Skip to main content
main-content
Top

About this book

This third edition of the late R.J. Salter's successful book has been revised and updated by N.B. Hounsell. Part I covers transportation planning, incorporating new methodological approaches and models. Part II covers highway traffic analysis and design, including updated sections on link and junction design, together with new computer aided design packages. Part III concentrates in traffic signals, with new chapters on microprocessor-based signal control and modern urban traffic control systems. This new edition consolidates the book's position as a practical text of traffic theory and practice, including many worked examples, for undergraduate and postgraduate students of transport and traffic engineering.

Table of Contents

Traffic Surveys and Prediction

Frontmatter

1. Introduction to the transportation planning process

Abstract
Large urban areas have in the past frequently suffered from transportation congestion. It has been recorded that in the first century vehicular traffic, except for chariots and official vehicles, was prohibited from entering Rome during the hours of daylight. While congestion has existed in urban areas the predominantly pedestrian mode of transport prevented the problem from becoming too serious until the new forms of individual transport of the twentieth century began to demand greater highway capacity.
R. J. Salter

2. The transportation study area

Abstract
The area of a country covered by a transportation study will depend upon the geographical distribution of the trip patterns which are of particular interest to the study organisers. The study area has been defined as the area within which trip patterns will be significantly affected by the implementation of any resulting transport proposals. It is important that sufficient thought be given to the boundary of the area because within the boundary the transport information is collected in considerable detail and the resulting cost of data collection will be largely related to the location of the boundary.
R. J. Salter

3. The collection of existing travel data

Abstract
Existing tripmaking or travel data may conveniently be classified into four groups according to the origin and the destination of the trip being considered. These four classes of movement are:
(a)
trips which have an origin within the external cordon and a destination outside the external cordon;
 
(b)
trips which have both an origin and a destination within the external cordon;
 
(c)
trips which have an origin outside the external cordon and a destination within it;
 
(d)
trips which have neither origin nor destination within the external cordon but which pass through it. Reference
 
R. J. Salter

4. The external cordon and screenline surveys

Abstract
The object of the external cordon survey is to obtain information on the trips originating outside the external cordon having origins within the external cordon or passing through the survey area.
R. J. Salter

5. Other surveys

Abstract
This survey is designed to measure the trips made by commercial vehicles within the internal area. The data is obtained by issuing drivers with forms on which they record each trip, together with trip origin, destination, purpose and parking details. The size of the sample depends on the size of the survey area and also on the variability of goods-use activity, ranging from 100 per cent for a small town to about 25 per cent for a large area. The population may be obtained from the land-use survey, which will indicate addresses at which commercial vehicles may be kept or alternatively excise licence records may be consulted. A typical questionnaire is reproduced in figure 5.1.
R. J. Salter

6. Trip generation

Abstract
Once the transportation survey has collected all the details of the existing trip-making pattern and the socio-economic, land-use and transportation-system characteristics of the survey area, the second stage in the transportation planning process is the development of relationships between the total number of trip origins and destinations in a zone and the total characteristics. It is assumed that these relationships will be true in the future and so, if land-use and socio-economic factors can be predicted, future trips can be estimated for any proposed transport system.
R. J. Salter

7. Trip distribution

Abstract
Trip distribution is another of the major aspects of the transportation simulation process and although generation, distribution and assignment are often discussed separately, it is important to realise that if human behaviour is to be effectively simulated then these three processes must be conceived as an interrelated whole.
R. J. Salter

8. Modal split

Abstract
Trips may be made by differing methods or modes of travel and the determination of the choice of travel mode is known as modal split. In the simplest case when a small town is being considered the choice is normally between one form of public transport and the private car, with the car being used for all trips where it is available. In such a situation most trips on the public transport network are captive to public transport and very little choice is being exercised. In the larger conurbations however the effect of modal split is of very considerable significance and is greatly influenced by transport policy decisions.
R. J. Salter

9. Traffic assignment

Abstract
Previously the estimation of generated trip ends has been discussed together with the distribution of trips between the traffic zones. Modal split methods also have been reviewed in which the proportion of trips by the varying travel modes are determined. At this stage the number of trips and their origins and destinations are known but the actual route through the transportation system is unknown. This process of determining the links of the transportation system on which trips will be loaded is known as traffic assignment.
R. J. Salter

10. The evaluation of transportation proposals

Abstract
The object of the simulation of the land-use/transportation process is to estimate the trips that will be attracted to a proposed future transportation system for a given pattern of land-use development. To compare effectively different transportation proposals it is imperative that each proposal should be evaluated.
R. J. Salter

Analysis and Design for Highway Traffic

Frontmatter

11. The capacity of highways between intersections

Abstract
The capacity of a highway may be described as its ability to accommodate traffic, but the term has been interpreted in many ways by different authorities. Capacity has been defined as the flow which produces a minimum acceptable journey speed and also as the maximum traffic volume for comfortable free-flow conditions. Both these are practical capacities while the Highway Capacity Manual1 defines capacity as the maximum hourly rate at which persons or vehicles can reasonably be expected to traverse a point or uniform section of a lane or roadway during a given time period under prevailing roadway, traffic and control conditions. The time period used in most capacity analysis is 15 minutes which is considered to be the shortest interval during which stable flow exists.
R. J. Salter

12. Headway distributions in highway traffic flow

Abstract
The concept of level of service in highway traffic flow illustrates the differences in the characteristics of the flow which may be examined by a study of the headways between vehicles. Time headways are the time intervals between the passage of successive vehicles past a point on the highway. Because the inverse of the mean time headway is the rate of flow, headways have been described as the fundamental building blocks of traffic flow. When the traffic flow reaches its maximum value then the time headway reaches its minimum value.
R. J. Salter

13. The relationship between speed, flow and density of a highway traffic stream

Abstract
Theoretical relationships between speed, flow and density When considering the flow of traffic along a highway three descriptors are of considerable significance. They are the speed and the density or concentration, which describe the quality of service experienced by the stream; and the flow or volume, which measures the quantity of the stream and the demand on the highway facility.
R. J. Salter

14. Traffic speed distributions and estimation

Abstract
One of the fundamental parameters for describing traffic flow is the speed, either of individual vehicles or of the traffic stream. It is of importance in work connected with the theory of traffic flow because of the fundamental connection between speed, flow and concentration when the movement of a traffic stream is being considered.
R. J. Salter

15. Highway link design

Abstract
The geometric design of new road links in the United Kingdom is now undertaken in accordance with the Department of Transport Departmental Standard TD9/93 ‘Highway Link Design’. The overall requirement is that an alignment is provided for a design speed which is consistent with the anticipated vehicle speeds on the road.
R. J. Salter

16. Intersections with priority control

Abstract
Intersections are of the greatest importance in highway design because of their effect on the movement and safety of vehicular traffic flow. The actual location of intersection is determined by siting and design and the act of intersection by regulation and control of the traffic movement. At an intersection a vehicle transfers from the route on which it is travelling to another route, crossing any other traffic streams which flow between it and its destination. To perform this manoeuvre a vehicle may diverge from, merge with, or cross the paths of other vehicles.
R. J. Salter

17. Driver reactions at priority intersections

Abstract
Interaction between traffic streams is an important aspect of highway traffic flow. It occurs when a driver changes traffic lane, merging with or crossing a traffic stream. Probably it takes place most frequently when priority control is used to resolve vehicular conflicts at highway intersections.
R. J. Salter

18. Capacities and delays at priority intersections

Abstract
In contrast to traffic signal-controlled junctions, where saturation flows and resulting junction capacities can be easily calculated, the estimation of the practical capacity of priority type junctions presents considerable difficulties.
R. J. Salter

19. A simulation approach to delays at priority intersections

Abstract
Prediction of capacities and delays at highway intersections under priority control may be made by the use of empirical, mathematical or simulation models.
R. J. Salter

20. Roundabout intersections

Abstract
At low/medium levels of traffic flow the control of traffic movements at intersections may be achieved by priority control. As has been discussed previously the form of priority control in the UK is that minor road vehicles give way to major road vehicles. On the continent of Europe nearside priority is sometimes used where vehicles give way to traffic approaching from the right while in Australia off-side priority is used.
R. J. Salter

21. Grade-separated junctions and interchanges

Abstract
Grade-separated junctions have been defined1 as ones which require use of an at-grade junction at the commencement or termination of the slip roads. This at-grade junction, in the form of either a major—minor priority junction or a roundabout together with the slip roads, can produce a diamond junction, a half-cloverleaf junction or a roundabout junction.
R. J. Salter

22. Merging, diverging and weaving at grade-separated junctions and interchanges

Abstract
Merging, diverging and weaving occur at and between many types of junctions in both urban and rural situations. Most frequently these traffic actions are associated with junctions on routes with a high standard of geometric design such as is found on the British motorway and trunk road system.
R. J. Salter

23. Queueing processes in traffic flow

Abstract
Queueing theory originally developed by A. K. Eriang in 1909 has found widespread application in the problems of highway traffic flow1.
R. J. Salter

24. Geometric delay at non-signalised intersections

Abstract
Delay to vehicles at intersections is an important factor in the choice of intersection type and frequently attention is focused on queueing delay which occurs at peak flow times. There is however a further source of delay which is found regardless of traffic conflicts, caused by vehicles slowing down and subsequently accelerating as they negotiate the junction. As this form of delay is determined by the size and shape of the junction, it is referred to as ‘geometric delay’. Because geometric delay takes place throughout the whole of the day it is often a substantial portion of the whole of the delay.
R. J. Salter

25. The environmental effects of highway traffic noise

Abstract
In our industrial society the number of sources of sound are steadily increasing and when these sounds become unwanted they may be classed as noises. Sound is propagated as a pressure wave and so an obvious measure of sound levels is the pressure fluctuation imposed above the ambient pressure.
R. J. Salter

26. The environmental effects of highway traffic pollution

Abstract
During recent years there has been a widespread attempt to reduce air pollution from all sources. In the United Kingdom the Clean Air Act of 1956 has resulted in a noticeable decrease in coal consumption and a reduction in air pollution from domestic and industrial sources. During this same period there has been a marked increase in the volume of road traffic and consequently an increase in pollution from this source. The National Society for Clean Air1 estimates that, during 1956 in the United Kingdom, the total coal and oil consumed was equivalent to 276 million tons (coal equivalent) and only 26 million tons of this were used for road or rail transport. It is however an increasing source of pollution, which is emitted in situations close to human activity. Approximately one-third of the carbon monoxide in the atmosphere is produced from vehicle exhausts.
R. J. Salter

27. Traffic congestion and restraint

Abstract
The motor car is an invention which, within half a century, has revolutionised our way of life. It has made possible a dispersal of dwellings far exceeding that of the railway age, it has offered a wide choice of employment situations and has increased the scope of recreational activities to a remarkable extent.
R. J. Salter

Traffic Signal Control

Frontmatter

28. Introduction to traffic signals

Abstract
Traffic signals are used at many at-grade junctions, particularly in urban areas, to maximise traffic efficiency and safety by separating conflicting traffic movements in time.
R. J. Salter

29. Warrants for the use of traffic signals

Abstract
A decision to use signal control in urban areas in preference to roundabout control or as a means of increasing traffic capacity at priority intersections may be made from the overall viewpoint of traffic management when urban computer control is being implemented.
R. J. Salter

30. Staging and phasing

Abstract
In the control of traffic at signal-controlled intersections, conflicts between streams of vehicles are prevented by a separation in time. Green signals are therefore shown to different sets of movements (or streams), so that conflicting ones do not receive a green signal simultaneously unless permitted in some circumstances (such as opposed right-turning traffic). Traffic streams therefore have to be divided into separate sets so that all streams in each set always receive identical signal indications. This procedure by which the streams are separated is known as ‘phasing’.
R. J. Salter

31. Signal aspects and the intergreen period

Abstract
The indication given by a signal is known as the signal aspect. The usual sequence of signal aspects or indications in Great Britain is red, red/amber, green and amber. The amber period is standardised at 3 s and in all new signal installations the red/amber at 2 s.
R. J. Salter

32. Signal control strategies

Abstract
Signal control strategies are implemented at the junction by the traffic signal controller which may be isolated (stand alone) or linked either locally to one or more adjacent junctions or, in the case of Urban Traffic Control, to a central computer which controls all the traffic signals. Recently manufactured signal controllers use electronic timing to determine the length of the signal stages and electromechanical interlocking relays to switch the signals on and off in the specified sequence. Signal timings can be adjusted according to vehicle detection, remote computers (UTC) or nearby linked controllers, or varied according to time of day.
R. J. Salter

33. Geometric factors affecting the capacity of a traffic signal approach

Abstract
The capacity of a signal-controlled intersection is limited by the capacities of individual approaches to the intersection. There are two types of factors which affect the capacity of an approach: geometric factors which are considered in this chapter, and traffic and control factors which are discussed subsequently.
R. J. Salter

34. The effect of traffic factors on the capacity of a traffic signal approach

Abstract
The effect of traffic factors on the capacity of a traffic signal approach is usually allowed for by the use of weighting factors, referred to as ‘passenger car units’, assigned to differing vehicle categories. As a consequence the saturation flow of a signal approach or of a single approach lane is expressed in passenger car units per hour (pcu/h).
R. J. Salter

35. Determination of the effective green time

Abstract
In Chapter 34 the concept of effective green time was introduced as a means of determining the number of vehicles that could cross a stop line over the whole of the cycle comprising both red and green periods.
R. J. Salter

36. Optimum cycle times for an intersection

Abstract
The length of the cycle time under fixed-time operation is dependent on traffic conditions. Where the intersection is heavily trafficked cycle times must be longer than when the intersection is lightly trafficked.
R. J. Salter

37. The timing diagram

Abstract
Previously that optimum cycle time has been calculated which would result in minimum overall delay when employed with fixed-time signals. The intergreen periods have also been selected previously and the remaining calculation, before the whole sequence of signal aspects can be described, is to calculate the duration of the green signal aspects.
R. J. Salter

38. Early cut-off and late-start facilities

Abstract
Where the number of right-turning vehicles is not sufficient to justify the provision of a right-turning phase but where right-turning vehicles have difficulty in completing the traffic movement, then an early cut-off or a late-start of the opposing phase is employed.
R. J. Salter

39. Opposed right-turning vehicles and gap acceptance

Abstract
Right-turning vehicles present particular problems at signal-controlled intersections. On a signal approach, right-turn vehicles may be given an exclusive right-turn lane or they may be mixed with straight ahead vehicles. If vehicles do not have to give way, that is, the flow is unopposed, then the maximum discharge or saturation flow can be calculated using the relationships given in Chapter 33. When the flow is opposed and the right-turn vehicles are mixed with straight ahead vehicles then once again the relationships given in Chapter 33 can be used to calculate the saturation flow. If however the right-turn flow is opposed and is not mixed with straight ahead vehicles then the following simplified approach to the calculation of the right-turn discharge can be used. In this case some right-turn vehicles are able to turn through gaps in the opposing straight ahead flow and the remainder must turn in an early cut-off intergreen period at the end of the green period.
R. J. Salter

40. The ultimate capacity of the whole intersection

Abstract
It was explained in Chapter 35 that the capacity of an approach is dependent on the lost time during the cycle. When the whole intersection is considered the capacity is also dependent on the total lost time on all the phases because the remainder of the time is shared equally between the phases and used as running time.
R. J. Salter

41. The optimisation of signal-approach dimensions

Abstract
The procedures previously outlined have optimised cycle time and green times so as to produce minimum overall delay for an intersection where the physical dimensions of the approach highways were fixed.
R. J. Salter

42. Optimum signal settings when saturation flow falls during the green period

Abstract
On some traffic-signal approaches it cannot be assumed that saturation flow will remain constant throughout the greater part of the green period. The effect of blocked right-turning movements and of approaches that are wider at the stop line than on the remainder of the approach is to reduce the flow rate over the stop line as the green period proceeds.
R. J. Salter

43. Delay at signal-controlled intersections

Abstract
Two approaches are in common use for calculating traffic delays at signal-controlled junctions. The first method, based on steady-state queueing theory, is relevant to situations of relatively constant traffic demand where demand does not exceed about 90 per cent of capacity. More recently, time-dependent queueing theory has been developed, applicable to all traffic states including time-varying demand and ‘oversaturation’.
R. J. Salter

44. Average queue lengths at the commencement of the green period

Abstract
In the design of traffic signals it is often desirable to be able to estimate the queue length at the beginning of the green period. The queue length at this period of the cycle will normally be the greatest experienced because during the green period the queue is being discharged at a rate which for practical purposes must be greater than the flow on the approach.
R. J. Salter

45. Programs for traffic signal design

Abstract
A number of computer programs have become available in recent years as an aid to the design of signal-controlled junctions. These can be used to help determine the optimum geometric design of the junction, calculate appropriate signal settings for control purposes and predict queues and delays at the junction. Such programs are widely used and are a valuable aid to the traffic engineer, but do not remove the need for experienced design engineers. For example, the choice of stage sequencing, phase combinations, geometric details and so on still require the skill of the designer; the signal programs then provide a powerful tool for evaluating different alternatives and for optimisation. Other issues outside the scope of signal programs, but vital in the design process, include:
(1)
‘designing for safety’ (for example, correct intergreen/minimum green settings and minimisation of conflicts);
 
(2)
adequate provision for all road users, including heavy vehicles (for example minimum turning circles), buses (for example, bus stops, priority facilities) and cyclists/pedestrians (for example, special protection/signalling facilities);
 
(3)
other design issues, such as junction location, signing, lighting and environmental impact.
 
R. J. Salter

46. The co-ordination of traffic signals

Abstract
When several traffic signal-controlled intersections occur along a major traffic route, some form of co-ordination is necessary to prevent, so far as is possible, major road vehicles stopping at every intersection. Alternatively, or in addition, the signals may be co-ordinated to minimise delays to vehicles. Sometimes linking between signals is carried out to prevent queues stretching back from one intersection to the preceding signals.
R. J. Salter

47. Time and distance diagrams for linked traffic signals

Abstract
When the flexible progressive system of co-ordinating traffic signals is employed, then it is frequently desirable to construct a time-arid-distance diagram to estimate the best offset or difference in the start of the green time, between adjacent signals.
R. J. Salter

48. Platoon dispersion and the linking of traffic signals

Abstract
A trial-and-error approach such as is involved in the preparation of a time-and-distance diagram can produce reasonable progression along a major traffic route. Where however it is desired to minimise delay at signal-controlled intersections in a network a more rigorous approach is desirable.
R. J. Salter

49. The prediction of the dispersion of traffic platoons downstream of signals

Abstract
The minimisation of delay by the adjustment of the offset of the green signal depends on the ability to predict the arrival rate of vehicles at the stop line. In a network controlled by traffic signals the platoons of vehicles travelling towards the stop line will have been discharged initially from a traffic signal upstream. For this reason the prediction of the dispersion of vehicles downstream from traffic-signal approaches is of considerable importance.
R. J. Salter

50. The delay/offset relationship and the linking of signals

Abstract
By the use of the technique of calculating the delay to vehicles on a traffic-signal approach from the difference between the demand and service distributions, as discussed in Chapter 48, and the prediction of the demand distribution from a knowledge of platoon diffusion relationships as described in Chapter 49, it is possible to obtain a relationship between the offset of one signal relative to another and the delay to vehicles passing through both signals.
R. J. Salter

51. Urban traffic control systems

Abstract
During the late 1950s, proposals were made that the linking and co-ordination of traffic signals, which had already been applied to produce green waves of traffic along major routes, could be extended to a network of highways over an area. Originally referred to as the area ‘control of traffic’, co-ordination over a network is now more usually termed ‘urban traffic control’.
R. J. Salter
Additional information