B roads
and 4.5m on lower road classes. The headroom should be 6m under high-power cables and 5 m
under low-power cables. The minimum headroom over footways and cycleways should be 2.5m.
An addition to the normal headroom is needed at crests with radii below 700m see table 4.1.
Crest radius (m)

100 200 300 400 500-700
Addition (m) 0.12 0.06 0.04 0.03 0.02
Table 4.20 Additions to headroom at crests

Lateral clearance, see figure 4.3 above, is the minimum permitted distance between the edge of
the traffic lane, the footway/path or cycle way and the nearest fixed object, see figure 4.3. Fixed
objects must not be so close as to discourage the driver from making full use of the traffic lane.
The recommended lateral clearances are given in table 4.2.
Page 37
Revision)
Footways
Speed limit and cycle- Foot-
Impacting object paths1)
ways 1)
30 50 80 100
height lower 0.00 0.00 0.25 0.25 0.00 0.00
than 0.2 m
height higher 0.30 0.60 1.50 2.00 0.30 0.15
higher than 0.2 m
guardrail 0.30 0.60 0.60 0.60 0.00 0.00
roofs etc 1.00 1.00 1.00 1.00 0.50 0.50
1) including hard shoulders for walking and cycling
Table 4.21 Minimum lateral clearances for traffic lanes, foot- and cycleways
Minimum headrooms and clearances for traffic signs are given in the MOWHC Traffic Signs
Manual (under preparation) see Figure 4 .13.
Minimum (mm) Desirable (mm)

A 600 1000
B 300 600
C 300 600
D 1000 1800
E 2100 2100
1)
F 5000 5200
1) maximum allowed vehicle height 4.00 m
Could be decreased to 4.5 m on category C and D roads
Figure 4.13 Headrooms and clearances for traffic signs
4.3 Selection of Cross-section

The most important traffic safety factors for cross-section selection in rural areas are:
speed limit
the need to separate vulnerable road users from other traffic.
Other factors of major importance are:
Page 38
Revision)
road category
predicted traffic flows for the design year and design level-of-service
road user costs
construction and maintenance costs
intrusion and other environmental impacts.
The cross-section may vary over a particular length of road due to controlling factors such as
volumes of vulnerable road users, terrain and geotechnical conditions. The basic requirements
are, however, that changes in cross-section shall not be made unnecessarily, that the cross-section
standard should be uniform within each sub-section and that any changes should be effected
gradually and logically over a transition length. Abrupt or isolated changes in cross-section lead
to increased hazards and reduced safety as well as capacity.
In certain cases, however, it may be necessary to accept isolated reductions in cross-section
standards, for example when an existing narrow structure has to be retained because it is not
economically feasible to replace it. In such cases a proper application of traffic signs and road
markings is required to warn motorists of the discontinuity in the road.
The capacity and level-of-service design criteria (see 3.3.3), depend heavily on overtaking sight
distances along the road. Level-of-service could be improved by use of climbing lanes and also,
sometimes, by use of overtaking lanes.
The HDM IV-procedure could be used to optimise the cross-section and alignment selection.
Traffic flow forecasts, design level-of-service and calculation procedure recommendations are
given in section 3.
In order to simplify the selection and design of cross-section elements and promote uniformity in
standards, a set of standard cross-sections have been laid down, and guidelines are presented for
the selection of the appropriate cross-section.
4.3.1 Standard cross-sections for rural roads

Single carriageway roads
High-standard single carriageway roads with speed limits of 100km/h, normally category A
roads, should have 3.5m traffic lanes and 1.5m hard shoulders. Where the volume of pedestrians
on any section is such that a separate footway is provided (see section 3.3.2) it may be acceptable
to reduce the width of the paved shoulder to 1m.
Single carriageway roads of category B (normally with a 80km/h speed limit) should have 3.25m
traffic lanes and 1.5m hard shoulders. Category C and D roads will normally have 3.0m traffic
lanes and 1.5m hard shoulders. Where the volume of pedestrians on any section is such that a
separate footway is provided (see section 3.3.2) it may be acceptable to reduce the width of the
paved shoulder to 1m or 0.5m.
Where the design year AADT is below 200 pcu/day it may be acceptable to have 2.0m wide
traffic lanes with 1.0m unpaved shoulders.
Figure 4 .14 shows the typical cross-section for single carriageway roads. See section 4.4 for
details of variations on this typical cross-section.
Page 39
Revision)
c a rr ia g e w a y
su p p o rt su p p o rt
s trip s h o u ld e r tra f fic la n e tra f fic la n e s h o u ld e r s trip
e d g e s tr ip f o r ro a d m a r k in g s e d g e s tr ip f o r ro a d m a r k in g s
Paved Unpaved
Road type Traffic lanes Total carriageway width
shoulders shoulders
Class RI 7.0 m 1.5 m 1) - 10.0 m
1)
Class RII 6.5 m 1.5 m - 9.5 m
1)
Class RIII 6.0 m 1.5 m - 9.0 m
Class RIV 4.0 m - 1.0 6.0 m
1. If a separate footway/cycleway is provided the paved shoulder may be reduced in width.
Figure 4.14 Typical single carriageway rural road cross-section
Dual carriageway roads

The level-of-service standards for Category A and B roads in flat terrain can only be met by
building a dual carriageway road once the design year volume exceeds about 13000 pcu/day.
However the safety benefits of dual carriageway roads can make them worthwhile at lower
threshold volumes, especially if the speed limit is to be 100km/h.
A dual carriageway with a speed limit of 100km/h should have 3.5 m traffic lanes and 2.0 m
outer paved shoulders. The median should be 9m wide if it is narrower than this consider
installing a median barrier.
A dual carriageway with a speed limit of 80km/h should have 3.25 m traffic lanes and 1.5 m
paved shoulders. All dual carriageways should have separate footways/cycleways.
Figure 4 .15 shows the typical cross-section. See section 4.4 for details of variations on this
cross-section.
Page 40
Revision)
d iv id e r
edge s e p a ra te
edge edge edge
s tr ip fo o t- a n d
s trip tra ffic la n e s s trip m e d ian s trip tra ffic la n e s c y c le w a y
o u te r in n e r
in n e r h a r d o u ter h ard
o u te r h a rd s h o u ld e r
s h o u ld e r s h o u ld e r
Paved Total carriageway

Road type Traffic lanes Median 1)
shoulders width
4.0 m 20.0 m
Class RV 2 x 7.0 m 2.0 m
9.0 m 29.0 m
4.0 m 18.5 m
Class RVI 2 x 6.5 m 1.5 m
6.0 m 20.5 m
1. Including inner hard shoulder. Consider installing a median barrier where the median is less
than 9.0m wide see section 9.2.9
Figure 4.15 Typical dual carriageway rural road cross-section
4.3.2 Standard road types for main through roads in built-up areas
It must be made very clear to drivers entering the built-up area that they have to slow down. This
can be done by completely changing the appearance of the road, so that it looks narrower and
more constricted. This will help reinforce the signed speed limits. Two typical cross-sections are
recommended as follows, see figures 4.7 and 4.8. See section 4.4 for details of variations on
these typical cross-sections.
Class T1:
This is a single carriageway road with traffic lanes 3.0 to 3.25m wide. There are raised, kerbed
footways for pedestrians.
Outer hard
Road type Traffic lanes Footways Median Total width
shoulders
Class TI 3.0 3.25 m none >=2.0 m - 10-10.5 m
F o o tw a y C a rria g e w a y F o o tw a y
> = 2 .0 m 6 - 6 .5 m > = 2 .0 m
Figure 4.16 Typical cross-section for single carriageway roads in built-up areas
Page 41
Revision)
Class T2:
This is a dual carriageway road with traffic lanes 3.25m wide (speed limit > 50km/h), or 3.0m
wide (speed limit = 50km/h). The outer hard shoulders are 1.0m wide (speed limit >50km/h) or
0.5m wide (speed limit = 50km/h). The median should have a minimum width of 2.0m, but this
can be reduced to an absolute minimum of 1.2m width where space is very restricted. There are
raised, kerbed footways on both sides of the road.
Outer hard
Road type Traffic lanes Footways Median Total width
shoulders
Class TII 3.0 - 3.25 m 0.5-1.0 m 2.0 - 3.0 m 2.0m varying
F o o tw a y tr a ffic la n e s M e d ia n F o o tw a y
tra f fic la n e s
m in im u m 2 m 6 - 6 .5 m m in im u m 2 m
2 .0 m 6 - 6 .5 m
h a r d s h o u ld e r h a r d s h o u ld e r
0 .5 - 1 .0 m 0 .5 - 1 .0 m
Figure 4.17 Typical cross-section for dual carriageway roads in built-up areas
4.4 Detailed Design of Cross-section

4.4.1 Traffic lanes
The traffic lane width should preferably be 3.0 to 3.5 m on rural roads. Traffic safety studies
support 3.5m lane widths, especially on high speed roads. Traffic lane widths should be as
narrow as possible in urban areas to discourage speeding.
The 0.10m edge line road markings should be located adjacent to the traffic lane on a 0.2m wide
edge strip. The edge strip is rotated with the traffic lanes in curves. The 0.10m centre line is
located on the traffic lanes.
On two-way single carriageway roads the carriageway should be cambered, i.e. it should slope
down to the shoulders from the centre line. Advice on the steepness of the slope, i.e. the crossfall
is given in Table 4 .22.
Pavement type Crossfall %
Earth 5
Gravel 4
Bituminous seal 3
Bituminous concrete 2.5
Table 4.22 Recommended crossfalls

(Adapted from Austroads)
Page 42
Revision)
4.4.2 Shoulders
Shoulders are used for a number of purposes such as:
to be used by pedestrians and cyclists
space for edge line markings, minimum 0.2 m
safety margin to enable drivers to recover control
moisture egress barrier
support the bearing capacity of the travelled way
space for broken-down vehicles.
It is recommended that all shoulders be paved, though exceptions may be made for very low
volume roads. Paved shoulders are a safety feature and they also contribute to structural integrity
and lower maintenance costs.
Shoulders intended for use by pedestrians and cyclists must be paved and be at least 1.5m wide
(absolute minimum 1m wide). Where there is a lot of pedestrian and cycle traffic the shoulder
may be widened to 2.0 metres, but it would be much safer to provide a separate
footway/cycleway. Use of a different surfacing can make the shoulder more visible, and this
helps drivers to avoid straying onto it by mistake. However, do not make the surface of the
shoulder rougher than that of the traffic lane, because this will discourage pedestrians and
cyclists from using it.
Paved shoulders should normally have the same slope (crossfall) as the traffic lanes and should
be superelevated with the traffic lanes in curves.
4.4.3 Footways and cycleways

The conventional practice is to assume that pedestrians and cyclists can use the shoulders, but it
is much safer for them to be on a separate footway, or combined footway/cycleway. At high
flows there can be conflicts between cyclists and pedestrians, but these are not as dangerous as
conflicts with motor vehicles. Combined footways/cycleways should be 3.0m wide (2.0m
absolute minimum). It is important for footway and cycleway surfaces to be at least as smooth as
the adjacent traffic lanes and shoulders preferably smoother. See also sections 8.1 and 8.2.
Footways and cycleways in rural areas

The footway/cycleways should be separated from the carriageway by a grass strip or similar, at
least 2.0m wide. Gradients should be gentle, preferably less than 4%. On embankments the
footway/cycleway can be benched into the fore slope. The planting of shade trees can help
encourage people to use the facility.
Page 43
Revision)
c a rria g e w a y
F o o tw a y / s h o u ld e r tr a ffic la n e tr a f fic la n e s h o u ld e r
c y c le w a y F o o tw a y /
c y c le w a y
d iv id in g g r a s s
s trip o r s im ila r
Figure 4.18 Separate foot and cycleway on rural roads
Footways and cycleways in built-up areas

Raised, kerbed footways should be provided in the larger built-up areas. Cycleways, where
necessary, should be constructed behind the footway.
F o o tw a y C a rria g e w a y F o o tw a y
> = 2 .0 m 6 - 6 .5 m > = 2 .0 m
Figure 4.19 Raised, kerbed footway in urban areas
A simpler and cheaper alternative is to have the footway at the same level as the traffic lane, but
separated by a barrier kerb or low wall see Figure 4 .20. This means it can function as a
combined footway/cycleway. Gaps are left in the separator to allow drainage and access to
roadside premises. The separators should be painted white to make them more visible at night,
and care should be taken to avoid starting the separator where speeds are high or visibility is
poor. If necessary, fit reflectors to the end of the separator.
F o o tw a y /c y c le w a y C a rria g e w a y F o o tw a y /c y c le w a y
> = 2 .0 m 6 - 6 .5 m > = 2 .0 m
se p a ra tin g s e p a ra tin g
k e r b o r s im ila r k e r b o r s im ila r
Figure 4.20 Footway/cycleway on physically separated shoulders
Page 44
Revision)
4.4.4 Service roads

In the larger trading centres and towns it is recommended that service roads be provided. A
typical service road design is illustrated below. The local access traffic is kept separate from the
through traffic, and the service road provides space for parking, unloading and loading, bus
stops, and informal trading.
F o o tw a y F o o tw a y S e rv ic e ro a d
> = 2 .0 m M a in ro a d > = 2 .0 m ty p ic a lly 6 .0 m
Shop
D ra in D ra in
Figure 4.21 Service roads
4.4.5 Medians on dual carriageways

The median includes the inner shoulders and the central island. It separates the traffic directions.
The main functions of the median are:
to separate opposing flows of traffic thus reducing the number and severity of head-on
collisions
to provide a refuge for pedestrians crossing the road - mostly in urban areas
to provide protected space for right turn bays, traffic signs and lighting columns.
Refer to Figure 4 .22 for recommendations on median design and width on high-speed rural dual
carriageway roads. Where the median is less than 9.0m wide consider installing a median barrier
see section 9.2.9.
Medians on urban dual carriageways should normally be designed to function as a refuge for
pedestrians. The median should have a minimum width of 2.0m, but this can be reduced to an
absolute minimum of 1.2m where space is very restricted. A 2.0m width will also give sufficient
space for most signs, signals and lighting columns. If a protected right-turn lane is to be provided
the median will have to be at least 4.2m wide (3.0m for the traffic lane and 1.2m for the traffic
island). Median barriers should not normally be necessary on urban dual carriageways with
speed limits of less than 80km/h.
Page 45
Revision)
If e m b a n k m e n t
M e d ia n w id th
s p e e d lim it d e s ira b le m in
100 9 6
80 6 4
In n e r h a rd
s h o u ld e r
Figure 4.22 Median designs at speed limit 80 and 100 km/h
4.4.6 Cross-section over bridges and culverts

The design of bridges and culverts is outside the scope of this Manual, but it is obvious that
safety and capacity will be affected if the road cross-section is not maintained across these
structures. The key points to consider are:
Any significant narrowing of the traffic lane or shoulder is dangerous, especially on high-
speed roads;
When roads are being upgraded and widened, the bridges and culverts will normally need to
be widened as well;
If the shoulder is not continued across the structure, vulnerable road users will move out into
the traffic lane in front of fast-moving vehicles and there will be a risk of collisions;
Footways are conventionally provided on structures with parapets, but the accident risks need
to be assessed carefully, especially where footways can only be provided by omitting the
shoulders; take account of the relative volumes of pedestrians and cyclists, the speed and
volume of motorised traffic and the length of the span;
Where footways are provided they should be a minimum of 1m wide, and they will usually
be quite high (150-200mm) in order to better protect the bridge parapet, in which case the
ends must be stepped or ramped down and flared away from the edge of the traffic lane;
It is best to segregate the vulnerable road users from vehicles by means of a safety barrier
see Figure 4 .23 below and this is the preferred solution for bridges on Class A and Class B
roads;
Where, exceptionally, a single lane bridge is planned the traffic lane should be a maximum
of 3.7 m wide between kerbs in order to avoid confusion over whether the bridge is for one-
way or two-way traffic.
See section 9.4 for advice on the design of bridge parapets
Page 46
Revision)
P e d e s t r ia n
p a ra p e t R e in f o r c e d c o n c r e t e
s a fe ty b a r r ie r
F o o tw a y
1m
2 m m in im u m C a rr ia g e w a y
Figure 4.23 Segregated footway on bridges
(Note: if cyclists are likely to use the footway increase the parapet height to 1.4 m)
4.5 Roadside Area Design

4.5.1 Clear zone
The clear zone is a safety zone adjacent to the traffic lanes. It provides space for a driver to
recover control of his vehicle if he is in danger of running off the road. The clear zone must be
kept free of rigid objects (such as posts, trees etc with a diameter over 0.10m) and other hazards,
such as steep slopes, open drains, etc. If the hazard cannot be removed from the clear zone it
should be shielded by safety barrier.
There are a number of empirical studies in Europe and US indicating major safety benefits from
clear zones, see Figure 4 .24. It is obvious that the need for clear zones increases with speed and
curvature.
The following clear zone widths, measured from the edge of the traffic lane, are considered to
give an acceptable standard of safety. Traffic volume is also a factor, as, generally, the higher the
traffic volume the greater the frequency of run-off-road incidents which supports the use of
wider clear zone widths.
Speed Standard
limit Desired Minimum
70 5m 3m
80 6m 4m
100 9m 6m
Table 4.23 Clear zone widths
Foreslopes steeper than 1:3 cannot be counted as part of the clear zone because they are too
steep. Slopes that can be traversed safely by out-of-control vehicles need to be at least 1:4 or
gentler. Slopes between 1:3 and 1:4 are marginal; the normal practice is that half the width of
these slopes is counted as part of the clear zone see Figure 4 .25.
Page 47
Revision)
100 C u m u la tiv e fre q u e n c y (% )
50
S w e d a ta w ith o u t fix e d o b je c t
30
S w e d a ta w ith fix e d o b je c t
G M te s t d a ta
5 10 15 20 25
F in a l lo c a tio n fr o m c a r r ia g e w a y ( m )
Figure 4.24 Empirical data on the final position of vehicles that have run off the road
A v a ila b le c le a r z o n e :
1 .5 + 4 .0 + 0 .5 * 3 .0 + 2 .0 = 9 m
1 :2
1 :4
1 :3
1 .5 1 .0 4 .0 3 .0 2 .0
Figure 4.25 Example how to calculate clear zones
The clear zone widths given in Table 4 .23 should be increased at sharp bends on high-speed
roads refer to Figure 4 .26.
4.5.2 Side slopes

The design of side slopes is of great importance for traffic safety, as this can determine whether
drivers that run off the road are able to recover control. The shallower the slope, the safer it will
be. And the transition from the shoulder to the foreslope must be smooth enough to prevent the
vehicle becoming airborne. A safe transition is also needed between the foreslope and the back
slope so as to avoid causing the vehicle to rollover.
Page 48
Revision)
Figure 4.26 Clear zone correction factor for bends
Figure 4 .27 shows the principles of smooth roadside area design. The foreslope should be 1:4 or
gentler. The transition at the top and toe of the slope should be smooth. The height difference H
between the shoulder and the support strip should not be more than 50mm
The backslope design in cuts with a cut drain should be designed with a 0.5m wide ditch bottom
followed by a 1:4 back slope for half a metre and then a 1:2 slope for 2.0m; this will help to
redirect a run-off vehicle to the roadside area.
C le a r z o n e w ith s lo p e s s te e p e r
C le a r z o n e w ith s lo p e s s te e p e r
- th a n 1 :3 e x c lu d e d - th a n 1 :3 e x c lu d e d S b
T ra ffic la n e - 1 :3 - 1 :4 in c lu d e d w ith T r a f fic la n e - 1 :3 - 1 : 4 in c lu d e d w ith
0 .5 * th e w id th 0 .5 * th e w id th 1
B a ck slo p e
H S f > = 4 if s p e e d
1 :2
lim it > = 8 0 k m /h
S f > = 4 if s p e e d lim it> = 8 0 H a rd s h o u ld e r 1
H a rd s h o u ld e r 1 F o re s lo p e
F o re s lo p e S u p p o rt s tri p 1 :4
S u p p o rt s trip H in g e p o in t
H in g e p o in t T o e o f s lo p e
>= >= >=
T o e o f s lo p e m in im u m b o tto m w id th 1 .0 m
0 .5 m 0 .5 m
0 .5 m a n d b a c k s lo p e 1 :4
Fill design Cut design
Figure 4.27 Principles of smooth road side area design on fills and in cuts
When the embankment (fill) height is greater than about 3.0m, the 1:4 foreslopes recommended
above become uneconomic. This is because a large amount of fill material will be needed and the
structure will extend over a large area thus increasing land acquisition costs. In these
circumstances the foreslope is best determined by the natural angle of repose and erodibility of
the material (often 1:1.5). Where steep foreslopes have to be used, consider installing safety
barrier see section 9.2.3.
Page 49
Revision)
The area in front of roadside obstructions such as bridge abutments, retaining walls, etc. should
be smooth, with a maximum change of deflection (w) of 24.5 degrees. Ensure that the lateral
clearance is adequate, and that the obstruction is outside the clear zone. If it is not, consider
installing a safety barrier.
W id th d u e to s ig h t d is ta n c e s ,
la te ra l c le a ra n c e s e tc
V e r tic a l
a b u tm e n t
w
S h a llo w c h a n g e o f
d e f le c tio n w < 2 4 .5 d e g r e e s
Figure 4.28 Design of area in front of obstructions
4.5.3 Side roads and culverts

Side roads are often built up on a little embankment so that they enter the main road on the same
level. This embankment can be an obstacle to vehicles that run off the road. And the culvert
carrying the main road drain under the side road will often have a large, solid headwall. Where
there is a culvert under the main road, the culvert headwall is often close to the edge of the
carriageway, especially if the road has been widened at some stage. These are hazards. With side
roads it is best to try and construct gentle embankment slopes and move the culvert further away
from the main road. In the case of the culvert under the main road it should be extended in order
to move the ends away from the carriageway edge. It is also important to assess whether culverts
really need large solid headwalls. It may be possible to provide a smooth opening instead. See
Figure 4 .29 and Figure 4 .30.
Figure 4.29 Access road design
Page 50
Revision)
Figure 4.30 Culvert end design
4.5.4 Side drains

Good drainage is essential to protect the road from damage. In rural areas the drainage often
consists of an open ditch parallel to the road, with culverts at intervals to discharge the flow into
local watercourses. Refer to the Hydrology and Drainage section of this Manual for advice on
designing drains that will cope efficiently with the estimated flow of water. Note also that, unless
carefully designed, these side drains can be a major hazard for out-of-control vehicles that run
off the road. Key points to consider in the design of safe side drains are:
There should be sufficient discharge points and culverts to ensure that the drain never gets
very deep
Open drains are best located outside the clear zone
With open drains, the slope next to the road should be flat enough to reduce the risk of errant
vehicles overturning (maximum gradient of 1:4, see section 4.5.2)
Open lined drains should be in the form of shallow dishes rather than steep-sided U or V-
sections (see Figure 4 .31)
In built-up areas channel drains deeper than 250mm should be covered for the safety and
convenience of both pedestrians and vehicles
The drain should terminate or discharge in a satisfactory manner without risk of causing
erosion or other problems
The drain should be capable of being cleaned and maintained easily.
It is not always easy to design drains that can cope with the expected flow and yet are safe,
affordable and easy to maintain, so compromises are often required.
C o n v e n tio n a l V e e -s h a p e d d r a in T r a p e z o id a l d r a in
L in e d w it h
s t o n e r ip - r a p
1 :1 1 :4
1 :1 1 :4
( 1 : 4 if d r a in is
( 1 : 4 if d r a in is
d e e p e r th a n 1 5 0 m m )
d e e p e r th a n 1 5 0 m m )
C h a n n e l d r a in D is h d r a in
D e s ir a b le m a x . d e p th o f 1 5 0 m m S u ita b le f o r b u ilt- u p a r e a s
Page 51
Revision)
Figure 4.31 Safe side drains
Page 52
Revision)
5 ALIGNMENT
5.1 General
The alignment is defined as the combination of horizontal and vertical geometric elements giving
the location of the road in the terrain. The horizontal elements used are circles defined by the
radius R, straights (tangents) and transition curves, normally clothoids. The vertical elements
used are grades and vertical radii R.
The alignment, the cross-section, road furniture, vegetation and the surroundings create the
framework for driver perception and behaviour such as choice of speed, choice of track and
lateral position on the road. The main design decision for alignment requirements is the speed
limit by sections of the road and associated design speeds. This design procedure is described in
section 3.3.
The basic assumption for road alignment is that the driver at design speed should be able to
perceive any possible road hazard on or close to the road to take action to avoid mishap. This
requires all the alignment to have good visual guidance and sufficient stopping sight distances,
see section 3.3.3. There is also a need to be able to overtake slower vehicles. Parts of the road
therefore require overtaking sight distances, see section 5.3.
It is judged important for traffic safety reasons to avoid, as far as possible dilemma sight
distances, i.e. in between stopping and overtaking sight distances, where the driver cannot see
quite far enough to be sure that it is safe to overtake. The driver should also be able to drive
comfortably due to inertia forces at the design speed relevant for horizontal alignment and sags.
These basic assumptions give minimum parameters for vertical and horizontal alignment
elements, see section 5.4. They also give a number of recommendations how to combine
elements to facilitate visual guidance and safe driving. A basic rule is to avoid sharp bends after
long tangents and other surprises to the driver.
The alignment and the cross-section should also be adapted to the surroundings to create a
stimulating driving task with a good rhythm, harmony and with varying views, trying to
maximise the length of road with sufficient overtaking sight distances and also obviously to give
a cost-effective design optimizing road user, investment and maintenance costs and other
important social and environmental impacts.
5.2 Alignment choice and terrain adaptation

5.2.1 Introduction
To locate a road in a landscape is a challenge with constraints and possibilities. It is a technical
and also an architectural process.
The main principle is to adapt the road to the surroundings considering technical requirements on
sight distances such as sufficient overtaking possibilities, minimum geometric elements and
visual guidance not to adorn or to emphasize. Three basic concepts unite and constitute the
technical and the architectural process:
Scale
Space
Rhythm.
Page 53
Revision)
The scale is the size of the landscape to locate the road in. Some typical landscape sizes are
illustrated in Figure 5.1.
Large-scaled landscape
Small-scaled landscape
(illustration from Swedish design guideline)
Figure 5.32 Description of the scale concept
Landscapes can be differentiated in types such as:
Flat landscape
Page 54
Revision)
Slightly hilly landscape
Hilly landscape
Figure 5.33 Examples of landscape types
Space limited by ridges and trees The eye view angle decreases with
speed
Speed
9 9
15 1 0 0 k m /h 15
20 20
8 0 k m /h
29 29
6 0 k m /h
37 37
S p a c e b u ild in g
Figure 5.34 Drivers space or room concept
Page 55
Revision)
The space or room is a defined part of the landscape as far as you can overview from a specific
point. The limitations of the space or room could be:
Terrain (mainly topography), vegetation, buildings
Road design, i.e. cross-section, horizontal and vertical alignment
Crossing bridges and road embankments.
The rhythm of a trip along a road the experience and enjoyment depends on the design and
how this design is located in the landscape. The designer should use the landscape combined
with the road alignment, cross-section and road side area to create a variation, rhythm, in
impressions and outlooks. The objective is to create a road that is enjoyable to drive.
Figure 5.35 Some examples on rhythmical landscape adaptation
The alignment should together with the cross-section, the roadside area, and the surroundings
create variation in outlooks for the driver and also support him in his driving task with visual
guidance. Outlooks should be long enough to be comprehensible at the design speed. A simple
rule of thumb is that outlooks should have at least the same length in metres as the design speed
in km/h representing some 4 to 5 seconds driving time.
The alignment is three-dimensional. It is of utmost importance to look at and treat the alignment
design as a space curve following as far as possible the laws of perspectives.
The road should have an inner and an outer harmony. The inner harmony means that the road
should have a satisfying, calm and graceful geometric form considered only as a space curve
without terrain. The outer harmony requires the space curve to be tuned with the terrain and in
harmony with the landscape. The geometric elements should have the same scale as the
surrounding terrain.
No harmony Harmony
Figure 5.36 Example of adaptation to the landscape
Page 56
Revision)
5.2.2 Methodology
The practical work with alignment choice and terrain adaptation can be divided into three steps:
Inventory of constraints and opportunities
Route planning
Detailed design of the alignment
Step 1: Inventory
The workshop for the road and the road designer is the terrain area. It is important to analyse the
terrain to understand its constraints and possibilities.
Constraints could be areas not allowed to be used or only to be used as constraints, e.g. existing
or planned buildings, rivers, roads, geotechnically difficult areas etc. Possibilities to create the
desired right scale and rhythm are the talent of the designer to combine the technical
requirements with the freedom given by the terrain area. It is important in the initial phase to get
good knowledge and a visual concept about the terrain. The terrain should be walked. Good
maps, terrestrial and aerial photos are essential but it should be stressed again that the solution of
the road design is found in the terrain.
STEP 2: Route planning

There are two alternative methods for identifying the route: the tangent method and the arcs
method.
Tangent method (see Figure 5.6)
The straight strategy gives the road line iteratively by defining straights and then to combine
these by arcs. An advantage could be effective use of sight distances.
Page 57
Revision)
Figure 5.37 Route planning by tangent method

Arc method (see Figure 5.7)
The other method starts by selecting arcs with radii that fit in well with the scale and form of the
landscape. These arcs are then linked together with transition curves, i.e. clothoids of larger arcs.
It is important in this strategy to avoid dilemma sight distances and to create sufficient
overtaking sight distances, see Section 5.1. This method is more likely to result in a route which
is well adapted to the terrain.
The road location and alignment procedure should start on the map by sketching suitable
alignment alternatives. Impressions and notations from the landscape and its characteristic forms
and properties should be used to create alternatives anchored in the landscape. The straight line is
normally not in harmony with the landscape.
The terrain adaptation must, as already stressed, be combined with the partly contradictive
requirement to create sufficient overtaking sight distances and to avoid dilemma sight distances.
Left-hand bends, even with large radii, will rarely have sufficient sight distance for safe
overtaking.
The designer should mentally visualize the three dimensional form. This could be supported by
using simple profile sketches to analyse the phasing between horizontal and vertical alignment.
Intersections, interchanges etc should be considered already in this stage.
Figure 5.38 Route planning by arc method
Page 58
Revision)
STEP 3: Detailed design

Having found a route in harmony with the terrain using the sketch technique above horizontal
geometric elements can be calculated and a first profile produced. Technical requirements should
also be checked such as:
minimum horizontal and vertical elements
combined elements for visual guidance
stopping, dilemma and overtaking sight distances
speed profiles.
Always check what percentage of the road will have overtaking sight distance. This will have a
major impact on safety and level-of-service at medium high traffic flows - see section 3.3.3.
The coordination between vertical and horizontal alignment should be checked using perspective
images. It is important to learn how to select points for perspectives and how to interpret
perspective images. The perspective images could be used to judge if minor or major
adjustments are needed.
Phasing
Six basic forms for the space curve can be defined. There are a number of long-established rules
on how to combine these elements in different terrain situations. Some of these are summarised
here.
G e o m e try
S p a c e e le m e n t
H o r iz o n ta l V e r t ic a l
S t r a ig h t w it h c o n s t a n t g r a d e
H o r iz o n t a l
or
c o n s ta n t g r a d e
S t r a ig h t in d e c lin e
C oncave bend
S t r a ig h t sag
S tr a ig h t o n c r e s t
C onvex bend
c re s t
B e n d w ith c o n s ta n t g r a d e
H o r iz o n ta l
or
c o n s ta n t g r a d e
B e n d in d e c lin e
C oncave bend
A rc sag
B e n d o n c re s t
C onvex bend
c re s t
Page 59
Revision)
Figure 5.39 Basic space elements
A straight, especially not a long one, should not be preceded by a small arc, see section 5.4.4. A
transition curve is recommended in this case. It could be a clothoid or a bigger arc with a radius
at least twice as big (R2/R1 < 2.0).
The clothoid gives a form that facilitates the driver to choose his lateral position in the lane in the
curve reducing short-cutting. It also gives a smoother alignment. The narrower the road the more
important is the use of transition curves to create lane designs used by the driver and to create a
harmonic road design adapted to the surroundings.
P R O F IL E
R
TP
S tr a ig h t
TP
R
A
C lo th o id
TP A
S t r a ig h t
Figure 5.40 Comparison of eye impression at straight arc and straight clothoid - arc
It is important that deviations between the elements of the alignment are large enough. The
minor change in direction the larger is the arc needed to avoid kinks in the alignment.
P R O F IL E
H O R IZ O N T A L R
TP TP
P R O F IL E
H O R IZ O N T A L R
TP
TP
Figure 5.41 Road with and without optical kink

It is also important that vertical elements arc lengths are large enough to avoid kinks, especially
for concave vertical curves (sags). The illustration in Figure 5 .42 shows a comparison of
transitions with from 2 % to + 3 % with a 210 m long vertical radius with R 4000 m and a 900
m long vertical curve with 18000 m radius
Page 60
Revision)
P R O F IL E
Rv
H O R IZ O N T A L
P R O F IL E
Rv
H O R IZ O N T A L
Figure 5.42 Comparison of alternative transitions from 2 % to + 3 %

Two consecutive arcs bending in the same direction connected with a short straight should be
avoided vertically as well as horizontally as illustrated in Figure 5.12.
P R O F IL E
H O R IZ O N T A L 1
S T R A IG H T
R2
S T R A IG H T
R1
S T R A IG H T
H O R IZ O N T A L 2
S T R A IG H T
S T R A IG H T
Figure 5.43 A short straight between two curves should be avoided
The combination of horizontal curves and vertical grades is critical, see Figure 5.13. A horizontal
curve could terminate anywhere on the vertical curve, case A in Fig. 5.13, and should preferably
start in the upper part, case D. A later start is acceptable, case E, and could be improved with a
transition curve case F. The start tangent point of a horizontal curve should be located on the
upper part of the first sag curve in the grade, not on the lower part.
A start of the horizontal curve before the sag, case B, could be interpreted as if the sag is a
horizontal contra curve in the large horizontal curve. A start of the sag in the beginning of the
horizontal curve creates a kink. A sag within a horizontal curve should have a large arc.
Page 61
Revision)
Figure 5.44 Illustration of combination of horizontal curves and vertical grades

Another critical design is the S-curve. The ratio between the horizontal radius R h and the vertical
radius Rv should be as small as possible, at least with Rh / Rv < 1/5 to 1/10. Moreover, the
changeover should be vertically and horizontally reasonably coordinated. The vertical curve
should be small or large in order to maximize overtaking sight distances.
P R O F IL E 1 w it h la r g e c o n v e x v e r t ic a l r a d ii
I1 I2 R v 3 ( la r g e )
R v 1 ( la r g e )
R v2
P R O F I L E 2 w i t h s m a ll c o n v e x v e r t i c a l r a d i i
R v 1 ( s m a ll) R v 3 ( s m a l l)
R v2
T u r n in g a r e a s
R3
H O R IZ O N T A L
A2
R1 A1 R2 RL
IN F
S p a c e e le m e n ts
H O R IZ O N T A L A N D P R O F IL E 1
Figure 5.45 Coordination of vertical and horizontal elements in turning area

Some dangerous alignment combinations to avoid are:
sharp horizontal curves after sharp crests the horizontal curve should start before the crest
to give visual guidance
Page 62
Revision)
consecutive horizontal curves with diminishing radii.

Bridges are important to coordinate in the chosen harmonic alignment. Straights should, if
possible, be avoided, see the two upper parts of Figure 5.15 below. Bridges in curves improve
the opportunity for drivers and passengers to experience and enjoy the visual impact of the
bridge.
Bridge not adapted to the road spline board effect
Bridge on straight and crest does not create any bridge experience
Bridge in horizontal curve improves the experience of the bridge
Page 63
Revision)
Figure 5.46 Examples of adaptation of bridge alignment

Note: the bridge parapet shown is indicative only -, see section 9.4 for advice on the design of
bridge parapets
Larger bridges, crossing rivers or valleys, should preferably be located and aligned to give the
driver a view of the bridge and not only the surface - see Figure 5.16.
Figure 5.47 Bridge alignment without and with a bridge view
5.3 Sight distances

Two sight distance requirements are defined:
stopping sight distance
overtaking sight distance.
Clearing obstructions to visibility, such as trees, bushes, old walls and buildings can be highly
cost-effective. On sloping ground it may be worth cutting into the slope to improve sight
distances. isolated slim objects such as lamp columns or sign supports of width 550 mm or under
can be ignored. Lay-bys should, wherever possible be sited on straights or on the outside of
curves, where stopped vehicles will not obstruct sightlines.
Page 64
Revision)
Stopping sight distance

The desirable and minimum stopping sight distances (SSD) are given in Table 5.1. Desirable
stopping sight distance should normally be available throughout the length of the road. The
minimum stopping sight distance can be accepted on short sections where the terrain or other
problems make it difficult to obtain greater visibility. Even lower sight distances can be
considered if combined with warning devices such as warning signs, local speed limit signs,
rumble strips etc.
Speed Stopping sight distance SSD

limit Desired Minimum
30 40 25
50 70 50
70 110 85
80 135 110
100 195 165
Table 5.24 Desired and minimum stopping sight distances on level ground
Stopping sight distance shall be measured from a drivers eye height of between 1.05 m and 2.0
m to an object height between 0.2 and 2.0 m above the road surface, as shown in Figure 5 .48. It
shall be checked in both the horizontal and vertical plane, between any two points in the centre
of the lane on the inside of a horizontal curve.
2 .0 0 m 2 .0 0 m
1 .0 5 m
0 .2 m
(from UK design guidelines)
Figure 5.48 Measurement of stopping sight distance
Actual stopping distances are affected by the road gradient, so the stopping sight distances must
be adjusted. Table 5.2 gives the stopping sight distances that apply where the road is on a grade.
Page 65
Revision)
Speed Grade
km/h -8 % -6 % -4 % -2 % 0 % 2% 4% 6% 8%
30 25 25 25 25 25 25 25 25 25
40 40 40 40 40 40 35 35 35 35
50 55 55 55 50 50 50 50 50 50
60 75 75 70 70 70 65 65 65 65
70 100 95 90 90 85 85 85 80 80
80 125 120 115 110 110 105 105 100 100
90 155 150 145 140 135 130 125 125 120
100 190 185 175 170 165 155 155 150 145
110 235 225 215 205 195 190 180 175 170
120 280 270 255 245 235 225 215 210 200
130 340 320 305 290 275 265 255 245 235
Table 5.25 Stopping sight distances on grades
Overtaking sight distance

The full overtaking sight distances (FOSD) needed for safe overtaking on single carriageway
roads are shown in Table 5.3. The Table also gives the visibility distances below which no
overtaking centre line markings should be used (refer to MOWHC Traffic Signs Manual). The
FOSD-values relate to a situation in which the overtaking vehicle is following another vehicle
and has to accelerate to overtake it. The sight distance for the use of the no overtaking
markings is based on a less demanding standard of safety in which the overtaking vehicle is
already at the speed of the overtaken vehicle, and the driver can abandon the overtaking
manoeuvre at any time up to the point where he has drawn alongside the other vehicle.
Sufficient visibility for overtaking shall be provided on as much of the road as possible,
especially where daily traffic flows are expected to approach the maximum design flow.
Visibility distance below

Speed limit FOSD which no overtaking centre
line markings shall be used
50 300 150
80 500 240
100 700 300
Table 5.26 Overtaking sight distances
Page 66
Revision)
FOSD shall be available between points 1.05 and 2.0 m above the centre of the carriageway as
shown in Figure 5.18 and shall be checked both in the horizontal and the vertical plane. When
determining the length of no overtaking centre lines the sight distance should be checked from an
eye height of 1.05 m to an object height of 1.3 m.
(adapted from UK design guidelines)
Figure 5.49 Measurement of full overtaking sight distance
FOSD is considerably greater than stopping sight distance, and the provision of FOSD can
normally only be economically justified in relatively flat terrain which permits the design of a
flat and relatively straight road alignment. It will frequently be more economic to design single
carriageway roads so as to provide clearly-identifiable overtaking sections with FOSD in
relatively level areas, interspersed with clearly-identifiable non-overtaking sections in more
difficult terrain. It is important to avoid creating sections where the sight distance is not quite
enough for safe overtaking i.e. dilemma zones.
Calculating the lateral clearance (offset) needed to achieve sight distances on

curves
Sight distances can be checked as outlined below. The principle is to compare the required and
existing lateral offset with regard to the sight obscuring terrain height h over the road surface as
illustrated in Figure 5 .50. The method presumes that the element arc lengths are longer than the
required sight distance. The result is on the safe side if the arc is shorter.
h
s ig h t lin e a
Figure 5.50 Offset and sight distance
Page 67
Revision)
Calculation procedure
Step 1: Select what required sight distance (S) is to be checked
Step 2: Required offset (a1) for actual horizontal radius Rh and sight distance S is given in
Figure 5 .51 for stopping sight distance and Figure 5 .52 for overtaking sight
distance
Figure 5.51 Required offset at various horizontal radius and stopping sight distances
a1
30
25
20
15
10
S=900
5 S=500 S=700
S=350
0
0 5 10 15 20
R/1000 S I DVN. XLS
Page 68
Revision)
Figure 5.52 Required offset at various horizontal radii and overtaking sight distances
Step 3. Terrain height, h, for actual vertical radius, R v, and sight distance is given in Figure 5 .
53 for stopping sight distance with object height 0.2 m and in Figure 5 .54 for
overtaking sight distances
Figure 5.53 Terrain height for stopping sight distance
Figure 5.54 Terrain height (h) for overtaking sight distances

Page 69
Revision)
Step 4: Existing offset (a2) can be calculated according to Figure 5 .55 below
a
2
h
c
b
a = b + c * m + c * n + h *n
2
b = f r o m c e n t r e o f t h e in n e r t r a f f ic
la n e f o r s t o p p i n g s i g h t d i s t a n c e
fr o m t h e c e n t r e o f t h e r o a d fo r
o v e r t a k in g s ig h t d i s t a n c e
Figure 5.55 Available offset (a2)
5.4 Horizontal alignment

The horizontal alignment is built of tangents and arcs. Transition curves (spiral curves) are
conventionally used to link tangents with arcs.
5.4.1 Tangents
Tangents or straights are defined by their start and end points and the bearing. Long tangents are
monotonous and likely to introduce boredom and reduce safety. It is also difficult to judge the
speed and distance of on-coming vehicles. Short tangents will also reduce safety and are likely to
result in an awkward-looking kink in the alignment.
5.4.2 Camber and superelevation
Tangent sections on single carriageways should normally be cambered in order to facilitate
drainage. This means that the carriageway surface should slope down (usually at 2.5%) from the
centre line to the edges. At curves the camber should be gradually replaced by a single slope that
falls towards the inside of the curve (called superelevation) in order to make it easier to drive
through the curve. Alternative strategies for camber on tangents and rotation to superelevation in
curves are shown in Table 5 .27.
T w o - la n e t w o - w a y
O n e - w a y o n e - la n e T w o - o r m u lti- la n e o n e - w a y
T w o - la n e o n e - w a y
C am ber on
ta n g e n t
N o rm a l
r o t a tio n
fo r
s u p e r e le v a t io n
A lt e r n a t iv e
r o t a tio n s
fo r
s u p e r e le v a t io n
Table 5.27 Rotation axle alternatives
The dynamic force system on a car driving at speed V in a super elevated curve is shown below.
Page 70
Revision)
N = normal force
F = centripetal force
G = inertia
fs = side friction coefficient
= super elevation or
camber
Figure 5.56 Dynamic force system in curve

E = superelevation or camber (tan )
2
V V = speed (m/s)
fs E
gR
g = 9.82 (m/s2)
R = horizontal radius (m)
The superelevation in curves is recommended to be chosen as one of the following alternatives:

adverse camber -2.5 %
minimum +2.5 %
4%
maximum 5.5 % (in special cases such as loop ramps 7 %)
Adverse camber causes driver discomfort and should be removed on all curves of less than 2500
m radii (speed limit 80 km/h) and 4000 m radii (speed limit 100 km/h). Superelevation is not
needed on very large radius curves, but it may sometimes be provided for aesthetic reasons.
Superelevation is decided due to speed limit and curve radius, see table 5.5 below.
5.4.3 Curve radii

Some of the factors to be taken into account when determining curve radii and arc lengths are:
traffic safety
level of service and vehicle dynamics
visual guidance and overtaking sight distances
terrain adaptation and aesthetics
costs and intrusion
Once curve radii fall below 1000 m (speed limit 100 km/h) or 800 m (80 km/h) the accident risks
start to increase rapidly. The critical value is usually around 600 m. Consistency of successive
curves is another important factor affecting safety. If you cannot avoid using a very sub-standard
curve, try and put in a mid-range radius curve before it, to get drivers to start slowing down. You
must ensure that drivers do not have to slow down sharply (more than 25km/h) in order to safely
negotiate the curve.
Table 5 .28 gives the minimum horizontal radii at that can be used in order to provide good, safe
driving conditions at 2.5, 4 and 5.5 % superelevation, with enough friction available for
emergency stops. Larger radii than this should be provided where possible.
Page 71
Revision)
Radius
Speed Environment Super Desirable Minimum
(km/h) elevation
(%)
50 Urban 2.5 160 100
4 140 90
80 Rural 2.5 700 500
4 400 300
5.5 370 270
100 Rural 2.5 800 600
4 700 530
5.5 600 500
Table 5.28 Desirable and minimum horizontal radii for safe driving
Formula:
Rv2/(0.28*exp(-0.03456V)+E)*g
v=design speed (m/s)
V=design speed (km/h)
E=super elevation (%, - adverse)
g=earth inertia
(adapted to the speed limit and design concept from Kenyan, Swedish, and German design guidelines, similar to UK, Australia and SATCC)
Very large radii will be needed if it is necessary to maintain overtaking sight distances. Table 5 .
29 shows the minimum radii required to achieve sight distances on a right-hand curve. Table 5 .
30 shows the same for left-hand curves.
It is advisable to avoid very short arcs, because they tend to create awkward-looking kinks in the
alignment. Arc lengths should preferably correspond to at least 5 seconds driving time at the
speed limit (3 secs absolute minimum) - see also section 5.2.
Radius
Sight distance Lateral clearance (edge of traffic
(m) lane to back slope)
4.5 m 5m 6m
350 1500 1400 1300
500 3100 2900 2700
700 6100 5700 5300
900 10100 9300 8700
Table 5.29 Minimum radii required to maintain overtaking sight distance on right-hand
flat curves
Page 72
Revision)
Sight Radius
distance
Vehicle type being overtaken
Car 12 m truck 18 m semi trailer
350 3500
500 4500 7000 9000
700 8000 13000 17000
900 11000 18500 26000
Table 5.30 Minimum radii required to maintain overtaking sight distance on left-hand flat
curves
5.4.4 Transition curves

Transition curves are used to:
achieve a smooth change of direction between arcs with different radii or between
tangents and arcs
encourage drivers to follow a dynamically correct lane track, without short cutting
create a smooth and visually-harmonic alignment.
Transition curves also provide a convenient place on which to transition between camber and
superelevation and back again.
The dynamics that make transitions necessary increase with decreasing road width and radius
and increase with speed. Transition curves are recommended wherever the resulting radius Rr is
below the values in table 5.8 below.
1
Rr
1 1

R1 R2
with R1 < R2 with R2 = for straights and inflexion points giving Rr = R1.
Speed Resulting radius (m)

(km/h) Desirable Minimum
80 500 400
100 750 650
(adapted to the speed limit and design concept from Kenyan and Swedish design guidelines)
Table 5.31 Minimum values for resulting radius Rr without transition curve
Clothoids
The clothoid is the name given to one form of transition curve. This is the most commonly used
form of transition in highway design and is defined mathematically as follows:
Page 73
Revision)
A2 = R L
where A = clothoid parameter (m)

R = final radius (m). Start point in the straight tangent with R=
L = clothoid length (m)
A = R L
R
Ym R
L
R
Y
R = S ta rt ta n g e n t
I n f le x io n p o in t Xm Tk
Tl
X
R = f in a l r a d iu s X m = a b s c is s a o f c e n tr e p o in t
R = c ir c u l a r s h i f t Y m = c e n t r e Y - c o o r d in a t e t
= d ir e c tio n c h a n g e X = f in a l c l o t h o i d e X - c o o r d in a t e
Tl = lo n g ta n g e n t Y = fin a l c lo th o id e Y - c o o r d in a te
Tk = s h o r t t a n g e n t Figure 5.57 Clothoid elements
The parameter of the clothoid and the length of the transition curve recommended are decided by
the rate of increase of centripetal acceleration (m/sec 3) k, see formula below. K should normally
be in the interval 0.3 to 0.6 with a standard value of 0.45 m/sec3.
A v3 / k
where A = clothoid parameter (m)
v = design speed (m/s)
k = rate of increase of centripetal acceleration, (0,45 m/s3)
(adapted from Swedish and German design guidelines)
The recommended transition length should apply to the following requirements:

at least as long as the length needed for the change of superelevation, see section 5.4.5
not exceeding the rate of increase of centripetal acceleration
aesthetic balance between parameters and arc lengths
Minimum lengths due to rate of change of centripetal acceleration are given in Table 5.27.
Page 74
Revision)
(adapted to the speed limit and design concept from Kenyan, Swedish and German design guidelines)
Figure 5.58 Minimum length due to rate of change of centripetal acceleration

Aesthetic considerations recommend that
R/3 < A < R
0.1R < L < R
The first condition limits the change of direction of the transition arc to 3.5 to 32 gon. Using
circles as transitions 0.1 R should be used for small and R for large circles.
A simple clothoid could be used as transition curve between a tangent and a circle. The circular
displacement R should be at least 0.25 m to avoid kinks.
LR
A
R TP
R = S ta r t ta n g e n t
TP
Figure 5.59 Simple clothoid

The compound curve consists of two arcs with a connecting clothoid. The major circle must
surround the minor circle with separate centre points.
R2 > R1
c lo th o id e
L
TP
B a s e ta n g e n t
TP
Figure 5.60 Compound curve
Page 75
Revision)
The compound curve could be substituted with three circles with R1 < R2 < R3.
The S-curve is a combination of two clothoids between two reverse circles. The relationship
between the clothoid parameters could be controlled by aesthetic aspects on harmonic alignment,
differences in size of superelevation and design of superelevation transition. With A < 200 the
ratio A1 / A2 1.5 is recommended, see also X. A1 / A2 > 1.5 could give an unbalanced design.
R =
U 1 L1
=
U 2 L2

A1
A2
L1
TP B a s e ta n g e n t
TP TP
L2
C lo t h o id e s
Figure 5.61 S-curve
5.4.5 Superelevation run-off

The superelevation run-off is normally achieved by rotating the road along a longitudinal axle,
normally the profile line located in the road centre as shown in Table 5.4 above. Other locations
could be chosen due to drainage, driving comfort, aesthetics and visual guidance.
A safety problem with this transition is the risk of aquaplaning (tyres losing contact with the
surface due to a thin film of water) in S-curves due to insufficient drainage conditions in the
inflexion area. The longitudinal gradient of the road should be in the interval 0.5 to 3.0 % on
parts with super elevation below 2.5 %. The resulting grade should be at least 0.5 %.
Transition curves and super elevation run-off should be coordinated on the same length. Longer
transition curves are sometimes desirable due to aesthetic reasons than needed according to the
calculation procedure below. The run-off over +2.5 and -2.5 %, U 0, corresponding to E=5 %,
should be as short as possible to improve drainage conditions - see Figure 5.31.
Page 76
Revision)
F IG U R 6 .6 .3 - 1
H o r iz o n t a l a lig n m e n t A R
R =
S u p e r e le v a t io n
L e ft e d g e
C e n tr e lin e
R ig h t e d g e
Uo L e ft e d g e
S u p e r e le v a t io n
d ia g r a m
L e ft a n d rig th
edge R ig h t e d g e
L>=U
H o r iz o n t a l a lig n m e n t R2
A2
A1
R1
S u p e r e le v a t io n L e ft e d g e
C e n tr e lin e
E=0%
R ig h t e d g e
S u p e r e le v a t io n R ig h t e d g e L e ft e d g e
d ia g r a m
(adapted to the speedL limit and design concept from Kenyan, Swedish and German design guidelines)
e ft e d g e Uo Uo R ig h t e d g e
Figure 5.62 Transition due to drainage and aesthetics for tangent-clothoid-radius
Perspective images should be used to control superelevation run-offs to avoid undesirable visual
effects together with horizontal and vertical alignment. The result could be changes of the
location of the transition or the sizes of the elements. Bridges are especially important if located
on run-offs from positive to negative camber. The result of lifting the outer edge could be an
impression that the bridge has sunk. An option is to keep the outer edge constant and to lower the
inner edge instead.
The run-off length is controlled by

The speed limit and design speed chosen
The rotation width
The resulting superelevation difference.
Page 77
Revision)
The following procedure to decide on the run-off length (U) is recommended:
Step 1: Decide on the rotation width (Wr).

Rotation width is the width used by traffic at design speed to be rotated along the rotation axle.
S in g le c a r r ia g e w a y D u a l c a r r ia g e w a y
R o t a tio n a x le R o t a t io n a x le
W r W r
HS L L HS
M HS L L HS
Figure 5.63 Rotation width calculation
Step 2: Decide on superelevation difference (E).

The largest difference in superelevation is to be used in transition from normal to adverse
camber/super elevation.
Step 3: Calculate the superelevation run-off length (U) according to figures due to speed limit.
The run-off arc length (L) if chosen should be at least as long as the super elevation transition
length needed (U): L U.
Transition arc lengths longer than required superelevation transition length U+20 m with
resulting superelevation containing super elevation 0 % should be designed with minimum
length U0 for the interval 2.5% to minimize aquaplaning risks. The remaining superelevation
difference is levelled on the rest of the length.
Figure 5.64 Superelevation run-off length U speed limit 50
Page 78
Revision)
Figure 5.65 Super elevation run-off length U speed limit 80
Figure 5.66 Superelevation run-off length U speed limit 100
Superelevation run-off design

Superelevation run-off designs are recommended for the following cases:
Tangent clothoid - circle
S-curve
Compound curve
Tangent circle
Page 79
Revision)
Two circles.
Transition from camber to superelevation on the combination tangent-clothoid-circle is normally
performed over the whole clothoid. The run-off should be split in two parts if U+20<L with
transition from -2.5 % to + 2.5 % on minimum length U 0, see figures above, and the rest on the
remaining part of the clothoid U1 = L- U0.
D e s ig n r o a d c e n tr e lin e
R o a d s u r fa c e c e n tr e lin e
R
A
U1 A 2 = R * L w ith L > = U
U + 2 0 < L ; U = U 0+ U
L 1
U U +20>=L; U =L
U0
A
RL
Figure 5.67 Superelevation transition design on tangent clothoid - circle
The S-curve circle clothoid clothoid circle have two alternative designs:
propeller
displaced zero-points.
as shown below
D e s ig n r o a d c e n t r e l i n e D e s ig n r o a d c e n tr e lin e
R o a d s u r fa c e c e n tr e lin e R o a d s u r fa c e c e n tr e lin e
R1 R1
Aa Aa
U1a U1a
La La
Ua Ua
U0 U0
0% 0% A a A a
A b A b
U0 U0
Lb Ub Lb
Ub
2
U1b U1b
A a = R 1* L a L a> = U 1a
2
A b A b
A b = R 2* L b L b> = U 1b R2 R2
Propeller design Displaced zero-points

Figure 5.68 Alternative S-curve superelevation transition design
Page 80
Revision)
The choice should be made due to aquaplaning risks, driving comfort, aesthetics and visual
guidance. The figure below gives contours and longest theoretical drainage lengths for a 9 m
wide minimum superelevation transition in 1 and 5 % grades using propeller and displaced zero
points. Note that:
The drainage length increases with longitudinal grade
The displaced design gives, at 1 %, less total water area, less depth and shorter drainage
lengths, at 5 %, larger water area but shorter drainage lengths.
1 % longitudinal grade
propeller
displaced
1 % longitudinal grade
propeller
displaced
(adapted from Swedish design guidelines)
Figure 5.69 Drainage tracks at 9 m rotated width with propeller and displaced design
Superelevation transitions on compound curves should be done rectilinear over the whole
transition arc.
R1
A
A
R2 D e s ig n r o a d c e n t r e lin e
R o a d s u r f a c e c e n t r e lin e
Figure 5.70 Super elevation run-off on compound curve.

It is recommended to superelevate the major curve even if large enough for camber.
Page 81
Revision)
The superelevation run off at the combination tangent circle is recommended to be located to
the tangent part of the curve. At 4 or 5.5 % super elevated curves could the part of the transition
with 4.0 % or higher superelevation be located on the curve.
D e s ig n r o a d c e n tr e lin e
R o a d s u r f a c e c e n tr e lin e
R
RL
Figure 5.71 Superelevation run-off on tangent - circle.
Superelevation transition between two circles should have maximum 1/3 of the length on the
minor circle.
U /3
R1
R2
U R 1< R 2
D e s ig n r o a d c e n t r e lin e
R o a d s u r f a c e c e n tr e lin e
Figure 5.72 Superelevation run-off between two circles
Page 82
Revision)
5.4.6 Curve widening

The track of a vehicle is wider when it is turning than when it is travelling straight. Sometimes
curve widening is needed. Track width depends on a number of factors:
curve radius (R) and angle of change of direction (), see Figure 5.42 below
vehicle type, length and axle configuration
what traffic situation to design for
speed.
C u rv e w id e n in g
C e n tre p o in t
ra d iu s
A n g le o f d ir e c tio n c h a n g e
Figure 5.73 Widening
The extra track width compared with a tangent needed for two meeting semi-trailers and two
meeting 12 m buses are shown in figures below. Category A and B roads should be designed to
permit two semi-trailers to pass each other on curves.
W id e n in g (m )
4 .0
a n g le o f d ir e c tio n c h a n g e
>=50
3 .0 =20
=15
=10
=5
2 .0
1 .0
100 200 300 400 500

R a d iu s (m )
(given by the design vehicles, see section 3.3.4)
Figure 5.74 Extra width needed to allow two semi-trailers to pass on a curve
Page 83
Revision)
(given by the design vehicles, see section 3.3.4)
Figure 5.75 Extra width needed to allow two 12 m buses to pass on a curve
Where widening is applied the road marking centre line should be centred between the edge
lines. The extra width should be added to the inner edge of the carriageway and should be
adapted smoothly as illustrated below for the case tangent circle.
c o n s tr u c t io n lin e
TP
w id e n in g
TP
Tangent w id e n in g t r a n s itio n
Ri
R
Figure 5.76 Road widening design for tangent - circle
5.5 Vertical alignment

5.5.1 General
The longitudinal profile consists of a series of vertical curves and straight grades. Convex
vertical curves are known as summits or crests, concave vertical curves as sags. These elements
are determined taking into account:
Traffic safety
Driving dynamics
Visual guidance
Overtaking opportunities
Terrain adaptation and aesthetics.
Various curve forms can be used as vertical curves. The parabola has traditionally been used
because of the ease of manual calculations and is used throughout this Manual. Other forms,
Page 84
Revision)
particularly those more suited to computer calculation are equally satisfactory. Vertical curves
are here specified with a radius parameter R defined in the parabola as Y = X 2/(2 * R) as
illustrated for R = 500 m in Figure 5 .77 below.
Figure 5.77 Vertical radius with R = 500 m

Parabolas could alternatively be specified with the parameter K = length required for 1 % change
of grade. The relationship R=100*K could be used to transfer between the alternatives. The
knowledge on safety effects of vertical profile claims that grades in the range 2-3 % increase
accident rates by some 20 % and 3-4 % by some 30 %. Profile adjustments to improve
overtaking sight distances are normally difficult and expensive to carry out and maybe
counterproductive as illustrated in the example below.
Vertical radii Rv 4500, Rv 7000 and Rv 14000 create sight minima around 150 m, 200 m and
300 m. Minimum sight distances increase from some 200 m at Rv 4500 to some 300 m at Rv
7000 and some 700 m at Rv14000. The adjacent sight distances are simultaneously decreased.
The result could easily be longer minimum sight distances but decreased lengths with overtaking
sight distances. Climbing lanes will usually be more cost-effective.
O v e r ta k in g s ig h t d is t a n c e (m ) S ig h t d is t a n c e p r o f ile
1600
1200
800
400
0
0 /0 0 0 0 /5 0 0 1 /0 0 0 1 /5 0 0 2 /0 0 0
L e n g th (m )
E le v a tio n (m ) P R O F IL E
40 R v=4500
30 R v=7000
R v=14000
20
10
0
(adapted to the 0speed
/ 0 0 0 limit and 0design
/ 5 0 0 concept1 from
/ 0 0 0 Kenyan, Swedish
1 / 5 0 0 and German
2 / 0 0 0 design guidelines)
L e n g th (m )
Figure 5.78 Example of effects of vertical alignment adjustment
Page 85
Revision)
Driver perception depends on the arc length. Good visual guidance and aesthetic alignment
require sufficiently long arcs, see section 5.2.
5.5.2 Convex vertical curves (crests)

Minimum radius for convex vertical curves due to stopping, overtaking and no overtaking line
sight distances are given in Table 5.9. The sight distance values given require the arc lengths to
be longer than these minimum values. Shorter arc lengths have to be checked.
Speed Radius (m)

Stopping Overtaking
(km/h)
desirable minimum desirable no overtaking
centreline
markings
50 1100 600 11000 5500
80 4500 3000 32000 15000
100 10000 7000 65000 24000
Table 5.32 Minimum radius for convex vertical curves (crests)
5.5.3 Concave vertical curves (sags)

Discomfort is felt by a person subjected to high vertical accelerations, with 0.5 m/sec 2 being the
standard critical value. Minimum radii for comfortable driving are given in Table 5.10 . Drainage
could be a problem in sags. Rainwater will pour down the adjacent grade into the sag especially
if in a superelevation changeover, see section 5.45 for advice on how to ease this problem.
Speed Radius (m)

(km/h) Desirable Minimum
50 600 400
80 1300 1000
100 2000 1500
Table 5.33 Minimum radii for concave vertical curves (sags)
5.5.4 Gradients
Passenger car speeds are relatively unaffected by gradient, and the horizontal alignment tends to
determine car speeds. However, truck and bus speeds are markedly affected by gradient. Grades
less than 3% give crawling speeds over 60 km/h for a design heavy truck with effective
power/mass ratio 5 W/kg - see Figure 5 .79. There are also indications that grades below 3 % are
reasonable due to traffic safety. Long, steep down hills are very dangerous and need very careful
Page 86
Revision)
design, preferably with arrester beds (= specially designed run-off roads for vehicles with brake
problems).
V (k m /h ) D o w n h ill( % ) U p h ill ( % )
-1 0 -8 -6 -4 -3 -2 -1
90
0
80 + 1
70
60 + 2
50
+ 3
40
+ 4
30
+ 6
20 + 8
+10
10 = U p h ill d e c e le r a t io n
= A c c e le r a tio n
0
0 500 1000 1500 2000
L (m )
Figure 5.79 Heavy truck speed behaviour in grades
Suggested maximum gradients for different speed limits and types of topography are given in
Table 5 .34. It should be stressed that these are guidelines only. The optimization of the design
of a specific road may suggest some other maximum gradient. Climbing lanes (section 5.6) are
an option.
Speed Max. grade (%)

km/h Flat Rolling Mountainous.
50 6-8 7-9 9-10
80 4-6 5-7 7-9
100 3-5 4-6 6-8
Table 5.34 Maximum grades

5.6 Climbing and overtaking lanes

A climbing lane is an additional outside lane added to a single or dual carriageway in order to
improve overtaking opportunities, capacity and safety because of the presence of a steep uphill
gradient. It is inserted into the carriageway by means of entry and exit tapers to the left of the
continuous lane so that slow moving vehicles have to merge into the faster traffic at the
termination point.
Page 87
Revision)
An overtaking lane serves the same objectives without a steep gradient. Climbing lanes are
obviously more cost-effective than overtaking lanes as the probability of an overtaking need
normally is higher and the overtaking length is shorter.
Figure 5.80 Climbing lane outside the ordinary lane

Climbing lanes should be considered if the design truck speed decreases more than 20 km/h
under the truck speed limit, normally 80 km/h in rural conditions. This gives the threshold
criteria shown in Table 5 .35 on combinations of grade and length which fulfill these criteria.
(example: An average grade of 2% requires a length of 1500 m before the design vehicle has
dropped its speed 20 km/h). The speed profile method below could be used to assess truck speed
profiles more thoroughly.
Average Minimum length

grade (%) (m)
2 1500
3 500
4 300
Table 5.35 Threshold criteria for climbing lanes

Long steep down grades are very dangerous due to risk of overheating and brake failures. It is
recommended that down grades longer than 1000 m with average grades over 5% be reviewed
for the need of descending lanes.
It is difficult to give specific traffic flow and alignment warrants when to justify climbing lanes.
Two positive effects are obvious, improved level-of-service and traffic safety, to be weighed
against costs and intrusion. The traffic safety effect is estimated to be some 20 30 % according
to a number of studies. Level-of-service effects could be estimated using the US Highway
Capacity Manual and depends to a large extent on traffic flow, ratio of heavy vehicles and over
all alignment. The traffic flows in Table 5 .36 are suggested based on manuals from SATCC,
Kenya, Sweden and UK.
Page 88
Revision)
Grade AADT
% design year
3-4 3-5000
5-6 2-4000
Table 5.36 AADT-volumes to justify crawling lanes

The climbing (and descending) lane is sometimes not effectively used, especially when traffic
flows are heavy, because the drivers of slower vehicle fear that they will not be allowed to merge
with the faster vehicles when the climbing lane ends. The position of the lane-drop must
therefore allow the slower vehicle to gain enough speed to merge. The climbing lane therefore
should have full width over the section with design truck speed below 60 km/h with entry and
exit tapers according to the table below. The design should be smooth. At lane-drops the sight
distance should exceed that required for no overtaking centreline markings, and should
preferably be much more than this.
Speed Entry taper Exit taper (m)
limit (m)
(km/h)
80 150 200
100 200 300
Table 5.37 Entry and exit taper widths

The climbing lane width should normally be 3.5 m. but this can be relaxed to 3.0 m where space
is limited. The paved shoulder width should be unchanged, as the shoulder will continue to be
used by pedestrians and cyclists.
Overtaking lanes can be used to improve overtaking sight distances on flatter sections. The
length of an overtaking lane should be in the range 1000 m to 2000 m.
Speed profile
The speed profile graph can be used to assess truck speed behaviour on combined vertical
alignments as shown in the following example.
Page 89
Revision)
H (m )
70
P3 R e a l p r o file
60 P4 A p p r o x im a t e p r o file
50
A p p r o x im a t iv e p r o file
40 E le m e n t L ( m ) L ( m ) i (% )
P2
30 P1 150 0
150
P2 1000 6
20 1150
P3 310 0
10 1460
P1 P4 Long -2
0
0 500 1000 1500 2 0 0 0 L (m )
0% R v 5000 6% R v 7000 -2 %
P r o f ile
Figure 5.81 Example profile
Calculate the speed profile for a heavy truck with power-mass ratio 5 and a start speed of 80
km/h on the following profile with a 6 % uphill preceded by a straight and a 5000 m sag and
proceeded by a 7000 m crest and 2 % downhill. Approximate the profile to straights and grades
by extending the vertical curve tangents as illustrated in the figure above.
The truck speed will be affected from the start point of section P2. This element, 1000 m with
grade 6 %, will give the speed performance according to the segment AB on the + 6 %-curve
starting at 80 km/h with a length of 1000 m giving an end speed of some 28 km/h (crawling
speed for 6 %).
The truck will start accelerating on the third element P3, 310 m and flat, with speed behaviour
given by the segment CD on the 0 %-curve starting from some 28 m/h reaching some 56 km/h.
The truck will obtain its start speed after some 400 m on the final downhill with speed behaviour
according to segment EF on the -2 % curve.
Page 90
Revision)
V (k m /h ) D o w n h ill ( % ) U p h ill ( % )
-1 0 -8 -6 -4 -3 -2 -1
90
0
A
80 F + 1
70
60 + 2
E
D
50
+ 3
40
+ 4
30 B
+ 6
C + 8
20
+10
10
0
0 500 1000 1500 2000
L (m )
Figure 5.82 Example to assess truck speed profile

The resulting speed profile is illustrated below with proposed location of the crawling lane, entry
and exit tapers.
V (k m /h )
100
90
80
A F
70
c lim b in g la n e p o s it io n e x it t a p e r
60
e n try ta p e r D,E
50
40
30
20 B ,C
10
0
0 500 1000 1500 2000 L (m )
0 % Figure
R v 5 0 0 05.83 Resulting
6% R v 7 0 truck
heavy 00 - 2 % profile
speed P r o file
Page 91
Revision)
6 INTERSECTIONS
6.1 Introduction
6.1.1 Requirements
Intersections are the most critical element in the road network. Because they involve users
sharing road space (unless they are grade-separated intersections) their capacity is limited and
there is an inherent risk of collisions. A large proportion of road crashes in Uganda occur at
intersections.
The basic principles of good intersection design are:
minimize the number of conflict points, and thus the risk of accidents
give priority to major traffic movements, through alignment, signing and traffic control
separate conflicts in space or time
control the angle of conflict; crossing streams of traffic should intersect at a right angle or
near right angle
define and minimise conflict areas
define vehicle paths
ensure adequate sight distances
control approach speeds using alignment, lane width, traffic control, or speed limits
provide clear indication of right-of-way requirements
minimise roadside hazards
provide for all vehicular and non-vehicular traffic likely to use the intersection, including
goods vehicles, public service vehicles, pedestrians and other vulnerable road users
simplify the driving task, so that road users have to make only one decision at a time
minimise road user delay.
6.1.2 Design Procedure

Designing an intersection is a highly complex task and it is advisable to follow a standard
procedure. The design procedure should be used for new intersections as well as for upgrading of
existing intersections. There should be one procedure for preliminary design and one for detailed
design.
The objective with the preliminary design is to select the intersection type and location and to
make a draft intersection drawing and traffic control plan. The objective with the detailed design
is to do the geometric design and to make a detailed intersection drawing and traffic control plan.
As with all road projects, it is recommended that a safety audit be done before the scheme is
finalised and built.
6.2 Location
Intersections should be located where they can be clearly seen and easily understood by drivers
on all the approach roads and where the provision of desirable, as opposed to minimum, safety
standards are possible and economical to obtain. Crests, gradients and curves should be avoided.
T-intersections on the outside of a curve will have much better visibility than those that are
located on the inside of a curve.
Page 92
Revision)
Intersections should not be located where it is difficult or expensive to provide adequate

visibility or driving comfort. Locations which should be avoided are for example where
earthworks are heavy, near bridges, on small radius curves, on the outside of superelevated
curves, on high embankments, steep grades (>3%) or on crests.
Careful location as well as landscaping of the surrounding terrain and planting can be used to
improve the visual guidance and visibility at intersections. This can be of great importance for
the perception and comprehension of the intersection.
Figure 6.84 Example of planting at an intersection
6.3 Distance between Intersections

6.3.1 General
The number of intersections should be as low as possible to decrease travel times and increase
safety. There are no strict guidelines for the minimum distances between intersections on
different road categories. However, a rule of thumb is that the minimum distance between
intersections on main roads should be ten times the design speed (km/h).
In areas with many minor side roads they should if possible be combined before being connected
to the main road. This can be achieved by building a service road parallel to the main road.
6.3.2 Staggered intersection

The minimum distance between minor roads in a right/left (first right then left turn between the
minor roads) staggered intersection should be 50 metres to allow long vehicles to complete the
right turn from the minor road.
In a left/right staggered intersection the minimum distance should be at least 100 metres and
preferably longer,- see section 6.6.4 - to make it possible to provide right turn lanes in the major
road.
Where there is a lot of cross traffic (from one minor road to the other) the right/left stagger is
preferable. This is because, once the driver has turned into the main road, he can proceed to the
exit without impeding other traffic. The left/right stagger involves vehicles turning right out of
Page 93
Revision)
the main road across the path of oncoming traffic, and this is a particularly hazardous
manoeuvre.
> 50 m
> 100 m
Figure 6.85 Minimum distances between minor roads in a staggered intersection
6.4 Intersection Types

6.4.1 Access roads
Access roads are used to connect properties etc to the road network. Accident risk increases with
the frequency of access roads, so they should, as far as possible, be discouraged on primary and
secondary roads (categories A and B).
Where the access road joins the main road the corner radii should be between 6 and 10 metres. If
the access is heavily-used, such as at the entrance to an industrial site, the intersection standards
set out in this section should be applied. The approach to the main road along the access road
should be level with the surface of the main road for the last 5-10 metres.
6.4.2 Proposed intersection types

This Manual covers only at-grade intersections. At-grade intersections can be classified into two
main intersection categories depending on the form of control used. For each category, there are
a number of intersection types. The classification of at-grade intersections is shown in Table 6 .
38.
Traffic control
Intersection
Intersection types
category
Major road Minor road
Priority intersection Priority Stop or give way A Unchannelised T-intersection

sign B Partly Channelised T-intersection
Page 94
Revision)
C Channelised T-intersection
D Roundabout
Control intersection Traffic signals or give way sign E Signalised intersection
Table 6.38 Classification of at-grade intersections

6.4.3 Priority intersections
Priority intersections will be adequate in most rural situations. This Manual gives advice on the
design of three types of T-intersection:
Unchannelised T-intersection (A)
The unchannelised design is suitable for intersections where there is a very small amount of
turning traffic. It is the simplest design and has no traffic islands.
Partly Channelised T-intersection (B)
The partly channelised design is for intersections with a moderate volume of turning traffic. It
has a traffic island in the minor road arm. In urban areas, the traffic island would normally be
kerbed in order to provide a refuge for pedestrians crossing the road.
Channelised T-intersection (C)
The fully channelised design is for intersections with a high volume of turning traffic or high
speeds. It has traffic islands in both the minor road and the main road.
Typical design of priority intersections in rural areas is shown in Figure 6 .86.
Unchannelised Partly channelised Channelised
Figure 6.86 Typical design of T-intersections
Page 95
Revision)
The crossroads form of priority intersection must not be used. It has a very high number of
conflict points, and has a much higher accident risk than any other kind of intersection. Existing
crossroads should, where possible, be converted to a staggered intersection, or roundabout, or be
controlled by traffic signals.
Control intersections
Control intersections are mostly used in towns and trading centres. However, roundabouts can be
used in rural areas in intersections between major roads or other intersections with high traffic
volumes. There are two types of control intersections:
Roundabout (D)
Roundabouts are controlled by the rule that all entry traffic must give way to circulating traffic.
The ratio of minor road incoming traffic to the total incoming traffic should preferably be at least
10 to 15% . Roundabouts can be of normal size, i.e. with central island radius 10 m or more, or
small size, i.e. with central island radius less than 10 m.
Signalised intersection (E)
Signalised intersections have conflicts separated by traffic signals. No conflicts are allowed
between straight through traffic movements.
Typical design of control intersections is shown in Figure 6 .87.
Roundabout Signalised intersection
Figure 6.87 Typical designs for control intersections
6.5 Selection of Intersection Type

6.5.1 General
These selection guidelines mainly deal with traffic safety. Other important impacts such as
capacity / road user costs, environmental issues, investment and maintenance costs should also
be taken into consideration. Capacity, delays, queue lengths, road user costs and also exhaust
emissions could be estimated using standard software such as Oscady, Picady and Arcady (UK)
SIDRA (Australia) or Capcal (Sweden) but they have not been calibrated for Ugandan
Page 96
Revision)
conditions, so they could give misleading results. Some traffic flow threshold values for capacity
are given in figure 6.6.
The safety requirement for intersections can be defined as an interval where the expected number
of accidents should not exceed a desired level and must not exceed a maximum level. If the
expected number of accidents does not exceed the desired level, a priority intersection should be
selected. If the number exceeds the maximum level, a control intersection should be selected.
Between the two defined levels, a control intersection should be considered. The traffic flow
threshold values presented in the following figures 6.5 and 6.7 are based on this concept using
general European traffic safety research results on the relationship between speed and incoming
traffic flows on the major and minor road.
The selection is divided into two steps; selection of intersection category (priority or control) and
selection of intersection type. It is based on the following assumptions:
Priority intersections can be safe and give sufficient capacity for certain traffic volumes and
speed limits
If a priority intersection is not sufficient for safety and capacity, the major road traffic must
also be controlled.
Depending on location, traffic conditions and speed limits, different types of priority or
control intersection should be selected.
6.5.2 Selection of intersection category

Safety
The selection of intersection category should mainly be based on safety. The selection can be
made by using diagrams with the relationships between the safety levels and the average annual
daily approaching traffic volumes (AADT in veh/day, see section 3.2.2) based on accident
statistics. The diagrams shown in Figure 6 .88 are for T-intersections on 2-lane roads with 50, 80
and 100 km/h speed limit. The diagrams are, as already stated, based on general European
experience on relationships between speed, safety and traffic flows. They are judged reasonable
to be used in Uganda until sufficient local research is available.
M in o r r o a d a p p ro a c h in g A A D T , Q 3 v e h /d a y
6000 Q 3 5 0 k m /h
S e le c t c o n tro l
Q 1 Q 2 in te r s e c t io n
4000
C o n s id e r c o n tr o l
in te r s e c t io n
2000
S e l e c t p r i o r it y
5000 10000
M a jo r ro a d a p p r o a c h in g A A D T , Q 1 + Q 2 v e h /d a y
Page 97
Revision)
M in o r r o a d a p p r o a c h in g A A D T , Q 3 v e h /d a y
3000 S e le c t c o n tr o l 8 0 k m /h
in te r s e c t io n Q 3
Q 1
2000 Q 2
C o n s id e r c o n tro l
1000
5000 10000
M in o r r o a d a p p r o a c h in g A A D T , Q 3 v e h /d a y
3000 1 0 0 k m /h
S e le c t c o n tr o l
2000 Q 3
C o n s id e r
c o n tro l Q 1
in te r s e c t io n Q 2
1000
5000 10000
Figure 6.88 Selection of intersection category as to safety
Capacity
The selection of intersection category based on safety should be checked for capacity. It can be
made by using diagrams with the relationships between the capacity and the approaching traffic
volumes during the design hour (DHV in pcu/design hour, see section 3.2.2 and 3.3.3). The
diagrams shown in Figure 6 .89 are for T-intersections on 2-lane roads with 50, 80 and 100 km/h
speed limit. The desired level refers to a degree of saturation (actual traffic flow/capacity) of 0.5.
The acceptable level refers to a degree of saturation of 0.7.
The diagrams are based on Swedish capacity studies with findings similar to other European
countries. It is judged reasonable to be used in Uganda until sufficient Ugandan research is
available. Capacity could be checked more in detail using standard capacity software as already
stated with the general drawback that Ugandan capacity studies are as yet not available.
Page 98
Revision)
M in o r r o a d a p p r o a c h in g D H V , Q 3 p c u /d e s ig n h o u r
5 0 k m /h
400 C o n tr o l o r g ra d e -s e p a ra te d
A c c e p ta b le
in te rs e c tio n n e e d e d
D e s ire d
200
Q 3
Q 1
Q 2
0
0 500 1000 1500
M a jo r r o a d a p p r o a c h in g D H V ,Q 1 + Q 2 p c u /d e s ig n h o u r
8 0 k m /h
400 C o n tro l o r g r a d e -s e p a ra te d
A c c e p ta b le
D e s ire d
200
Q 3
Q 1 Q 2
0
0 500 1000 1500
400
1 0 0 k m /h
A c c e p ta b le C o n tro l o r g r a d e -s e p a ra te d
200
D e s ire d
Q 3
Q 1
Q 2
0
0 500 1000 1500
Figure 6.89 Selection of intersection category as to capacity
Page 99
Revision)
6.5.3 Selection of intersection type

Priority intersections
The selection of priority intersection type should mainly be based on safety. The selection can be
made by using diagrams with the relationships between the safety levels and the average annual
daily approaching traffic volumes (AADT in veh/day, see section 3.2.2) based on accident
statistics. The diagrams shown in Figure 6 .90 are for T-intersections on 2-lane roads with 50, 80
and 100 km/h speed limit. Crossroads should be avoided. The number of right turners should
obviously also impact the decision.
The diagrams are based on general European findings on safety effects of right turn lanes. It is
judged reasonable to be used in Uganda until sufficient Ugandan statistics are available. Note
however they are only a starting point for determining the most appropriate form of intersection.
6000 Q 3 5 0 k m /h
Q 1 S e le c t c h a n n e lis e d
Q 2
T - in t e r s e c t i o n
4000
S e l e c t u n c h a n n e li s e d
2000 o r p a rtly c h a n n e lis e d
5000 10000
3000 Q 3 8 0 k m /h
Q 1 Q 2
2000
S e le c t c h a n n e lis e d
1000
o r p a rtly c h a n n e lis e d
5000 10000
M a jo r ro a d a p p r o a c h in g A A D T, Q 1 + Q 2 v e h /d a y
1500 1 0 0 k m /h
Q 3
1000 Q 1
Q 2
S e le c t c h a n n e lis e d
500
o r p a r tly c h a n n e lis e d T -in te r s e c tio n
5000 10000
M a jo r ro a d a p p r o a c h in g A A D T, Q 1 + Q 2 v e h /d a y
Figure 6.90 Selection of priority intersection type as to safety
Page 100
Revision)
Partly channelised should normally be used if needed to facilitate pedestrian crossings and also if
the minor road island is needed to improve the visibility of the intersection.
Control intersections
Roundabouts are suitable for almost all situations, provided there is enough space. Roundabouts
have been found to be safer than signalised intersections, and are suitable for both low and
medium traffic flows. At very high traffic volumes they tend to become blocked due to drivers
failing to obey the priority rules. Well-designed roundabouts slow traffic down, which can be
useful at the entry to a built-up area, or where there is a significant change in road standard, such
as the change from a dual carriageway to a single carriageway.
Traffic signals are the favoured option in the larger urban areas. Co-ordinated networks of
signals (Area Traffic Control) can bring major improvements in traffic flow and a significant
reduction in delays and stoppages. However, they must be demand-responsive, in order to get the
maximum capacity from each intersection. Observance of traffic signals by Ugandan drivers is
reasonably good, and could be improved through enforcement campaigns.
For some traffic distributions, for example high traffic volumes on the major road, the total delay
can be shorter in a signalised intersection than in a roundabout. The diagram in Figure 6 .91
shows the traffic conditions for which signalised intersections are most suited, based on Kenyan
and UK experience.
M in o r r o a d a p p r o a c h in g A A D T , v e h /d a y
15 000
10 000
I n te rc h a n g e
needed
Ro u nd a b o u t
5 000
C o n s id e r
S ig n a lis e d
In te rs e c tio n
0
0 10 000 20 000 30 000
M a jo r r o a d a p p r o a c h in g A A D T , v e h /d a y
Figure 6.91 Selection of control intersection type

If a signalised intersection is considered due to planning conditions or traffic volumes, a capacity
analysis and economic analysis should be made. This should include road construction and
maintenance costs, accident costs, travel time costs, vehicle operating costs and environmental
costs.
6.6 Design of Priority Intersections

The following design items are covered in this chapter:
1. Alignment, camber and superelevation
2. Visibility
3. Corner curves
Page 101
Revision)
4. Traffic islands in the minor road

5. Right turn lanes with traffic islands in the major road
6. Left turn lanes.
6.6.1 Alignment, camber and superelevation

The two roads should preferably intersect at, or nearly at, right angles. Angles between 75 0 and
1200, with 00 representing the direction of travel on the main road, are also acceptable. Where the
intersection angle is outside this range (called skew intersections) the driver on the minor road
cannot get a good view of the main road traffic, and the risk of collisions increases. In this case
the minor road should be re-aligned see example in Figure 6 .92.
More complex layouts with more than four roads should be simplified by combining and
realigning the approaches.
C o r n e r r a d iu s t o b e
t h a t w h ic h w il l r e d u c e
s p e e d s to 1 5 k m /h
b e lo w th a t o n
a p p ro a c h
S tr a ig h t w ith le n g th
e q u a l to S S D fo r
d e s ig n s p e e d o f
p r e v io u s c u r v e
Figure 6.92 Realignment of skew intersection

Where the minor road approaches the main road on a grade the last 15 metres of the approach
shall be level with the surface of the main road.
The camber or superelevation of the major road should normally be retained across the junction
and the side road graded into the channel of the major road. Superelevation greater than 2.5%
should be avoided, if possible.
6.6.2 Visibility
Visibility splays on the minor road
Drivers approaching the intersection along the minor road need to be able to see whether the
main road is sufficiently clear of traffic to enable them to proceed. The required visibility
standards are set out in Figure 6 .93. The normal eye and object heights apply - see section 5.3.
It is not advisable to provide more visibility than this, as it might encourage excessive approach
speeds.
Where it is not possible to achieve the visibility in the minor road to let the intersection operate
with a Give way sign, see figure 6.10 below, it will be necessary to use a Stop sign instead.
The "Stop" sign could also be used to improve traffic safety in difficult locations.
Page 102
Revision)
L L
M a jo r r o a d
M in o r r o a d
Type of Control A L (m)

(m) 50 km/h 70 km/h 80 km/h 100 km/h
STOP control 5 110 170 205 260
GIVE WAY control 20 120 195 235 315
Figure 6.93 Visibility splays for Stop and Give way regulation
(adapted mainly from Swedish design guidelines)
6.6.3 Corner curves and swept paths

Corner ccurves and channel widths are determined by the dynamics of the design vehicles. Rural
intersections should normally be designed for the design vehicles illustrated in section 3.3.4. It is
recommended that a 3-centred compound curve be used at corners. The ratio R 1:R2:R3 should be
2:1:3. R2 should normally be 12 metres, as shown in Table 6 .39, but can vary from 8 to 12
metres depending on the turning characteristics for the vehicles expected to use the intersection,
see illustration in figure 6.11 below.
Radius R1 R2 R3
Ratio 2 1 3
Normal value (m) 24 12 36
Table 6.39 Details of compound curve
6.6.4 Traffic islands in the minor road

General
A traffic island in the minor road helps to:
increase the visibility of the intersection for traffic approaching along the minor road (this
can improve safety considerably in some cases)
make it easier and safer for pedestrians to cross the road
guide drivers in the correct path to take (discourage short-cutting)
discourage high approach speeds
provide protection for vehicles waiting to enter the main road
Traffic islands shall normally be kerbed and raised. They should have an area of at least 4.5 m 2
so that they can be easily seen. The standard design of rural traffic islands in the minor road is
Page 103
Revision)
shown in Figure 6 .94 and Figure 6 .95. Nevertheless, even when these designs are used, the
intersection layout must be carefully checked to ensure that there is adequate clearance for all
types of vehicles that are likely to use the intersection.
At very lightly-trafficked intersections it may not be necessary to provide a traffic island in the
minor road.
Partly channelised intersection

The standard design for a partly channelised rural intersection is shown in Figure 6 .94. The
traffic island could be smaller in urban areas with measures to facilitate pedestrian crossings, see
section 8.1.2.
Note that there is no provision for acceleration and deceleration lanes. These are not normally
necessary, and in some circumstances can be unsafe. Refer to section 6.5.5 for more advice.
RC Control radius normally 15 metres

Page 104
Revision)
W1 Minor road lane width, but not less than 3.0 metres
W2 Normally 5.5 metres + offset to kerbs if any
Figure 6.94 Traffic island in the minor road in partly channelised intersection
(adapted mainly from the Kenyan guidelines)
Channelised intersection
The standard design for a rural channelised intersection is shown in Figure 6 .95. There is an
extra lane in the main road for right turning traffic. This enables the right turning vehicles to
slow down and stop clear of the through traffic stream. The dividing island also provides some
protection for these vehicles while they wait for a gap in the opposing traffic stream.
The traffic islands in the main road may be formed using road markings or kerbed islands.
Kerbed islands are better respected than road markings but they are a hazard, especially at night
on high-speed (unlit) rural roads. If road markings are used consider laying rumble strips within
the islands, and installing reflective studs this can help improve driver compliance.
Note that there is no provision for acceleration and deceleration lanes. These are not normally
necessary, and in some circumstances can be unsafe. Refer to section 6.6.6 for more advice.
Page 105
Revision)
cL M I N O R R O A D
2 .5 2 .5
R C
R C 2 .0 - 4 .0
R 3 R 1
R = 1 .5 R
R 2
a p p r o x . 2 5 .0
2
R 1 R 3
16
0
W 1
S tr a ig h t
R = 0 .2 5
W 2
S tr a ig h t
0 R = 2 5 0
20
R = 2 5 0
RC Control radius normally 15 metres

W1 Minor road lane width, but not less than 3.0 metres
W2 Normally 5.5 metres + offset to kerbs if any
Figure 6.95 Traffic island in the minor road in a channelised intersection

(adapted mainly from Kenyan design guidelines)
6.6.5 Right turn lanes

Right turn lanes should be provided in channelised intersections. They should be designed
according to the following principles.
Main design features
A right turn lane consists of the main design features shown in Figure 6 .96.
intersection cross-section design
right turn lane design
taper design
Page 106
Revision)
cross-section widening.
LR
g h o s t is la n d le n g th
LC LS
C h a n n e l w id th
X
w id e n in g le n g th L w C h a n n e l w id th
Y
in te r s e c tio n
c r o s s - s e c tio n
ta p e r
d e s ig n
Figure 6.96 Main design features in right turn lane design
Major road cross section

Excessive intersection widths should be avoided in order to discourage high speeds and
overtaking. The major road cross section at the right turn as indicated as section X-Y in Figure 6
.96 is shown in Figure 6 .97.
X Y
W S1 W L1 W R W T W L2 W S2
S E C T IO N X -Y
Figure 6.97 Major road cross section
The maximum cross-section width is:

W = WS1 + WL1 + WR + WT + WL2 + WS2.
through lane widths, WL1 and WL2, should normally be unchanged through the intersection.
If they are 3.5 m on the approaches to the intersection, they could be slightly narrowed to
discourage high speeds and overtaking, otherwise the width should be kept.
the right turning lane width, WR ,should normally be 3.5 m. It could be decreased to 3.0 m
in urban areas
Page 107
Revision)
the traffic island width, WT, depends on island type:

- island created with road markings: normally 0.35 m for double centre line
- kerbed island: space needed for:
Pass left side only traffic sign (sign no. R75.1), 0.4 to 0.9 m
lateral clearances, minimum 0.3 m
an inner hard shoulder, if needed, in the opposite direction, 0.25 to 0.5 m for an edge
line
The total width will vary from minimum 1.0 m to 2.0 m.
paved shoulder widths, WS1 and WS2, normally 1.5 m wide, should be narrowed to 0.5 m in
order to discourage overtaking in the intersection. Separate footways should be provided
for pedestrians so that they do not have to walk on the shoulder.
Where there are many long vehicles turning right into the main road consider extending the
width of the island so that it provides them with some protection if the driver decides to make the
turn in two stages (i.e. crosses one major road traffic direction at a time).
Widening of the main road
The main road cross-section has to be widened over the length LW in order to create the traffic
island and the turning lane. The width of the widening depends on the maximum intersection
width and the adjacent cross-section width. The widening should be done over a taper of
preferably 1:30 (speed >80 km/h) or 1:20 (speed 80 km/h). It is advisable to lengthen the
island, if the intersection is located on a crest or in a horizontal curve, as this will make the
intersection more visible to approaching traffic.
Length of the right turn lane
The right turn length LR, see Figure 6 .95, consists of the following sections:
Deceleration section, LC
Storage section, LS
The normal length of these sections is shown in Table 6 .40.
The total length can sometimes be extended in rural areas to allow higher approach speeds to the
right turn lane. Shorter sections could be necessary in urban areas due to limited space available.
At busy intersections it may be necessary to refer to the intersection capacity calculations in
order to estimate queue lengths.
Speed (km/h) 50 >50
Length of deceleration section, LC (m) 30 50
Length of storage section, LS (m) 30 30

Total length of right turn lane, LR (m) 60 80
Table 6.40 Normal length of right turn lanes
Page 108
Revision)
6.6.6 Left turn lanes

Left turn deceleration lanes
Left turn deceleration lanes allow left turning major road traffic to slow down and leave the
major road without impeding the following through traffic. However, they have less operational
and safety benefits than right turn lanes. They may be appropriate at intersections on high-speed
roads, especially if they are on gradients or there is a large volume of left-turning traffic.
A left turn lane consists of:
diverging section
deceleration section.
Left turn deceleration lanes should normally be designed according to Figure 6 .98. They should
be 3.5 m wide excluding shoulder. The minimum length of the diverging section is dependent
on the speed. The minimum length of the deceleration section is dependent on the speed and the
exit radius. The exit radius should normally be 25 m, and there should be a triangular island in
the mouth of the minor road to channel the turning vehicles. A Give Way sign (sign no.R62)
and marking (sign no. M32) must be provided at the entrance to the minor road to indicate that
the left turning traffic does not have priority over the traffic turning right from the main road.
R 6 2 S ig n s m a n u a l
Speed (km/h) 50 >50

LC Length of diverging section 30 m >60 m
LD Length of deceleration section 30 m 50 m
Figure 6.98 Design of left turn deceleration lanes

(adapted mainly from Kenyan design manual)
Left turn acceleration lanes

Left turn acceleration lanes allow minor road traffic to accelerate fully before entering the major
road. However, they are generally not safe to use on single carriageway roads. They may be
appropriate at intersections on high-speed, multi-lane dual carriageway roads, where the
consequences of dangerous merging are not likely to be as severe as on single carriageway roads.
The volume of left-turning traffic and the gradient, if any, should also be taken into account.
Where they are used the corner radius should be 25 m and the lane should be 3.5 m wide. A
typical design for major road speeds of 70 to 80 km/h is shown in Figure 6.16.
Page 109
Revision)
m e d ian
3 .5 m
L m e rg e
if 8 0 k m /h
1 :1 0 L a c c ( if 8 0 k m /h 1 0 0 m ) 6 m
R 25 m
Figure 6.99 Design of left turn acceleration lanes
6.7 Design of Roundabouts

The following design items concerning roundabouts are covered in this chapter:
1. General requirements
2. Design principles
3. Sight distances
4. Central island and circulating carriageway
5. Entries
6. Exits
7. Combination of entries and exits
8. Pedestrian and cycle crossings.
6.7.1 General requirements

The following features are generally considered necessary for a roundabout to perform safely
and efficiently:
it must be easily seen and identified when drivers are approaching it
the design must encourage drivers to enter the intersection slowly (<50km/h) and keep a low
speed throughout this is crucial to safety
the layout must be simple and easy to understand
it must be clearly signed and marked
adequate, but not excessive sight distance, must be provided at all entry points to enable the
driver to observe the movements of conflicting vehicles, pedestrians and cyclists.
lighting is very beneficial for safety.
6.7.2 Design principle

Speed control
Approaching vehicles must be slowed down to 50km/h or less and this can be achieved in
various ways including the use of a large centre island, offsetting the entry roads, and deflecting
the entry roads sharply to the left as they join the circulating carriageway. In practice all
roundabouts must be designed with some entry deflection.
To keep the speed down to 50 km/h or less, the roundabout should be designed so that it is not
possible to drive through it on a path (see Figure 6 .109) with a radius of more than 100m.
Page 110
Revision)
Number and alignment of entry roads

Roundabouts work best with four arms or entries, but they can also be used where there are three
or five entries. More than five legs should not be considered.
Ideally, the entry roads should be equally-spaced around the perimeter with a minimum angle of
600 between them.
In three arm intersections, the angles between the entry roads can be adjusted by displacement of
the central island from the intersection point of the centre lines of the connecting roads or by
deflection of the road alignments.
In five arm intersections, the space for the extra connection can be created by making the central
island elliptical or by increasing the radius of the central island to at least 20 m. However,
elliptical central islands can be confusing.
Figure 6.100Three and five arm roundabout
6.7.3 Visibility and sight distances

Roundabouts should preferably be located where approaching drivers can have a good overview
of the roundabout with its entries, exits and circulating carriageway. Stopping sight distance
must be available on all approaches.
Roundabouts can be made more visible from a distance by moulding the centre island into a
dome and by planting the centre with shrubs not thicker than 100 mm in diameter. Advance
direction signs (map-type with an accurate diagram of the roundabout approach and layout) are
essential. Large chevron signs must be installed on the centre island opposite each approach.
The visibility splays shown in Figure 6 .101 must be provided to allow drivers to judge whether
it is safe to enter the roundabout. It must be possible to see vehicles at the preceding entry and
the following exit as well as the nearest parts of the circulating carriageway. However, drivers
should not be able to see the preceding entry from more than 15m before the give way line, as
this might encourage excessive approach speeds.
Page 111
Revision)
s h r u b s p o s s ib le
60m
>=50m
5m
Figure 6.101Required visibility for entering a roundabout

(mainly adapted from UK guidelines)
Once in the roundabout drivers must be able to see the area shown in Figure 6.19. Signs and
landscaping on the centre island should be designed and located so that they do not obstruct the
view more than absolutely necessary as illustrated to the right above
S t o p p in g s ig h t
Figure 6.102Required visibility for drivers within a roundabout

(mainly adapted from UK guidelines)
Centre island and circulating carriageway

Definitions
The dimensions of roundabouts are defined by the following radii and widths shown in Figure
6 .103:
Edge of carriageway radius, Re
Central island radius, Rc
Inner central island radius, Ri
Circulating carriageway width, B
Traversable area (small roundabouts only).
Page 112
Revision)
Width of circulating
carriageway B
Edge of carriageway
radius Re
traversable area if small roundabout
Inner central island radius

Central island
radius Rc Ri if small roundabout
circulating central island circulating

carriageway carriageway
inner central island
traversable
area
Figure 6.103Roundabout radii and widths
Normal roundabouts (centre island radius 10 m)

Central island radius
The central island radius should normally be between 10 metres and 25 metres. It is difficult to
control speeds if the roundabout is larger than this, and this would mean that cyclists and other
vulnerable road users would be at risk. In most cases the size of the site will determine the size
of the roundabout.
There is no simple relation between the central island radius and safety and the possible capacity.
The capacity is also depending on other design parameters, such as the width of the circulating
roadway and the entry and exit angles and widths.
Width of the circulating carriageway
The width of the circulating carriageway depends on whether it is to be one lane or two lane.
Normally, one lane roundabouts are designed for a semi-trailer, and two lane roundabouts are
designed for a semi-trailer and a passenger car.
For normal (central island radius 10 metres or greater) one lane roundabouts and two lanes
roundabouts the central island radius, the edge of carriageway radius and the width of the
circulating carriageway are determined by the diagram in Figure 6 .104.
Check that the circulating carriageway is no more than about 1.2 x the maximum entry width.
Very wide carriageways encourage unsafe speeds.
Page 113
Revision)
Figure 6.104Radius of central island and circulating carriageway radius in normal

roundabouts
(consequence of the design vehicles used based on International agreements on vehicle manoeuvrability)
Small roundabouts (central island radius >10 m)

Where space is limited, such as in built-up areas, a slightly different design of roundabout is
needed.
Island radius
Small roundabouts shall have an inner central island radius of at least 2 metres.
Width of the circulating carriageway
The problem with small roundabouts is that it is difficult to control car speeds because the
circulating carriageway has to be very wide in order to accommodate semi-trailers and long
vehicles. The solution is to build a centre island with an outer fringe which is traversable by long
vehicles see Figure 6.20. The traversable area should be a maximum of 40mm high, have a
rough surface (to discourage light vehicles), and be edged with a mountable kerb. The intention
is that light vehicles will go around the outside of the traversable area, thus forcing the drivers to
travel slowly. Drivers of long vehicles will be able to negotiate the roundabout by letting the rear
wheels cross the traversable area.
Advice on central island radii and traversable area are given in Figure 6 .105.
Page 114
Revision)
Figure 6.105Roundabout radii in small roundabouts
6.7.4 Entries
Number of entry lanes
One lane roundabouts are preferred from a safety viewpoint. For higher traffic volumes, a 2-lane
circulating roadway may be necessary. The diagram Figure 6 .106 shows the need for two lanes.
E n t e r in g t r a f f ic , v e h /D h
c ir c u la tin g
1000
e n t e r in g
a p p ro ach A
T w o la n e s
500
C o n s id e r t w o la n e s
O n e la n e
0
0 500 1000 1500
C ir c u la t in g t r a ff ic , v e h /D h
Figure 6.106Number of entry lanes

The need for two lanes must be checked for each entrys entering and circulating flows during
the design hour. If two lanes are necessary for one entry, the whole roundabout should be
designed with two lanes. An alternative design to increase the capacity for one entry can be to
Page 115
Revision)
use a separate left turn lane as shown in Figure 6 .107. With this arrangement care must be taken
to ensure good visibility and signing at the merge otherwise cyclists and other vulnerable road
users could be put at risk.
Figure 6.107Alternative design to increase the capacity for an entry
Approach alignment
As previously stated, entry deflection is essential in order to reduce the speed of approaching
vehicles to 50km/h or less. The size of the deflection is dependent on the alignment of the entry
and should normally be at least one lane wide (3.5 m). Figure 6 .108 shows one way of
achieving entry deflection. Avoid making the deflection too sharp as this could cause vehicles to
overturn or overshoot (i.e. driver unable to stop at the give way line).
The entry road must be level with the circulating carriageway for a distance of at least 15 m
before the give way line.
Figure 6.108Design of approach deflection
Entry radius
The entry radius should normally be in the range 15 20 m. It should never be less than 10m.
Large entry radii will result in inadequate entry deflection and must not be used.
Entry width
The entry width is depending on the main entry radius. The entry widths in Table 6.4 should
normally be used for one and two lanes roundabouts respectively. The transition to normal lane
width should be at least 30 metres long.
Page 116
Revision)
Entry width
Number
Design vehicle(s) Entry radius >15
of lanes Entry radius
15 m m
1 Semi-trailer 6.5 m 6.0 m
2 Semi-trailer + passenger car 10.0 m 9.5 m
Table 6.4 Entry widths
6.7.5 Exits
Number of exit lanes
The number of exit lanes can be decided according to Table 6 .5.
Exiting traffic Number of

veh / Dh lanes
< 750 One
750 1 050 Consider two
1 050 1 500 Two
Table 6.5 Number of exit lanes

Exit curve
The exit should be designed to give smooth traffic flow. The main radius in the exit curve should
be between 50 and 100 metres for a normal roundabout. If there is a pedestrian crossing on the
exit the radius should be 50m or smaller, in order to control speeds.
Exit width
The exit widths in Table 6 .6 should normally be used for one and two lanes roundabouts
respectively. The transition to normal lane width should be 75 - 100 metres long.
Number
Design vehicle(s) Exit width
of lanes
1 Semi-trailer 5.5 m
2 Semi-trailer + passenger car 7.0 m
Table 6.6 Exit widths
Combination of exit and entry curves

Driving paths
The alignment of the connecting roads can make it necessary to adjust the exit and entry curve
radii. If larger radii than normal are used, it must be checked that all possible 2 metres wide
Page 117
Revision)
driving paths for passenger cars fulfill the requirement R1<=R2<=R3<=100 metres to achieve
speed control, see Figure 6 .109.
R3
R2
R2
R3
R1 R1
Figure 6.109Driving paths for passenger cars

Alignment between entry and exit
It is preferable to avoid reverse curvature between the entry and the following exit - see Figure 6
.110. For roundabouts with big central islands or long distances between entry and exit, this can
be difficult to avoid. If possible, the alignment of the connecting roads should be adjusted.
Figure 6.110 Alignment between entry and exit
Pedestrian and cycle crossings

Pedestrian/cycle crossings are normally placed according to one of the two alternatives shown in
Figure 6 .111.
G iv e W a y L in e
> 6 m
G iv e W a y L in e
A lt . 1 A lt . 2
Figure 6.111Location of pedestrian crossings
Page 118
Revision)
In Alt 1 the give way line is placed after and in Alt 2 before the pedestrian crossing. The
advantages and disadvantages with the Alt 1 compared to Alt 2 are mainly the following:
With a distance between the crossing and the give way line, vehicles can yield for the pedestrian
crossing and the roundabout separately. This improves capacity. The traffic safety effects are
questioned. Some traffic safety researchers claim Alt 2 to be superior.
With the pedestrian crossing at a distance from the roundabout, an exiting vehicle can give way
to a pedestrian without blocking the roundabout with obvious capacity advantages.
A disadvantage is that the traffic island may have to be extended and widened to accommodate
pedestrians and cyclists. Another disadvantage is that pedestrians have to make a bigger detour.
6.8 Design of Signalised Intersection

6.8.1 Introduction
This section deals with geometric layout of signalised intersections and connections to the signal
control strategy. Close co-operation is necessary with the signal control and lighting engineers
throughout the design process, especially in the early stages, to optimize and coordinate
geometric layout and signal control strategy.
Signal control at an intersection, properly designed, can enhance traffic safety and efficiency by
reducing congestion and conflicts between different vehicle movements. The major advantages
compared to priority-controlled intersections are:
the maximum waiting time is fixed and known (if capacity is not reached)
the available capacity is distributed fairly between approaches
the driver on the minor road does not have to make a judgement on when it is safe to
proceed.
Most of the safety problems that arise with signalised intersections are related to drivers passing
the signal at red, and rear-end collisions at signal changes from green to red. This has
implications for signal visibility and timings.
Primary conflicts between motor vehicles must be separated in a signalised intersection, see
Figure 6 .112 below. Motor vehicles passing a steady green signal or green arrow signal must
not encounter any primary conflicts, but lower order conflicts, i.e. with turning vehicles, may be
acceptable in some circumstances.
p r im a r y c o n flic t lo w e r o r d e r c o n f lic t
Page 119
Revision)
Figure 6.112 Primary motor vehicle conflicts

The control strategy of a signalised intersection is called the phases or the stage sequence. An
example of a stage sequence for a T-intersection with a protected right turn (controlled by a
green arrow) - is shown in Figure 6 .113.
S ta g e 1 2 3
Figure 6.113 Protected right turn stage sequence

The control strategy can work on fixed timings or be vehicle-actuated - adapting to traffic
conditions by detectors. Vehicle actuated (demand-responsive) signals are much more efficient,
and because of this drivers are more likely to comply with them. Each stage has a minimum and
maximum green time. There should always be an intergreen period between conflicting stages to
allow for safe stage changes. The length of the intergreen depends mainly on the size of the
intersection, speed limit and whether pedestrians and cyclists are involved. The time period
between two consecutive starts of the same stage is the cycle time.
6.8.2 Control strategy and layout

Signalised intersections should normally be restricted to roads with a speed limit 50 km/h. Where
signals are needed on roads with speeds higher than 50km/h additional equipment is needed to
ensure safety. Signals should never be installed on roads where the speed limit is higher than
70km/h.
Protected right turns are preferable from a safety viewpoint. They give positive control and are
easy to understand. The disadvantage is that they use up significant intersection capacity, so
waiting times are longer.
Pedestrian crossing signals may be provided at signalised intersections. They should have their
own stage, during which there should be no conflicts with vehicle movements.
The number of traffic lanes with permitted traffic directions and signal control type with stages
and location of signal heads should be decided due to capacity, traffic safety, road user costs,
environmental and other impacts and investment and maintenance costs. The capacity analysis
should be based on expected traffic volumes during the design hour, normally both morning and
evening peaks.
Page 120
Revision)
The following safety requirements must be coordinated with the geometric layout:
Protected right turns (i.e. without conflicts) must have right turn lanes
rig h t tu rn w ith o u t c o n flic t
Figure 6.114 Right turn lanes with protected right turn
Permissive right turns (i.e. with conflict with opposing straight forward traffic) can have
separate lanes.
rig h t tu rn w ith o p p o s in g
s tra ig h t fo rw a rd tra ffic
Figure 6.115 Right turn lanes with permissive right turn
6.8.3 Visibility
Each traffic lane shall have clear vision of at least one primary signal head associated with its
particular movement from the desirable stopping sight distance, 70 m at 50 km/h and 110 m at 70
km/h speed limit. It is also important that the desirable stopping sight distance is available to all
possible queue tails given by the capacity calculation. The warning sign for traffic signals must
be used where the visibility is marginal.
(adapted from UK guidelines)
Figure 6.116 Visibility requirements on intersection approach
Page 121
Revision)
The intersection intervisibility zone is defined as the area bounded by measurements from a
distance of 2.5 m behind the stop-line extending the full carriageway width for each arm, as
indicated in Figure 6 .117. Designers should aim to achieve the greatest level of intervisibility
within this zone to permit manoeuvres to be completed safely once drivers, cyclists and
pedestrians have entered the zone.
Visibility along the intersecting road must be at least equal to the standards for STOP signs, as
set out in Figure 6.10. This is to ensure a minimum level of safety when the signals are out of
order.
Minor obstructions to visibility caused by slender projections such as lighting columns, sign
supports, signal posts, controller cabinet and guardrails may be unavoidable. When placing signs,
street furniture and planting, consideration should be given to ensure that their obstructive effect
is minimised.
Figure 6.117 Intervisibility zone without pedestrian crossing

6.8.4 Lane design

Traffic lanes should normally be 3.0 to 3.5 m wide. Nearside (kerb) lanes that are well-used by
cyclists should be widened to 4 m if possible. The lane width can be narrowed to 2.75 m, if space
is very limited, but only if there are few trucks or buses.
The required lane lengths depend on estimated queue lengths to be decided based on the capacity
analysis.
Entry lanes for right turners are needed, as already stated, if protected right turns are to be used.
Additional entry lanes for through traffic will improve capacity and level-of-service, but the
larger intersection area can result in the need to set longer intergreen periods.
The entry taper Lin of a kerbed entry lane should be minimum 30 m (taper 1:10) to allow a design
semi-trailer to cope with it. Tapers can be narrowed to 1:5 to allow more queuing space within
the same total length.
Page 122
Revision)
L queue L in
Figure 6.118 Right turn lane design

Minimum design measurements for a right turn with a ghost island are shown in Figure 6 .119.
7 .5 m 1
d ir e c t 10
ta p e r g h o s t is la n d ta p e r
Figure 6.119 Ghost island layout

Opposing right turns, especially permissive right turns (i.e. with opposing traffic) on the main
road, should be aligned opposite each other to improve visibility to meeting vehicles, to avoid, if
possible, safety problems as shown in Figure 6 .120.
Figure 6.120Sight shadow design problem at permissive right turns

(adapted from Swedish guidelines)
The number of straight ahead entry and exit lanes should be balanced in order to reduce conflicts
caused by traffic merging or diverging within the intersection visibility zone. Lane drops should
take place beyond the visibility zone over a distance of at least 100 m for a single lane reduction.
The lane drop may be carried out on either the nearside or offside dependant on traffic
conditions.
Page 123
Revision)
>= 100 m
Intervisibility
zone
Figure 6.121Lane drop design principles

Slip lanes (for left turners) can be signalised or uncontrolled (give way signs and markings).
They can be used when left turn manoeuvres for large vehicles have to be facilitated, see
example below. Uncontrolled slip lanes improve the efficiency of the traffic signal control, as
intergreens can be decreased, especially at high left turn volumes. Uncontrolled traffic should be
separated with a triangular separation island.
ta p e r
set back
Figure 6.122Left turn slip lane with taper to facilitate large vehicles
If left turn slip lanes are used, a consistent design approach should be adopted for ease of
understanding. Uncontrolled slip lanes can be confusing for pedestrians. Uncontrolled and
controlled pedestrian crossings should not be mixed within the same intersection.
6.8.5 Swept paths and corner curves

Corner curves and channel width design depend on what design vehicle and design level-of-
service is chosen, see section 3.3.4.
Signalised intersections with very low volumes of large trucks and buses could have simple 6 m
corner radiuses to minimise the intersection area and optimise the signal control strategy. The
radius should be increased to 10 m if 12 m rigid trucks or buses are common. The following
Page 124
Revision)
combinations of tapers and corner radii can be used in urban areas to accommodate semi trailers,
see Figure 6 .123.
M in o r M in o r
ro a d ro a d
1
T
R
M a jo r A
T ro a d R
M a jo r
1 ro a d
B
R T A B
Urban 10 5 30 30
Rural* 15 10 25 25
* or triple radii curves, see section 6.6.2
Figure 6.123Combinations of tapers and corner radii (adapted from UK guidelines)
It is essential to ensure that adequate turning radii are provided for the swept paths of all types of
vehicles using the intersection. Swept paths must be checked for all permitted turning
movements to control locations of traffic islands, signals etc, see examples below. The example
on the left indicates that there is an unnecessary taper; the example on the right indicates that the
stop-line must be set back.
Simple swept path templates are not recommended for checking whether semi trailers can
negotiate intersections. The use of specialist computer software (such as AUTOTURN) gives a
much more accurate simulation.
S t o p - lin e s e t b a c k
U n e c e s s a ry t o a llo w s w e p t p a t h
c a r r ia g e w a y
Figure 6.124Examples of swept path checks

Nosings of central reserves and pedestrian refuges should be set back a minimum of 1.5 m,
measured from a line extended from the edge of the intersecting roads. Minimum clearances
should be provided, see section 4.2, and must be controlled if the superelevation is over 2.5 %.
Page 125
Revision)
6.8.6 Signals
There should be at least two signals visible from each approach, and stop-line usually comprising
a primary and a secondary signal (see also the Traffic Signs Manual, Volume 1). Where separate
signalling of turning movements is used this advice applies to the approach lane(s) associated
with each turning movement. One signal post can then display information for more than one
turning movement.
The primary signal should be located to the left of the approach a minimum of 1 m beyond the
stop line and in advance of crossing marks for pedestrians if any. The secondary signal should be
located within a 30 degree angle on a maximum distance of 50 m with priorities as shown in the
figure below.
( X ) p r io r ity
s e c o n d a ry
s ig n a l
(2 ) (3 )
(1 )
p rim a r y (4 )
s ig n a l
s tu d ie d a p p ro a c h
Figure 6.125 Signal location advice
The primary signal should preferably be located 0.8 to 1.0 metre from the edge of the
carriageway with 0.3 and 2.0 m as minimum and maximum. Recommended locations in relation
to the stop-line and a pedestrian crossing are shown below.
2 .5 m
1 .0 m
Figure 6.126 Primary signal location advice
The following alternative designs may be used where there are approaches with three or more
traffic lanes and protected right turns. The primary right turn arrow is mounted on the exit
separation island, Alt 1, or on an extra separation island in the approach, Alt 2, being more
expensive.
Page 126
Revision)
ALT 1 ALT 2
3 colour standard signal head

3 colour arrow
Figure 6.127Alternative signal signal head
locations for right turn lanes
The standard traffic signal head width is 300 mm (with 450 mm as oversize), which results in
island width requirements, including clearances, of 0.3 to 0.6 m from 0.9 m to 1.65 m. Wider
islands can be needed if they are also to serve as pedestrian refuges.
6.8.7 Pedestrian and cyclist facilities

Pedestrian crossings should be perpendicular to the edge of the carriageway to assist
intervisibility and to benefit visually impaired people. The footway should have a dropped kerb
see Figure 8.1.
Minimum measures for pedestrian refuges for pedestrian crossings timed to permit crossing in
one movement is shown below. The normal width should be 2.5 m, with 1.5 m as the absolute
minimum.
Page 127
Revision)
M in 1 . 5 m
C r o s s in g s r o a d
m a r k in g s
N o r m a lly
2 .5 to 5 m
Figure 6.128Traffic signal island and pedestrian refuge

Pedestrian phases should preferably not have conflicts with turning traffic. This could be
arranged with staggered pedestrian crossings as illustrated below.
Figure 6.129Example of a signal-controlled intersection with a staggered pedestrian

crossing
Page 128
Revision)
7 SPEED MANAGEMENT
7.1 Introduction
Major traffic safety problems arise when main roads pass through trading centres and towns.
This is because of the mix of long-distance high-speed motor vehicles with local access traffic,
parking and vulnerable road users. The safest solution and by far the most expensive is to build a
by-pass. If this is not possible, a number of traffic safety measures must be implemented, such
as:
Speed limit and speed control measures
Separate vulnerable road users from the motor traffic by providing footways and
cycleways
Separate the local traffic from the through traffic by providing service roads.
Through roads with heavy traffic can also be provided with a median to improve traffic safety.
U-turns should then preferably be achieved by use of roundabouts, which maybe false, i.e. no
connecting roads.
This section focuses on speed limits and speed control measures in built-up areas. However some
of the measures described may have uses in other situations, such as in advance of hazardous
bends or bridges. Advice on footways, cycleways and service roads is given in other sections.
7.2 Speed Management Principles

Through roads in trading centres and towns shall be speed limited to 50 km/h or less. On the
busier sections these speed limits must be reinforced by speed control measures. The speed
management principles for main roads through built-up areas are summarised in Figure 7 .130.
R u m b le s t r ip s S p e e d lim it 5 0 R u m b le s t r ip s
g a te C e n tra l s p e e d g a te
f o o t - a n d c y c le
c o n tr o lle d a r e a

w a y s e c t io n
Figure 7.130 Speed management principles on 50km/h through road
The standard sequence is:

rumble strips
gateway sign and gate
start of footways
speed controlled area normally speed humps.
The gateway sign is double-sided and combines the speed limit sign with a panel showing the
place name. The gateway sign should be preceded by rumble strips to alert drivers. Warning
Page 129
Revision)
signs are not normally necessary. The rural speed limit should be taken down to 50 km/h by an
intermediate speed limit.
The entrance to the built up area should be marked with a gate (see below). The cross-section
within part of and sometimes all the 50 km/h area should normally have separate footways.
Major trading centres and towns should also have service roads.
Details of the individual speed control measures are given below.
7.3 Speed Control Measures

7.3.1 Gates
It is necessary to emphasise the speed limit change at the entrance to the built-up area with a gate
to signal very clearly to motor vehicle drivers that driving conditions are to change. The gate
should preferably be designed according to speed control principles, i.e., the toughest vehicle
path for a passenger car, through the gate or portal should have an entry radius R1 below 100 m
for 50 km/h speed control and 50 m for 30 km/h speed control. Radiuses are not allowed to
decrease in the gate as the entry radius is to be the most speed controlling. The gate could be
one-sided with speed control only in the entry direction or two-sided with speed control also in
the exit direction. The design can be tapered or smoothed with circles.
entry speed controlled gate entry and exit speed controlled gate
smooth design taper design smooth design taper design
2 m passenger
c a r tr a c k 2 m passenger 2 m passenger
E x it r a d iu s 2 m passenger E x it r a d iu s c a r tra c k E x it r a d iu s c a r tra c k
E x it r a d iu s c a r tra c k
R 3 R3 R3 R3
S e p a ra te d fo o t S e p a ra te d fo o t S e p a r a te d fo o t S e p a ra te d fo o t
a n d c y c le w a y a n d c y c le w a y a n d c y c le w a y
a n d c y c le w a y
R2
R 2 R2
R2

E n tr y r a d iu s E n tr y r a d iu s E n t r y r a d iu s

R1
R 1 R 1

E n t r y r a d iu s

R 1
Figure 7.131The design of gates

The narrowing of the carriageway through the gate can put pedestrians and cyclists at risk of
being squeezed by motor vehicles. It is recommended that short footway / cycle by-passes be
built around the gates.
7.3.2 Speed control zone

Principles
The next step is to use speed controlling measures within the busier part of the 50 km/h speed
limit section.
Page 130
Revision)
The maximum intervals between speed control devices to achieve speeds in the range of:
30 km/h is preferably 50 m and not more than 125 m
50 km/h is preferably 125 m and not more than 175 m.
Speed control is most effectively achieved by humps. Speed control should preferably be located
where judged reasonable for drivers. Sometimes formal pedestrian crossings could be
implemented combined with humps, see section 8.1.2. Well-designed roundabouts are also very
effective speed control measures and are highly recommended. False roundabouts (i.e. where
there is no intersection) are worth considering.
Humps
The most effective device to use for speed control is the speed hump. Two alternative designs
have proved to be most effective. These are the circular hump and the plateau hump. The
geometric designs are shown Figure 7 .132 for length profile.
Length profile
circular
plateau
Figure 7.132Alternative design of humps
The circular hump is normally recommended. The plateau hump can be used in pedestrian and
cycle crossings flush with connecting foot and cycle ways.
The height of the circular hump should be 100 mm. Various hump radii and chord lengths are
given in Table 7.1. These are based on empirical studies into hump dimensions, speed, and driver
/ passenger discomfort. The 4 m long design giving car speeds in the range 30 km/h is
recommended in residential areas. Main through roads with large ratios of trucks and buses
should normally have the 6.0 m long design to ease discomfort for bus passengers the chord is
longer than the normal axle width.
Page 131
Revision)
R H e ig h t = 0 .1 0 m
le n g th
car (truck) Radius Length

speed level (m) (m)
20 (5) 11 3.0
25 (10) 15 3.5
30 (15) 20 4.0
35 (20) 31 5.0
40 (25) 53 6.5
45 (30) 80 8.0
50 (35) 113 9.5
(40) 180 12.0
(adapted from Danish guidelines)
Table 7.41 Detailed design of circular humps
The traffic level-of-service, especially for buses and trucks, can be improved if the hump entry
and exit is smoothed with a fillet as shown below.
H u m p s u rfa c e
S m o o th e d tra n s itio n
R o a d s u r fa c e
L e n g th (m ) -0 .7 5 -0 .5 0 -0 .2 5 0 0 .2 5 0 .5 0 0 .7 5
H e ig h t ( m m ) f r o m
0 2 5 10 6 2 0
h u m p a n d ro a d s u r fa c e
Figure 7.133Detailed design of hump transition

The height of the plateau hump should also be 0.10 m. Table 7.2 gives recommended ramp
grades and lengths based on empirical studies. The design with 1.0 m ramp length and 10%
grade giving car speed levels around 30 km/h is recommended for residential roads. Main
through roads with large ratios of trucks and buses should normally have the 6.0 m long design
to ease discomfort for bus passengers.
Page 132
Revision)
r 4 - 6 m
i 0 .1 0 m
car (truck) ramp length r

Grade i (%)
speed level (m)
0.7 14
25 (5) 0.8 12.5
30 (10) 1.0 10
35 (15) 1.3 7.5
40 (20) 1.7 6
45 (25) 2.0 5
50 (30) 2.5 4
(35) 3.3 3
(40) 4.0 2.5
Table 7.42 Detailed design of plateau humps
Note:
On a road with hard shoulders the hump must extend over the shoulder for a 1 m or so
Consider whether the hump will interfere with drainage. On roads with kerbed footways
you may have to stop the hump 100 150 mm before the kerb to create a drain. This
solution cannot be used at a flush pedestrian crossing.
100 - 150 m m
F o o tw a y H um p
Figure 7.134Example of hump drainage design at kerbed footway

Humps are only allowed on roads with speed limit 50 km/h or lower. They should always be
clearly marked, as illustrated in Figure 7.6, with chequerboard markers (sign no. M35) and hump
information signs (sign no. I02) in each direction of the road. Hump warning signs (sign no.
W24) might also be needed.
Page 133
Revision)
C h e q u e rb o a rd m a rk e rs
1 .5 m
H u m p le n g th
Figure 7.135 Markings and signing of humps

Rumble strips
Rumble strips are transverse strips across the road used to alert and warn drivers with a vibratory
and audible effect before a hazard such as a sharp bend, an intersection or a lower speed limit at
the entry to a trading centre. Warning signs are not normally needed when the strips are built to
the specifications given below.
Research in other countries indicates that speed reduction effects tend to be minor and also erode
over time. Reliance should therefore not be placed on using rumble strips alone to reduce speeds.
Rumble strips can be used for example in the following situations:
before a local speed limit
at an approach to a dangerous intersection
before a sharp bend
before a hump.
The following principles should be observed when using rumble strips:

rumble strips should normally be in groups of 4 strips
the height of the strips shall be no more than 10 15 mm
the strip width should be 0.5 m
one set of rumble strips is usually enough within 50km/h sections
the last or only strip should be located 30 to 50 m before the hazard
pre-warning sets can, if used, be located 20 to 80 m before the hazard depending on speeds
rumble strips should preferably have yellow thermoplastic lines across the top for better
visibility
strips should continue across the full width of the carriageway, including the shoulders but
be terminated so that they do not interfere with drainage.
Page 134
Revision)
Speed D
50 20
T o a llo w 70 80 In te rv a l 0 .5 - 1 m
d r a in a g e H a z a r d i. e .
and D 3 0 to 5 0
G a te
c y c le n a r ro w in g
passage s p e e d li m it
s h o u ld e r
0 .5 m
s t r ip d e ta il 0 .0 1 0 to 0 .0 1 5 m
Figure 7.136 Design of rumble strips
Rumble strips create disturbing noises and can cause vibration problems on soft ground. Avoid
installing them near houses, schools, hospitals, etc.
Narrowings and chicanes

Road narrowings and chicanes can help control speeds, but they tend to be less effective than
speed humps. The basic principles of one-sided and double-sided road narrowings are illustrated
in Figure 7 .137 below. Narrowings can put cyclists at risk of being squeezed by larger vehicles,
so it is best to provide a short by-pass for them.
One sided
S h o u ld e r o r fo o tw a y
narrowing
Double sided
S h o u ld e r o r fo o t w a y
narrowing
Figure 7.137Design of road narrowings
Table 7 .43 can be used to judge the relationship between speeds and meeting situations. The
conclusion is that a narrowing must be very tough to have some speed impact - 3.5 m is the
Page 135
Revision)
recommended width between kerbs for one-way traffic. The narrowing should be designed with
tapers at least 1:5.
Speed Meeting situation

km/h Truck and bicycle Two cars Truck and car Two trucks
30 4.0 m 4.15 m 4.95 m 5.9 m
50 4.5 m 4.5 m 5.5 m -
Table 7.43 Road width for different speeds and meeting situations
Another alternative is to build a kerbed island (min. width 1.2 m and length 5 m) in the centre of
the road, with 3.0 m 3.5 m wide traffic lanes either side. This could also function as a
pedestrian refuge, perhaps combined with a raised pedestrian crossing. The island must be well-
signed to avoid it becoming a hazard. As with all narrowings, consider whether cyclists may be
put at risk.
5 m
3 -3 .5
3 -3 .5
Figure 7.138Example of road narrowing with central island
Page 136
Revision)
8 OTHER ROAD FACILITIES

8.1 Pedestrian Facilities
In the past the needs of pedestrians were largely ignored, and this may be one reason why so
many pedestrians are killed and injured on our roads. Pedestrians have as much right to use the
road as motorists, and roads must be designed with their needs in mind. The first step is to
identify major pedestrian generators (markets, shops, schools, etc.) and determine which are the
most important pedestrian routes. The aim should be to develop a network of pedestrian routes
and crossing facilities that is convenient to use and avoids conflicts with vehicular traffic.
8.1.1 Shoulders and Footways

The conventional view is that pedestrians in rural areas can walk on the road shoulders. The
shoulder should be at least 1.5m wide, though 1m is just acceptable if there are constraints. The
surface must be well drained and be as smooth as the traffic lanes if not, pedestrians may prefer
to walk in the traffic lane. The implication of this is that low-cost chip seal shoulders may not be
a good investment. Letting pedestrians use the shoulders is not entirely satisfactory, as there is
nothing to protect the pedestrian from speeding traffic. This is of particular concern on high-
speed and / or high volume roads. In these situations it is preferable to provide a separate
footway several metres beyond the edge of the shoulder and separated from it by a grass strip
(see also Section 4 Cross-Section). Some criteria for the provision of footways are given in
Table 8.1 below but these should be used with caution in some circumstances footways can be
justified at lower pedestrian flows.
Pedestrian flow per day
Average daily
Location of footway Speed limit of 60 80 Speed limit of 80
vehicle traffic
km/h 100km/h
400 to 1,400 300 200
One side only
> 1,400 200 120
700 to 1,400 1,000 600
Both sides
> 1,400 600 400
Table 8.44 Criteria for provision of footways

Standard footway widths are:
Absolute minimum: 1m (two persons cannot pass each other)
Desirable minimum: 1.8m (two persons can pass each other closely)
Light volume: 2.25m (two persons can pass each other comfortably)
Heavy volume: 3.5m + (space for three persons)
In urban areas the footways are normally raised, and edged with barrier kerbs. Barrier kerbs
should normally be 100 150mm high. Higher kerbs (200mm) are sometimes used in order to
deter vehicles from parking on the footway, but these are not recommended for general use,
because they are too high for most pedestrians who will prefer to walk in the traffic lane. The
kerb should be lowered at all pedestrian crossings, and where private entrances, footpaths, and
Page 137
Revision)
cycle tracks enter the carriageway. The dropped kerb (Figure 8 .139) is particularly helpful to
disabled persons.
A lte rn a tiv e m a te ria l a t c ro s s in g
K erb fo r R a m p to e a se c ro ss in g
d ire c tio n a l a d v ic e fo r w h e e l c h a irs
Figure 8.139 Dropped kerb
8.1.2 Pedestrian crossing facilities

It is difficult to set down criteria for the provision of pedestrian crossing facilities. Factors to
take into account include:
the volume of pedestrians crossing the road
the speed of the traffic
the width of the road
whether there are a lot of children crossing
whether there are significant numbers of disabled pedestrians.
There are many things that can be done to help pedestrians cross the road, including:
Formal crossings
- uncontrolled (zebra) crossings
- controlled (signal-controlled) crossings
Humped pedestrian crossings
Build-outs
Refuge islands
Medians
Footbridges (see next sub-section)
Subways (see next sub-section).
The rules for the use of uncontrolled (zebra) and signal-controlled crossings are set out in the
law, and are fairly well understood though not always obeyed. These formal crossings should
be used where there are high volumes of pedestrians trying to cross wide and / or busy roads. If
formal crossings (zebra crossings especially) are used in places where there is no obvious need,
drivers will have even less respect for them than they have at present. The safety benefits of
zebra crossings are a little questionable given the lack of discipline amongst Ugandas drivers.
Yet, if discipline was better and all drivers stopped for pedestrians, it could lead to severe
congestion at the busier crossings. Signal-controlled crossings, though more expensive to install
and maintain, are likely to perform better, though there are not enough in Uganda to be sure of
this. Refer to the Traffic Signs Manual for details of the layout of these crossings.
Page 138
Revision)
Experience in other countries suggests that zebra crossings can be made to work better if the
crossing is marked on top of a plateau road hump (see Section 7, Speed Management). The hump
forces approaching vehicles to slow down and this gives the pedestrian a chance to step onto the
crossing and claim priority. Humped pedestrian crossings have not been tried in Uganda so it is
not clear how well they would work.
Crossing the road can be made a lot safer by means of simple, informal measures, such as build-
outs, refuge islands and medians. Figure 8 .140 shows the use of build-outs and refuge islands
with a zebra crossing, but they can also be used on their own.
Build-outs are useful on wide roads where there is roadside parking, because they extend the
footway further into the road thus improving intervisibility between pedestrians and drivers.
Refuge islands in the centre of the road enable pedestrians to cross the road in two stages, which
makes it much easier and safer. However, kerbed refuge islands are at risk of being hit by
speeding vehicles, so they must be well signed with R75, Pass left side signs and M53
traffic island markings. Alternatively, create refuges out of road markings (M53 traffic island
marking) and put rumble strips within them. Providing refuge islands along the whole length of
the road is the safest arrangement, and this will also help to discourage unsafe speeds and
overtaking. Refuge islands should be 1.5 2.0 m wide, or 1.2 m absolute minimum. Make sure
that there is sufficient width of carriageway left to enable traffic to flow freely. A width of 3.5 - 4
m is normally enough for one lane of traffic, but, if there are a lot of cyclists, this might need to
be increased to 4.5 m.
When considering the provision of crossing facilities at a site always check the intervisibility
between pedestrians and drivers. It should be at least equal to the stopping sight distance.
5m 3m
3 m -5 m
I0 1
P e d e s t r ia n
C r o s s in g
s ig n
B u ild - o u t
R 75 P ass
L e ft S id e M 5 3 T r a ff ic
S ig n R75 is la n d m a r k in g
M 53
1 .5 m 3 .5 m m in .
B u ild - o u t
I0 1
Figure 8.140Pedestrian crossing facilities
Refer to Section 4 Cross-section for advice on pedestrian facilities on bridges.
Page 139
Revision)
8.1.3 Footbridges and subways

Pedestrian footbridges and subways are not generally recommended, because they are
inconvenient to use, and have a number of other problems, including crime, vandalism, and
maintenance. However, they are very appropriate where the terrain is such that pedestrians can
use the footbridge or subway without having to climb or descend. Subways are preferable to
footbridges because the height difference is less, but they are much more costly, and the security
and maintenance problems can be worse. Whenever possible the footbridge or subway should be
in line with the normal path that pedestrians take when crossing the road. If pedestrians have to
diverge from their direct route they will be discouraged from using the facility. Pedestrian
barriers can be used to try and force pedestrians to use the facility, but local people will destroy
them if they feel that the detour is unreasonable. When designing footbridges and subways make
sure that they are easy to access and as pleasant and safe to use as possible. Subways should be
designed so that the pedestrian can see from one end to the other and can thus choose not to enter
if there is a potentially threatening situation. Recommended minimum dimensions for subways
are given in Table 8 .45.
Type of subway (m) Width (m) Height (m)
Narrow (short) 2.3 2.3
Standard 3.3 2.6
Wide (long) 5.0 2.6
Table 8.45 Minimum subway dimensions

Ideally, there should be a choice of stairs or ramps. Ramps should not normally be steeper than
5% (3% if used by cyclists) and should have a non-slip surface.
Recommended standards for stairs are:
Flights of stairs (between landings) should be limited to 20 steps (9 steps where there are
significant numbers of disabled persons)
Stair landing should be a minimum of 1.8m deep
Step should be of equal height throughout the subway or footbridge
Optimum dimensions for stairs are 300mm tread (horizontal) and 130mm rise (vertical)
Handrails (1m above the floor) should be provided on both sides of stairs and ramps central
handrails may be advisable where the width of the stairs or ramps exceeds 3m.
The minimum clearance required for footbridges above the carriageway surface is 5.0m. This
headroom must be maintained over the full width of the carriageway, including shoulders.
8.2 Bus Lay-bys

Bus stop lay-bys are recommended for new and upgraded roads, other than low-volume roads.
They enable buses to slow down and stop outside the traffic lane, and this greatly reduces the
risk of following traffic colliding with them or having to overtake in a panic. On busy roads bus
drivers may be reluctant to use them, because of the difficulty of getting back into the traffic
Page 140
Revision)
stream, but this problem can usually be overcome in liaison with the bus companies and the
traffic police. At busy stops the waiting passengers sometimes stand in the lay-by and so prevent
the buses from fully entering the lay-by. Pedestrian barrier can be installed to try and prevent
this, but local publicity campaigns can also be effective.
The standard design for a bus lay-bay is shown in Figure 8 .141, but the dimensions can be
adjusted to suit local traffic situations.
Bus lay-bys should preferably be located on straight, level sections of road with good visibility
(at least Stopping Sight Distance). They should be sited after intersections, to avoid stopped
vehicles from interfering too much with the view of drivers who want to enter the main road
from the minor road (see Figure 8.4). They should be sited after pedestrian crossings, for similar
reasons. Do not site bus lay-bys opposite each other, because this can cause safety problems due
to road blockage if both buses set off at exactly the same time. Stagger the lay-bys so that
departing buses move away from each other (Figure 8.4).
When siting bus lay-bys bear in mind that the existing bus stops will generally be located where
they are most convenient for the passengers, and it is usually very difficult to persuade
passengers and bus drivers to move to new stops, especially if they are more than 50m away. The
provision of bus shelters that protect passengers against the sun and rain can help make the new
bus stop more attractive. Ensure that pedestrian access is easy and convenient.
Figure 8.2 Design of bus lay-by
M 2 2 C o n tin u ity lin e M 21

It is of major safety importance to study how passengers can reach the lay-by in a safe way.
18m r 18m r 3 .2 5 m
12m r 12m r
15m v a r ia b le 20m
m in . 1 5 m
p la tf o r m fo r
p a s s e n g e rs
Figure 8.141Standard bus lay-by
Page 141
Revision)
Figure 8.142 Recommended location for bus lay-bys
At heavily used bus lay-bys the dripping of oil and diesel onto the road surface may result in
early failure of the pavement. In these situations it is preferable to use a concrete pavement.
8.3 Cycle Facilities

Cycling has health, environment, economic and other benefits, and in rural areas bicycles may be
the most common means of transport. So, all highway, traffic management, safety and
maintenance programmes should give consideration to the needs of cyclists. Although no
research has been done, it is likely that, as in other countries, they have an above average risk of
being involved in a road accident. It is important that more be done to reduce the danger of
cycling. The first step when planning highway programmes or projects is to assess the demand
and get some idea of journey patterns and volumes.
The conventional view is that cyclists in rural areas can use the shoulders, and this is acceptable
provided that the combined volume of pedestrians and cyclists is low (<400 per day) and the
shoulder is at least 1.5m wide. But, with heavier flows, and especially if there is high-speed
traffic and / or a high proportion of heavy goods vehicles, it will be better to provide a separate
cycleway or a combined cycleway and footway. Figure 8.5 shows the basic dimensions, and
Table 8 .46 gives recommended cycleway widths. Cycleways need to have a smooth surface
with good skid resistance.
.
1 .9 0
0 .7 5
L e n g th = 2 .0 0
Figure 8.143 Cyclist dimensions
Minimum Standard width Width for heavy

Type
width (m) (m) usage (m)
Cycleway
(separate from 2.0 2.5 3.5
carriageway)
Combined cycleway
2.0 3.0 4.5
and footway
Cycle lane
1.5 2.0 2.5
(one way)
Table 8.46 Recommended widths for cycle facilities

Page 142
Revision)
Speed management measures (see Section 7) in trading centres and other built-up areas will help
cyclists by reducing motor vehicle speeds, but the needs of cyclists must be considered from the
beginning. Otherwise features such as gates and narrowings could increase the dangers to
cyclists. If there are a lot of cyclists it may be worth providing a short by-pass (1m wide) which
will enable cyclists to avoid the speed reduction measure.
In the larger urban areas there may be a case for developing a network of cycle facilities linking
all the main travel generators. Complete segregation of cyclists and motorists is not necessary
they can mix safely at speeds up to 50 km/h providing there are not too many heavy goods
vehicles. The network can be based on the existing road network, but with special facilities, such
as cycleways and cycle lanes on difficult sections and at major intersections. The simplest way of
assisting cyclists is to provide a wide (4.5m) nearside lane. Cycle lanes (created with road
markings) should only be necessary on the busier roads. Providing for cyclists at major
intersections can be difficult, but measures which control speeds, reduce conflicts, and improve
visibility will be helpful for cyclists. All cycle facilities should be well signed for the benefit of
both cyclists and other road users.
Page 143
Revision)
9 ROAD FURNITURE AND OTHER COMPONENTS

9.1 Traffic signs and road markings
The use of traffic signs and road markings, including road studs and delineators, is described in
the Ministrys Traffic Signs Manual.
9.2 Safety Barriers

9.2.1 Principles
There are three main types of safety barrier:
wire rope
guardrail, with beams and posts made of steel
concrete barrier, made of reinforced concrete.
They are used to prevent vehicles from hitting or falling into a hazard - such as falling down a
steep slope, or falling into a river, hitting an obstruction near the edge of the road, or crossing a
median into the path of traffic on the other carriageway. These events happen when a driver has
lost control of the vehicle due to excessive speed, lack of concentration, tyre failure, collision,
etc.
Ideally, the safety barrier will;
prevent the vehicle from passing through the barrier (the vehicle will be contained)
absorb (cushion) the impact of the vehicle without injuring the occupants (no severe
deceleration)
re-direct the vehicle along the road parallel to the other traffic
enable the driver to retain control of the vehicle (no spinning or overturning of the vehicle).
No safety barrier yet invented will perform to this standard in every impact. There is great
variation in the impact circumstances (vehicle type, impact speed, impact angle, etc.,) so the
design of a barrier is inevitably a compromise.
The specification, installation and maintenance of safety barrier is a highly technical subject, and
this Manual can only give a brief introduction to the subject. Always seek advice from experts,
as safety barrier can be useless and even dangerous if not properly designed and installed.
Always purchase the components of safety barrier from a specialist manufacturer and obtain
their advice. If possible, arrange for them to install it, or supervise the installation.
9.2.2 Performance
Conventional safety barrier installations are designed for impacts by cars travelling at 65 km/h
hitting the barrier at a 25 degree angle. Barriers can be made that will cope with trucks and
buses, but the high cost means that they can only be justified in exceptionally risky situations.
Most barriers will not perform well when hit at a large angle - such as can happen when barrier is
installed on the outside of a sharp bend.
Maximum permissible deflection is an important consideration. Concrete barrier does not deflect
at all, so it is the best solution for shielding hazardous objects which are very close to the
Page 144
Revision)
carriageway, or for stopping out-of-control vehicles from crossing a narrow median. With steel
beam barriers the amount of deflection can be reduced to some extent by reducing the post
spacing or using two beams nested one inside the other.
Deflection
Category Type Comments
when hit
Expensive; technically complicated; quick
Flexible Wire rope 2m +
to repair
Steel beam Performs well on high-speed roads when
1 - 2m
weak post hit by cars; technically complicated
Semi-rigid Performs well in moderate-speed
Steel beam
0.9m situations; most common type in
strong post
developing countries
Expensive; high level of containment, but
No
Rigid Concrete barrier can cause severe deceleration when hit at
deflection
a large angle; low maintenance costs
Table 9.47 Safety barrier characteristics

9.2.3 When to use
When a roadside hazard is identified, the best solution is to remove the hazard. Where the hazard
is a drop, it will be worth considering whether the slope can be flattened to make it less of
hazard. If this cannot be done there may be a case for shielding the hazard with a barrier, but it is
important to remember that installing a barrier is a second-best solution, as there is no certainty
that it will perform successfully. Moreover, safety barrier is a hazard in itself. Collision with a
barrier can cause death and serious injury - particularly to riders of two-wheelers. This means
that safety barrier should only be installed when the consequences of an out-of-control vehicle
hitting the unprotected hazard are likely to be more severe than those of impact with the safety
barrier. However, this is very hard to judge. Vehicle occupants can sometimes die when the
vehicle they are in collides with an insubstantial object, or drops a very short distance, yet clearly
it is not economic to try and shield every hazard. The risk of a loss-of-control incident at a
particular spot increases with traffic volume, traffic speed, and road curvature, but not in any
predictable way.
On existing roads the main consideration will be the accident history. If collisions with the
hazard are occurring often there may be justification for safety barrier assuming the hazard
cannot be removed. In most situations steel beam strong post guardrail will be the best option.
Cost-benefit analysis can help determine whether it is worthwhile installing barrier.
Steel beam strong post guardrail is also the preferred type for use on new roads. It is
recommended that the following criteria be used:
For roads where cars travel in excess of 50 km/h:
At >2 metre drops at sharp bends (defined as those where the safe speed to negotiate the bend
is more than 15 km/h lower than the speed on the approach) where the side slope is steeper
than 1 in 4
Where there is a risk that vehicles could fall into a body of water deeper than 1 m or onto a
rail track
To shield any solid, substantial object within the clear zone
On medians less than 9 metres wide where the road has an ADT of >10,000
Page 145
Revision)
On straight sections or gentle bends on embankments as indicated in Figure 9.1- this level of
provision may be unaffordable, in which case give priority to bends with a radius of less than
450 m and high-speed, high-volume roads.
These are crude guidelines and there is scope to deviate from them if a case can be made. The
criteria should be re-assessed when more experience is gained.
S lo p e
( V e r t ic a l : H o r i z o n t a l )
1 :1 .5
1 :2
1 :3
1 :4
0 2 4 6 8 10 12 14 16 18
H e ig h t o f fill ( m )
(adapted from SATCC)
Figure 9.144Guardrail determination of need
9.2.4 Length of need

Steel beam safety barrier is often installed in lengths that are too short to be effective. This is
done to keep costs down, but the resulting installation may be completely useless. Generally you
need at least 30 m of steel beam strong post guardrail for it to perform satisfactorily. Figure 9.2
gives guidance on determining the length of need. Note that on a two-way single carriageway
road you have to consider both directions of travel - you cannot assume that vehicles will not hit
the downstream end of a barrier. One of the common faults on hazardous bends is to stop the
barrier at the point where the bend meets the tangent. Experience shows that some of the vehicles
that fail to negotiate the bend will run off the road just beyond the tangent point. Always try and
close off any gaps through which vehicles may fall.
Page 146
Revision)
C le a r a n c e f o r
g u a r d r a il d e f le c t io n
H a z a rd S e t-b a c k
C le a r w id t h lim it
(n e a r s id e ) C le a r w id t h lim it
(o ff s id e )
C le a r z o n e
G u a r d r a il
C le a r z o n e
L e n g th o f n e e d L e n g th o f n e e d
( n e a r s id e ) ( o ffs id e )
R u n - o u t le n g t h R u n - o u t le n g th
R e f e r to T a b le 4 .4 in S e c t io n 4 .5 fo r Speed R u n -o u t le n g th
d e ta ils o f c le a r z o n e w id th s (k m /h ) (m e tre s )
60 50 - 60
N o t e : T h is is a s ta r t in g p o in t f o r
e s t im a t in g le n g t h o f n e e d . E n g in e e r in g 80 80 - 90
ju d g e m e n t s h o u ld b e u s e d t o d e t e r m in e 100 1 0 0 - 11 0
t h e p r e c is e le n g t h o f g u a r d r a il t o b e
in s ta lle d . > 11 0 120
Figure 9.2 Determining the length of need (adapted from Australian guidelines)
9.2.5 Steel beam strong post guardrail
Steel beam strong post guardrail is the most common type of safety barrier used in developing
countries, and is available from many sources. The precise design varies in detail, but the basic
characteristics are: (refer to Figure 9.3)
steel beams with a W-shape (this is the part that comes into contact with the vehicle)
the beams are 4130 m long
the beams are mounted on steel posts that are set either 1900 mm or 3800 mm apart
the beams are mounted so that the centre of the beam is 550 mm above the height of the road
surface
there is a steel spacer block between the post and the beam to prevent the vehicle from hitting
("snagging") on the post ("snagging" will usually cause the vehicle to spin round).
When an out-of-control vehicle hits the barrier the beam flattens, the posts are pushed
backwards, and the tension in the beam builds up to slow the vehicle and redirect it back onto the
road. That is if it performs successfully. The speed, mass and angle of the vehicle is critical to
success. With heavy vehicles, high angles of impact and very high speeds the barrier may be torn
apart or crushed. The containment capability can however be increased by using two beams, one
mounted above the other.
Page 147
Revision)
Typical Post Detail

STEEL SPACER BLOCK (150 MIN. WIDE)
FIX TO POST WITH TWO BOLTS TO
PREVENT ROTATION
600 MIN. 600 MIN.

THERE MUST BE AT
LEAST 1000 CLEARANCE
BETWEEN THE BACK OF
310
THE POST AND ANY
SOLID OBJECT
550
OUTER EDGE OF
SHOULDER
STEEL POSTS TO BE HAMMERED

IN PLACE BY APPROVED METHOD.
1100 MIN.
IF THIS IS NOT POSSIBLE THE
POSTS CAN BE CONCRETED IN.
130
Rear Elevation
BEAM LENGTH: 4130
SPLICE BETWEEN W-BEAM BACKUP PLATE

TWO BEAMS ON INTERMEDIATE
POST
1905
Figure 9.3 Steel beam strong post guardrail - typical details

9.2.6 Installation of steel beam strong post guardrail
The beams must be overlapped in the direction of travel, so that if they come apart in an
impact there is not an end that can spear the vehicle
The beams must be bolted together with eight bolts and the whole structure must be rigid
The beam centre must be 550 mm +-5 mm above the adjacent road surface - if it is lower,
vehicles may ride over it; if it is higher, vehicles may go under it
The spacer block must be fitted to the post with two bolts, otherwise it may rotate in a
collision
There must be two layers of beam at each spacer block, so at the intermediate posts (i.e.
those where there is no beam splice) insert a short section of beam between the main beam
and the spacer block - this is often called a backup plate and it helps to prevent the beam
hinging or tearing at this point
If the posts and spacers blocks are made of steel channel they must be installed so that the
flat side faces the traffic - this reduces the risk of injury if they are hit by a person who has
fallen from a vehicle
Page 148
Revision)
There must be a space of at least 1000 mm between the back of the post and any rigid
obstacle - this can be reduced to 500 mm if the barrier is stiffened by putting in extra posts
(at 952 mm centres), putting two beams together (one nested inside the other) and using
extra large concrete foundations
When installed on top of an embankment there must be at least 600 mm between the back of
the post and the break of slope in order to have sufficient ground support for the post - where
this is not possible, you must use much longer posts
Do not install the guardrail behind a kerb, because when a vehicle hits the kerb it will be
pushed upwards and so will hit the guardrail too high - with a risk that the vehicle will go
over the guardrail
Set the guardrail back from the shoulder edge (or carriageway edge if there is no shoulder)
by at least 600 mm - putting it at the edge of the shoulder reduces the effective width of the
shoulder and increases the risk of minor damage.
9.2.7 End treatment for guardrail

The end of a steel beam guardrail installation is a major hazard, as out-of-control vehicles can
become impaled on it resulting in serious injuries for the occupants. Do not leave short (<80m)
gaps in guardrail - instead make it continuous. There is no wholly safe way of terminating
guardrail but the general advice is:
stiffen the end section by installing the posts at 1905 mm spacing, AND
flare the end section of the guardrail away from the edge of the shoulder until it is offset by at
least 1m - use a flare rate of at least 1 in 10 - this reduces the risk of a direct impact; AND
use a special impact-absorbing terminal piece or ramp the beam down sharply into the
ground.
On a two-way road both the upstream and downstream ends of the guardrail will need to be
terminated in the above way. One of the problems of ramped ends is that they can launch out-of-
control vehicles into the air, with disastrous consequences. Try and avoid this by ramping the
beam down sharply. Flaring is an effective way of reducing the risk of impact but this can be
difficult to achieve in some situations, such as on narrow embankments.
1 :1 0 P a r a b o lic fla r e L = 11 4 0 0
P o s t s p a c in g = 1 9 0 5
11 4 0
E n d s e c tio n to
b e ra m p e d D ir e c t io n o f t r a v e l
dow n
(all dimensions in mm)
Figure 9.4 Flaring of the end of guardrail to reduce the risk of impact
Page 149
Revision)
9.2.8 Transition from guardrail to bridge parapets and concrete barriers

Collisions with the ends of bridge parapets and concrete barriers are usually very severe. It is
essential that these obstacles be shielded so that out-of-control vehicles will be redirected along
the face of the parapet or concrete barrier. This is best done by installing a semi-rigid steel beam
guardrail on the approach - normally at least 30 m long. It must line up with the face of the
parapet / barrier and be strongly connected to it. The guardrail must be progressively stiffened so
that deflection is reduced to zero as the parapet / barrier is reached. This is called a transition
section. The stiffening is achieved by putting in extra posts, putting two beams together (one
nested inside the other) and using extra large concrete foundations. See Figure 9.5. A steel
connecting piece is used to bolt the end of the guardrail to the parapet or barrier - the design of
this will vary to suit the design of the parapet / barrier.
L a s t 4 p o s ts C o n c r e te b a r r ie r
s e t in c o n c r e t e o r p a ra p e t
S te e l p o s ts
S ta n d a rd 1 .3 m p o s t s p a c in g 1 m p o s t s p a c in g C o n n e c ti n g p ie c e
p o s t s p a c in g b o lte d to b a r r ie r
D ir e c t io n o f t r a v e l
Figure 9.5 Typical transition (W-beam guardrail to rigid object)
9.2.9 Median barriers

High-speed dual carriageway roads with medians less than 9m wide may need to have median
barriers to reduce the risk of cross-over accidents and/or to provide protection against collision
with obstacles (e.g. lighting columns). Studies have shown that they may be cost-effective when
the ADT exceeds 10,000. Median barriers should not normally be used on urban dual
carriageways with speed limits of less than 80 km/h. If such roads have a cross-over problem it
should be tackled through speed reduction measures.
Median barriers often take the form of two guardrail beams mounted back to back on one post
see Figure 9.6. They are not suitable where the median is narrower than 2.0 m because they
deflect too much on impact. Rigid barriers made of reinforced concrete are preferred in this
situation see Figure 9.6. As always with safety barrier it is a problem to terminate it safely. If
possible the barrier should be terminated at points where speeds are low, such as at roundabouts.
Failing this, the guardrail beams should be flared and ramped down, or at least be capped with a
protective end-piece (bull-nose terminal). Concrete barrier should be ramped down.
Page 150
Revision)
24 0
R e in f o r c e d
800
c o n c re te
R o a d le v e l
210
290
6 50
Back to back guardrail Rigid concrete barrier
Figure 9.6 Median barriers
9.2.10 Concrete barrier

Concrete barriers are strong enough to stop most out-of-control vehicles, and being rigid there is
no deflection on impact. This makes them suitable for use on narrow medians and where it is
essential to keep vehicles on the road, such as at bridges. Small angle impacts usually result in
little damage to the vehicle. However, large angle impacts tend to result in major damage to the
vehicle, and severe injuries to the occupants. Research has shown that the conventional profile
(commonly called New Jersey Barrier) tends to cause small vehicles to overturn, and the
preferred shape is now a vertical or near-vertical wall (see Figure 9.6). Concrete barrier generally
requires very little routine maintenance except after very severe impacts.
The ends of concrete barrier are very hazardous, so every effort should be made to terminate the
barrier where speeds are low. The end of the barrier should be ramped down. If approach speeds
are unavoidably high the end of the barrier should be protected by fitting a section (of at least 30
m) of semi-rigid guardrail.
9.3 Kerbs
9.3.1 Function
Kerbs have a number of useful functions:
they define the edge of traffic lanes, traffic islands and footways during both day and night
(they reflect vehicle headlights)
they support pavements and island structures so that edge break-up is avoided
they protect adjacent areas from encroachment by vehicles
they assist in drainage of the carriageway.
9.3.2 Types of kerb and their application
The main types of kerb and their applications are listed below: (see also Figure 9.7)
Page 151
Revision)
Kerb with integral

Mountable kerb drain
20
50 200 100 330
50 150 200
R20 75 R20
40 R20 157
25
R20 20 R25
140 R20
80 100
350 40
R20
200X140 250X200 153
Barrier kerb 450X350

180 20
R20
130 20 Semi-mountable kerb
200
380
R20
125
80 120
250 R20 120
180
125 R20 30
150X250 200X380 300 150
250
200X300
125 Flush kerb

250X125
Figure 9.7 Types of kerbs
Barrier kerb
This kerb is used to provide protection to footways, traffic islands, pedestrian guardrail, traffic
signs, etc. Kerbs on footways should have a height of no more than 125 150 mm above road
level. If they are higher than this pedestrians may prefer to walk in the road. Barrier kerbs should
not normally be used on roads with vehicle speeds in excess of 70 km/h.
Semi-mountable kerb
This kerb can be used in rural situations where high speeds would make the use of barrier kerbs
risky. They are useful in defining and protecting the edges of the carriageway and traffic islands
at intersections.
Mountable kerb
This kerb is used to define traffic islands and road edges in urban and rural situations where there
is a high risk of the kerb being hit by vehicles.
Flush kerb
This kerb is used to protect and define an edge which can be crossed by vehicles.
Kerb with integral drain
This is a neat and effective way of providing drainage in urban areas, and it reduces the risk of
water penetration into the edge of the pavement. Other types of kerbs can be designed with
integral drains.
Page 152
Revision)
9.4 Bridge parapets

A parapet is a protective fence or wall at the edge of a bridge or similar structure. There are two
broad types:
Pedestrian parapets: These are designed to safeguard pedestrians but are not intended to contain
vehicles. These are used where there is a safety barrier between the vehicle lanes and the
footway (see Figure 4.14). Figure 9.8 shows a typical design for a lightweight steel parapet.
There should not be any openings wider than 100mm, in order to prevent small children from
squeezing through the parapet. Standard height is 1.0 m from the footway surface to the top of
the parapet, but this should be increased to 1400 mm if cyclists are present.
Vehicle / pedestrian parapets: These are designed to contain vehicles and safeguard pedestrians.
They are usually made either of metal (see Figure 9.8) or reinforced concrete. Vehicle /
pedestrian parapets must be designed to contain out-of-control vehicles on the bridge and to
deflect them back into the traffic lanes without severe deceleration or spinning. The design event
is usually taken to be a 1.5 t vehicle travelling at 80 km/h and hitting the parapet at an angle of
200. Well-designed reinforced concrete parapets have the greatest containment capability and
should be used where containment is essential. The basic design principles are:
Parapets should present an uninterrupted continuous face to the traffic i.e. no projections
Metal parapets should be designed with horizontal rails set in front of posts, so that vehicles
brush against the rails and do not snag on the posts
As far as possible the horizontal rails on metal parapets should be continuous if joints are
unavoidable they must be strengthened so that there is no risk of them coming loose on
impact and exposing free ends (like a spear)
The ends of the bridge rails must be linked with a strong connecting piece so that there are no
exposed ends
The standard height for parapets is 1m, but this should be increased to 1400 mm for
cycleways, and 1500 mm for bridges over railways and other bridges where containment is
essential
Metal parapets should have no openings wider than 100mm if necessary the parapet should
be faced with wire mesh panels
Parapets should be designed so that they are difficult to climb i.e. no footholds or flat tops
Where kerbs are located in front of parapets, the height of the parapet should be increased to
take account of dynamic jump effects.
Reinforced concrete parapets should take the form of solid, continuous walls with no openings. A
metal hand rail is often fitted along the top of the wall in order to improve the visual appearance.
In most cases there will need to be guardrail at both the approach and departure ends of the
parapet in order to prevent out-of-control vehicles hitting the end of the parapet. This is
particularly important with reinforced concrete parapets, because of their rigidity. The guardrail
can also prevent out-of-control vehicles from going the wrong side of the parapet and dropping
into the river / railway / etc below this is a common incident at bridges where the approach is
on a bend. The guardrail should be at least 30m long and should continue the line of the traffic
face of the parapet. Refer to Figure 9.5 for the design of the transition section between guardrail
and parapet. It is possible to design a metal parapet that incorporates w-beam guardrail, and this
has the advantage that the guardrail can be extended off the bridge to protect vehicles on the
Page 153
Revision)
approach sections. The containment capability can be increased by using two beams, one above
the other.
Figure 9.8 Vehicle / pedestrian parapet - typical details
9.5 Traffic islands

9.5.1 Function
Traffic islands are a key element in the design of safe, efficient intersections. They can be used
to:
separate conflicting traffic streams
control the path of vehicles and reduce unnecessary areas of carriageway
provide segregated lanes for some vehicle types or some traffic movements
warn drivers that they are approaching an intersection
provide shelter to vehicles that are waiting to make a manoeuvre
slow vehicles down by deflecting them from a straight ahead path
assist pedestrians to cross the road
locate traffic signs and signals where they will be at least risk of being hit.
9.5.2 Design requirements

Traffic islands must help drivers to recognise and follow a safe path through the intersection.
This calls for care in location, alignment, sizing and construction details. The key requirements
are:
Page 154
Revision)
of sufficient size to be easily seen (min. 4.5m2)

shape should take into consideration the wheel tracks of turning vehicles, the radii of left and
right turns, island nose radii, etc (use vehicle turning circle templates)
where islands are needed on high-speed rural roads consideration should be given to creating
them with road markings driver compliance can be encouraged by infilling them with
rumble strips
where kerbed islands are to be provided on high-speed rural roads the kerbs should
preferably be of the mountable or semi-mountable type
kerbed islands can be made more visible by painting the kerbs black and white (500mm
sections)
pedestrian refuges should normally have barrier kerbs and be 1.5 m wide (1.2 m absolute
minimum)
traffic signs (typically sign no. R75 Pass Left Side Only) should be used on kerbed islands,
and they should be positioned so that there is at least 300 mm clearance between the edge of
the sign and the traffic face of the kerb
the nose at the approach end of an island should have a minimum radius of 0.6 m at other
corners the radius can be as small as 300 mm
on high-speed roads it is advisable to offset the edges of islands from the edge of the through
traffic lane by 0.6 m 0.9 m, to reduce the risk of collisions
road markings (typically sign no. M53 Traffic Island Marking) should be used to guide
drivers safely past the island.
9.6 Pedestrian barrier

9.6.1 Function
Uncontrolled pedestrian movements are a significant factor in urban traffic and safety problems.
Pedestrian barrier can bring big improvements by segregating pedestrians from vehicular traffic
and channelling them to safe crossing points. At intersections, barrier can:
reduce conflicts by channelling pedestrians to crossing points on the approaches
discourage buses, minibuses and boda-boda from stopping and parking within the
intersection
discourage delivery vehicles from loading or unloading within the intersection
discourage roadside vendors from occupying the road space in the intersection
Other applications include:

Schools barrier can be used to prevent children from running into the road from the school
gate
Bus parks, cinemas, stadiums, etc barrier can channel pedestrian flows at areas of heavy
pedestrian movement
Pedestrian crossings, subways, footbridges barrier helps channel pedestrians to the crossing
facility
Medians barrier can be used to deter pedestrians from using the median to cross the road,
though barriers on the footways are more likely to be effective.
Page 155
Revision)
Nobody likes walking further than they have to, so, although pedestrians can be guided to a
certain extent, do not try and force them to make unreasonably long detours. People will smash
the barrier if it is seen as too much of an obstacle.
9.6.2 Design
Ideally, pedestrian barrier should:
be effective
be strong and easily maintained
cause minimum damage to vehicles and the occupants when hit
not be hazardous to pedestrians, including the disabled
not interfere with visibility
look acceptable.
The choice is between steel railings and brick or concrete walls. Steel railings should be
designed with a minimum number of horizontal elements, because these are potentially
hazardous in a vehicle collision. The design principles are similar to those for steel bridge
parapets see Figure 9.8 with the added requirement for them to be see-through. The railings
should be about 1.0 m high and be rustproof. Brick or concrete walls are likely to be cheaper and
easier to maintain but they take up more space. All barriers should be set back (normally 300
mm) from the traffic face of the kerb to give adequate clearance for passing vehicles. Ends of
pedestrian barrier may need to be fitted with reflectors to make them less of a hazard at night.
9.7 Street Lighting

Street lighting contributes to road safety and improves personal security. It should be standard on
all main roads in urban areas. Priority should also be given to lighting areas with a high
proportion of night-time pedestrian accidents, such as bus parks, stadia, and entertainment
centres. The provision of street lighting can also improve safety at isolated major intersections in
rural areas.
Lighting columns can be a hazard to out-of-control vehicles. The lighting scheme should aim to
minimise the number of lighting columns and ensure that the poles are not located in vulnerable
positions. Standards for lateral clearance, clear zones, etc. must be respected.
Page 156
Revision)
10 PAVEMENT DESIGN (NO REVISION)
11 HYDROLOGY AND DRAINAGE (NO REVISION)
/var/www/apps/conversion/tmp/scratch_2/356011162.doc
Page 157
Revision)

Design Manual Ver 14设计守则

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Design Manual Ver 14设计守则

Hochgeladen von

Copyright:

Verfügbare Formate

THE REPUBLIC OF UGANDA

MINISTRY OF WORKS, HOUSING AND COMMUNICATIONS

ROAD DEVELOPMENT PROGRAM PHASE 2 PROJECT

CONSULTANCY SERVICES FOR THE PRODUCTION OF

Road Design Manual

Client: Lead Consultant:

Figure 5.26 Clothoid elements 72

Road Design Manual (RDM) Traffic Safety Revision

0.1 Main Requirements

0.2 Safety Considerations

50 km/h in a side-on vehicle/vehicle or vehicle/object collision

Reducing the severity of accidents

0.3 Environmental considerations

Effects related to the road as a physical feature

Effects related to the traffic

0.4 Economic considerations

0.4.2 Route selection

0.4.3 Detailed design

1 GENERAL (NO REVISIONS)

2 ROAD SYSTEM (NO REVISIONS)

Restrictions are mainly imposed by:

3.2 Basic Prerequisites

Expected traffic volumes.

3.2.1 Road category

Road Desired rural

Table 3.1 Functional road categories

3.2.2 Expected traffic volumes

Motor vehicle traffic

Table 3.2 Vehicle traffic volume measures

DHV = AADT * K or ADT * K

Table 3.3 Design Hour Volumes percentage

Volumes of vulnerable road users

Table 3.4 Classification of pedestrian volume

3.2.3 Terrain (topography) and geotechnical conditions

Table 3.5 Types of terrain

3.2.4 Road environment

Minor trading centre Major trading centre

Figure 3.2 Road environments

3.3 Design Criteria

3.3.1 Speed limits and design speed

Speed change (%)

Figure 3.3 Speed power law

Speed limits and design speed

Table 3.6 Normal speed limits

3.3.2 Separation of vulnerable road users from motorised traffic

Level D 55 km/h 1880 pcu/h Unstable flow

Table 3.7 Definition of levels of service

No Volume-capacity ratio (V/C) (2)

Table 3.8 Level of service criteria for two-lane roads

Selection of level of service

Table 3.9 Selection of level of service

Determination of level of service

Table 3.10 Lane width factor

Directional distribution 50/50 60/40 70/30 80/20

Table 3.12 Directional factor for two-lane roads

Level of Maximum AADT

Table 3.13 Maximum AADT for two lane rural roads

3.3.4 Design vehicles

The principal design elements affected are:

Passenger car 2.9 m 0.9 m 1.2 m 5.0 m 1.9 m -

Semi trailer combination see figure below 16.0 m 2.5 m 3.8 m

Table 3.14 Design vehicles

(adapted from EU agreements on vehicle manoeuvrability)

(adapted from EU agreements on vehicle manoeuvrability)