Traffic flow theory

(Jan Valentin)

Introduction
The knowledge of traffic characteristics is useful to the road engineers in developing roads, motorways and transportation plans, performing economic analyses, establishing geometric design criteria, selecting and implementing traffic control measures and evaluating the performance of transportation facilities. In this lecture presented traffic flow characteristics traffic speed, travel time, volume and density are fundamental for planning, design and operation of roads and motorways (highways) and transport facilities. Determination of relationships between concentration, density, speed and volume is of primary interest in traffic flow theory, which involves the development of mathematical relationships among the primary elements of a traffic stream flow, speed, density. The relationships help the traffic engineer in planning, designing and evaluating the effectiveness of implementing traffic-engineering measures on a road or highway system. The basic for further analyses are data collecting on several elements of traffic stream. One example of the use of traffic flow theory in design is the determination of adequate lane lengths for storing left-turn vehicles on separate left-turn lanes. The determination of average delay at intersections and freeway ramp merging areas is another example of the application of traffic flow theory. Another important application of traffic flow theory is simulation, where mathematical algorithms are used to study the complex interrelationships that exist among the elements of a traffic stream or network to estimate the effect of changes in traffic flow on factors such as accidents, travel time, air pollution and fuel consumption. Occasionally, single vehicles traverse the transportation facilities without significant interference from other vehicles. But the same facilities also experience simultaneous usage by streams of vehicles. The resulting traffic conditions range from almost free flow when only a few relatively unconstrained vehicles occupy a roadway to highly congested conditions when the roadway is jammed with slow-moving vehicles. The determinant of these traffic flow models is the car-following rule adopted by drivers in an attempt to maximize their speeds while maintaining an acceptable level of safety. They accomplish this by adjusting the distance between vehicles, depending on their speed. The basic variables that describe the prevailing conditions within a vehicular stream (flow, concentration, and mean speed) are introduced and the fundamental relationship between the three stream variables is postulated and applied to several traffic phenomena, including the propagation of shock waves in traffic. Consider the case of vehicles following each other on a long stretch of roadway or guideway. Furthermore, assume that these vehicles are not required to interrupt their motion for reasons that are external to the traffic stream, such as traffic lights, transit stations, and the like. In this case of uninterrupted flows the only interference that a single vehicle experiences is caused by other vehicles on the roadway. Figure 1 shows two typical stream vehicles traveling at a speed v and a spacing s between the front of the leading vehicle to the front of the following vehicle. As a general rule, the spacing between vehicles should be such that if a sudden deceleration becomes necessary for a leading vehicle, the following vehicle has ample time and distance to perceive the situation, react to it, and be able to decelerate safely without colliding with the stopping, leading vehicle. Figure 1(a) shows the locations of the leading and following vehicles described above at the moment when the leading vehicle begins to decelerate, and Fig. 1(b) shows the limiting acceptable conditions at the end of the stopping maneuver of the following vehicle. Parenthetically, the term vehicle may be taken to mean a vehicular train consisting of a number of articulated vehicles rather than a single vehicle. Using the following notation, a relationship between spacing, speed, and deceleration (assumed constant) can be developed:
1

s = v +
v dl df x0 L N

v2 v2 + NL + x0 2d f 2d l

(1)

= initial speed of two vehicles = deceleration rate of the leading vehicle = deceleration rate of the following vehicle = perception-reaction time of the following vehicle = safety margin after stop = length of vehicle = number of vehicles in a train

Figure 1 The combined choice of particular values for these variables has some important implications with respect to the level of safety provided by the systems operation. Three particular values of deceleration are relevant to the operations safety level: dn de = normal or comfortable deceleration = emergency deceleration = instantaneous or stonewall stop

The higher the level of safety the higher the required spacing has to be just to avoid a collision. On this basis alone it would seem reasonable to choose the safest level of operation. However, by increasing the level of safety, the capacity of the system (i.e. the maximum number of vehicles or passengers that can be accommodated during a given period of time) suffers. Consequently, a trade-off between safety and capacity exists.

Normal deceleration is related to passenger comfort. The instantaneous stop condition may arise when an accident or a stalled vehicle or other obstruction suddenly comes within the perception field of the subject vehicle. The safest level of operation occurs when the spacing between vehicles is such that the following vehicle can safely stop by applying normal deceleration even when the leading vehicle comes to a stonewall stop. A lower level of safety results when the spacing is selected so that the following vehicle would have to apply the emergency brake rather than normal deceleration in order to avoid a collision. The combination of leading and following vehicle decelerations designate various safety regimes.

Figure 2 Various safety regimes (relationship between spacing and speed)

1. Stream variables
Spacing and headways There are many situations that engineers encounter for which it is necessary to consider the behavior of individual vehicles in the traffic stream rather than the probability of delay and average delay for vehicles or pedestrians crossing a traffic stream and predicting the length of waiting lines at toll booths, traffic signals and entrances to parking facilities. Spacing and speeds of the vehicles make up the stream. Two measures are therefore of fundamental importance in traffic stream calculations spacing and time-headway between successive vehicles. The spacing is simply the distance between successive vehicles, typically measured from front bumper to front bumper. It is the reciprocal of density. Time-headway is the time between the arrival of successive vehicles at a specified point and it is the reciprocal of volume. For many light traffic situations, traffic can be described by the Poisson probability distribution:

m x e m P( x) = x!
where P(x) =

(2)

m V t

= = =

the probability that exactly x randomly arranged vehicles will be observed in a unit length of road, or the probability of arrival of exactly x vehicles in a unit length of time Vt/3600 = the average number of vehicles arriving in an interval of length t traffic volume (vehicle per hour) length of time interval (s)
3

It should be emphasized that the Poisson exponential distribution is applicable to random or free-flow traffic situations. More complex distributions have to be used for description of a wider variety of traffic flow conditions. Traffic concentration Traffic density, also referred to as traffic concentration k of the vehicular stream, is defined as the average number of vehicles occupying a unit length of roadway at a given instant or the ratio of the number of vehicles appearing on a photograph to the length of the roadway segment. This is an instantaneous measurement taken at the instant when the picture was taken. The dimensions of concentration are given in terms of vehicles per length of roadway. Generally it is expressed in units of vehicles per kilometer at an instant in time. Traffic density bears a functional relationship to speed and volume. Concentration is in the present recommended as the basic parameter for describing the quality of flow along freeways and other multilane roads. The relationship between spacing (or average spacing when not constant) and concentration is:

s=

1 k

(3)

Consider a stationary observer next to the roadway. Vehicles pass the observers location one after another at intervals of time defined as the headways h between vehicles. In the simple example described above, the headway between vehicles is constant and can be computed by dividing the constant spacing by the constant speed of system operation. Time headway (h) is the difference between the time the front of a vehicle arrives at a point on the highway and the time the front of the next vehicle arrives at that same point. Time headway usually is expressed in seconds. Space headway spacing (s) is the distance between the front of a vehicle and the front of the following vehicle. Traffic volume and rate of volume The number of vehicles counted at the point of observation (point along a roadway or traffic lane) divided by the total observation time (equivalent hourly rate at which vehicles pass a point on a highway lane during a time period) is defined as the stream flow q sometimes referred s volume V and measured in vehicles per unit time. A measure of the quantity of stream flow is commonly measured in units of vehicles per day, vehicles per hour, etc. Flow is a measurement at a point on the roadway over time. The relationship between headway and flow is

h=

1 q

(4)

Volume is then the total number of vehicles that pas a point on a highway during a given time interval. When the time interval is one hour, the unit of volume is vehicle per hour. In some circumstances, volume is also given as the number of passenger cars per hour, with all trucks and buses counted as an equivalent number of passenger cars. For example in the USA the Highway Capacity Manual gives equivalent factors for trucks and buses for varying types of roads, terrain conditions and velocities. Two measures of traffic volume are of special significance to the engineers average daily traffic (ADT) and design hourly volume (DHV). ADT is the number of vehicles that pass a particular point on a roadway during a period of 24 consecutive hours averaged over a period of 365 days. It is a fundamental traffic
4

measurement needed for the determination of vehicle-kilometers of travel on the various categories of rural and urban road systems. ADT values for specified road sections provide the engineer, planner an administrator with essential information needed for the determination of design standards, the systematic classification of roads as well as for the development of programs for improvement and maintenance. It is not feasible to make continuous counts 365 days a year along every section of a road system. ADT values for many road sections are therefore based on a statistical sampling procedure.

Figure 3. Traffic volume fluctuation (Source: Georgia Department of Transportation).

DHV is a future hourly volume that is used for design. Traffic volumes are much heavier during certain hours of the day or year, and it is for these peak hours that the road is designed. In order to design a road properly, it is necessary to know the capacity that must be provided in order to accommodate the known traffic volume. On a given roadway, the volume of traffic fluctuates widely with the time. Such variations during a day, week or year tend to be cyclical and to some extent predictable. The nature of the pattern of variation depends on the type of road facility. Urban arterial flow is, for example, characterized by pronounced peaks during the early morning and late afternoon hours, due primarily to commuter traffic. Rural roads tend to experience less pronounced daily peaks, but they may accommodate heaviest traffic flow on weekends and holidays because of recreational travel. As might be expected, the seasonal fluctuations are most pronounced for rural recreation routes. The term rate of flow accounts for the variability or the peaking that may occur during periods of less than one hour. The term is used to express an equivalent hourly rate of vehicles passing a point along a roadway or for traffic during an interval less than one hour (usually 15 minutes). The distinction between volume and rate of flow can be illustrated as follow (suppose the traffic counts were made during a study period of one hour): Time period 8:00 8:15 8:15 8:30 8:30 8:45 8:45 9:00 TOTAL Number of vehicles 1000 1100 1000 900 4000 Rate of flow (vehicles/hr) 4000 4400 4000 3600

The total volume is the sum of these counts or 4000 vehicles per hour. The rate of flow varies for each 15min period and during the peak period its value is 4400 vehicles per hour. Note that 4400 vehicles did not actually pass the observation point during the study hour, but they did not pass at that rate for one 15-min period. Suppose the example roadway section is capable of handling a maximum rate of only 4200 vehicles per hour. In other words, its capacity is 4200 vehicles per hour. Since the peak rate of flow is 4400 vehicles per hour, an extended breakdown in the flow would likely occur even though the volume, averaged over the full hour, is less than the capacity. Peak hour factor is defined as the ratio of total hourly volume to the maximum rate of flow within the hour. If there was no variability in flow rate during the hour, the peak hour factor would be 1,00. Figure 4. Variations of traffic volume (Source: Georgia Department of Transportation).

Speed and travel time The third basic measurement of traffic is that of average, or mean, speed. Speed of travel is simple and widely used measure of the quality of traffic flow. Basically speed is the total traveled distance divided by the time of travel. Travel time is its reciprocal value. In the time-distance diagram the speed of a vehicle at any time is the slope of the line. In the case of the uniform vehicular stream described above, all vehicles were assumed to operate at the same speed v. The problem of when, where and how to take speed measurements that are representative of the traffic stream is not trivial. For example, the speed of successive vehicles may be taken at a single point of the roadway over a long period of time. These speeds are also known as spot speeds. Alternatively, the speeds of all the vehicles occupying a length of highway may be taken at the same instant. In general we distinguish between three basic classes of speed of travel:
6

a) spot speed b) overall speed c) running speed All speeds vary with time, location, environmental and traffic conditions. Speeds vary also with the quality of traffic services, being generally higher along motorways and expressways as well as during times when traffic congestions are not a factor. For example during the petroleum crises in 1970s and 1980s the average speed decreased. Today the most important factor impacting speed is traffic density and problem of traffic congestions. At a given time and location, speeds are widely dispersed and can generally be represented by a normal probability distribution. The range of speeds decreases with increase in traffic volume. Spot speed is the instantaneous speed of a vehicle as it passes a specified point along a road. Spot speeds may be determined by manually measuring (with use of electronic or electromechanical devices like pneumatic tube detectors or radars) the time required for a vehicle to traverse a relatively short specified distance. Two common ways of computing the average, or mean, speed. The average of a series of measures of spot speeds can be expressed as time-mean speed or space mean-speed. The time mean speed Ut is the arithmetic mean of spot speeds of all vehicles passing a point during a specified interval of time

1 N U t = U i N 1

or

ut

n i =1

ui

(5)

where

Ui (ui) N (n)

= =

observed speed of i-th vehicle number of vehicles observed

The space mean speed is calculated on the basis of the average travel time it takes N vehicles to traverse a length of roadway D. The i-th vehicle traveling at speed Ui will take

ti =

D Ui

(6)

seconds to cover the distance D. Space mean speed Us is then defined as (harmonic mean)

Us =

1 1 1 U N 1 i
N

or

us =

nd

t i =1 i

(7)

where

ti N (n) D (d)

= = =

observed time fore the ith vehicle to travel distance d number of vehicles observed length of roadway section

In other words space-mean speed is the average of vehicle speeds weighted according to how long they remain on the section of road. Space-mean speed and time-mean speed are not equal. For general usage, no distinction is normally made between both speeds, for theoretical and research purposes WARDROP has shown in his calculations that

ut = u s +
where s2 =

s2
us

(8)

variance of the space distribution of speeds

7

Overall speed and running speed These speeds are measured over a relatively long section of a road between an origin and a destination. These measures are used in travel time studies to compare the quality of service between alternative routes. Overall speed is defined as the total distance traveled divided by the total time required, including traffic delays. Running speed is defined as the total distance traveled divided by the running time. The running time is the time the vehicle is in motion. Both speeds are normally measured by means of a test vehicle that is driven over the test section of a roadway at the average speed of the traffic stream. A passenger uses a stopwatch to record time of travel to various previously chosen points along the course.

2. Time-distance diagrams of flow

The vehicular variables (spacing, headway, vehicle speed) and stream variables (flow, concentration, mean speed) just described can be clearly illustrated via a time-distance diagram of the trajectories of the vehicles constituting a traffic stream. Following figure is such a diagram for the simple case of uniformly operated vehicles represented as particles. Since in this case the speed of the vehicles is constant, the timedistance plot for each vehicle is simply a straight line, the slope of which, dx/dt, equals the speed, v. A point on a plot represents the location of the subject vehicle at the corresponding instant of time. A horizontal line (line AA, for example) intersects a number of time-distance lines and the (time) difference between pairs of vehicles along the horizontal line is the headway between those vehicles. Also, this horizontal line represents a stationary observer whose location does not change with time. The number of vehicles that the observer would be able to count over a period of observation T is equal to the number of times the horizontal line AA intersects a vehicle time-distance line: The higher the number of vehicles counted during time T, the higher the stream flow.

Figure 4
8

A vertical line (BB) represents the conditions prevailing at a given instant. The difference between subsequent vehicles is the spacing between vehicles. Also, line BB represents an aerial photograph of the stream at that instant: The number of time-distance lines that are intersected by the line BB corresponds to the number of vehicles that would appear on a photograph of the roadway segment shown. The smaller the number of such vehicles, the lower the stream concentration. In general the time-distance diagram is a graph that describes the relationship between the location of vehicles in a traffic stream and time as the vehicles progress along the highway.

3. Fundamental equation of vehicular stream

If two vehicles are traveling at a spacing s and speed u, the headway between them is simply h=s/u. Substituting s and h in this relationship leads to the fundamental equation describing a traffic stream:

q = k us
where q k = = = average volume or flow (vehicles per hour) average density or concentration (vehicle per kilometer) space-mean speed (km.h-1)

(9)

us

This equation is fundamental for describing the speed-volume-density relationship. The product of higher speed times a lower concentration gives the resulting flow. Note that the units balance to vehicles per hour on both sides of this equation, which represents a three-dimensional relationship between the basic vehicular stream variables flow, mean speed and concentration. It is if the utmost importance to realize that the three variables vary simultaneously. Consequently, it would be generally incorrect to attempt to compute the value of one of the three variables by varying another while holding the third constant. To gain a clearer understanding of this phenomenon, consider the two-dimensional projection of the equation on the u-k, u-q and q-k planes, first for the simple case of uniform flow and then for the more complex case of highway traffic. The theory has been postulated with respect to the shape of the curve depicting the relationship between concentration and flow. 1. When the density (concentration) on the highway is zero, the flow is also zero as there are no vehicles on the highway. 2. As the density increases, the flow also increases. 3. However, when the density reaches its maximum, generally referred to as the jam density (kj), the flow must be zero, as vehicles will tend to line up and to end. 4. It follows, therefore, that as density increases from zero, the flow will also initially increases from zero to a maximum value. Further continuous increase in density, however, will result in continuous reduction of the flow, which will eventually be zero when the density is the jam density. Although a number of theoretical and analytical speed-density relationships have been published, the exact shape of the k u s curve has not been conclusively established. A model proposed by Greenshields assumed a linear relationship between speed and density. With that assumption, a parabolic volume-density model results. The three main curves are shown in following figure.

Figure 5. Flow, density speed relationships

Let us consider three key points on the curves point A represents a low-density, high-speed situation, low volumes exist because few vehicles are on the road. Point B is the point of maximum flow. For this condition, intermediate levels of speed and density occur. Point C represents the worst possible type of flow situation, with maximum density occurring and speeds approximating zero. Empirical research has indicated that speed decreases exponentially with increases in density.

4. Uniform flow
Substituting equation (2) into equation (1), solving for k in terms of u, and adjusting the units of concentration to vehicles per km leads to:

10

k = f (u ) =

1 u u2 u + + NL + x0 2d f 2d l
2

(10)

Excepting the hypothetical case, the relationship between speed and concentration is seen to be monotonically decreasing as should be expected: the higher the speed, the longer the required spacing and, consequently, the lower the concentration. The conditions around very low concentration and very high speed are referred to as the free flow conditions and the maximum speed at zero concentration is known as the free-flow speed uf. The equation (10) also shows speed to approach infinity asymptotically as concentration approaches zero, for all practical purposes there exists a maximum speed, which depends on the technological characteristics of the system. The equation (10) could be modified for q:

q=

u u u2 u + + NL + x0 2d f 2d l
2

(11)

Each curve indicates zero flow at zero speed, meaning that, since no vehicle is moving, zero vehicles per unit time flow by a point on the facility. At the high-speed end, the flow exhibits a decline because of the increasingly longer spacing requirements for safe operation. The maximum flow (qmax) shown on each curve is the capacity of the roadway or guideway at the specified safety regime. The units of capacity are the same as the units of flow, that is, vehicles per unit time and not simply vehicles. Capacity occurs at an intermediate speed um and not at maximum (i.e. free) flow: Up to um, increasing speed corresponds to increasing flow; beyond um, increasing speed is associated with decreasing flow. Hence, in this range, a trade-off exists between speed and flow: Higher speeds can be attained only by sacrificing the throughput capability of the highway or guideway . Finally, the relationship between flow and concentration can be examined by solving equitation (8) for u in terms of k and multiplying both sides by k to obtain:

q = k u (k )

(12)

Next figure shows a typical plot of this relationship for a chosen safety regime and, for the sake of discussion, the hypothetical train as well. The free-flow end of this figure (i.e. low flow and low concentration) corresponds to the high-speed end of previous figure. At the other end of the diagram, concentration attains its maximum value, flow is zero, and speed is also zero. In other words, the roadway or guideway is occupied by as many vehicles as it can hold, but no vehicle is moving. Hence, no flow is developed. These conditions correspond to a traffic jam, where maximum "packing" of stationary vehicles occurs. The value of concentration at that end is denoted by the jam concentration kj. Again, maximum flow or capacity occurs at intermediate values of speed um and concentration km. The horizontal line AA intersects the q-k curve at two points. Although the flow is the same at these points, the concentration is different. Also, the speeds corresponding to these two points are different (see previous figures). Point 1 represents conditions that are closer to free-flow, whereas point 2 represents conditions that are more congested. If a straight line is drawn from the origin to a point on the q-k curve, the slope of this line is simply equal to q/k, which is equal to the mean speed u. Therefore it is possible to specify the numerical values of the three basic stream variables (q, k, and u) by using only one of the three diagrams. It is customary (especially in highway traffic analysis) to use the q-k diagram for this purpose.

11

Figure 6. Speed-concentration curves for four of the five safety regimes

Figure 7

12

5. The case of highway traffic flow

Figure 7 shows a q-k curve for some safety regime, say b. As discussed earlier, operating conditions that do not lie exactly on the curve are quite possible. For example, all points associated with safety regime d lie above the regime b curve shown. Thus, it is the desired level of safety that fixes the q. k. and U points on a particular curve and not the physical capabilities of the system. Consider, for example, the limiting hypothetical case of a continuous train that operates on a closed loop at a constant speed U and that is never required to decelerate. In this case, considerations of safe stopping are not relevant. Theoretically, the concentration of the continuous train can be kept at jammed conditions on the track. Point C represents a stationary train at jam concentration, zero flow, and zero speed. If the train is operated at some constant speed u, its concentration remains at jam concentration, but the flow becomes finite (see point D). If the operating speed is higher, as exemplified by the slope of line BE, the conditions associated with point E result. Thus the vertical line at jam concentration represents this hypothetical case. Clearly, it is physically impossible to operate the system at any of the conditions shown to the right of this line.

Figure 8 Now consider line AB. The slope of this line represents a high speed. This line would represent situations where the same high speed can be maintained at all values of concentration, a situation that is approached in the case of car racing: Irrespective of concentration, the speeds that race-track drivers sustain are very high. Of course, in the case of car racing the predominant consideration is not safety but speed. Points below this race-track line are also attainable. Thus, for a given transportation technology, triangle ABC encloses the area on the q-k plane within which it is physically possible to operate the system. Within this triangle, the conditions described by points below as well as above the q-k curve shown are physically possible. Even

13

points lying on the k-axis may be given a physical interpretation. For example, point H may represent a sparsely occupied parking lane at concentration below jammed conditions, zero speed, and zero flow. In this case drivers make their own decisions regarding trade-off between safety and speed (the typical q-k diagram). What gives rise to the q-k curve in a typical travel situation is the trade-off between a desire to get to ones destination quickly as possible (maximize speed) on one hand and getting there safety (reflected by the preferred safety regime) on the other. Some drivers keep close to the car in front of them and try to increase their speeds when possible, whereas others keep unusually long spacing by stressing safety more than speed.

Figure 9 In addition, highway vehicles are not identical but exhibit a great amount of variability in size and technological attributes. The upshot of all these individual differences is a statistical clustering of points representing the stream conditions around a q-k curve. The next figure illustrates the general form of the u-k, u-q and q-k diagrams corresponding to highway flow. The u-k relationship is monotonically decreasing, reflecting the rule that drivers follow on the average as they follow one another. The rule of the road suggested by many city traffic ordinances of keeping a distance of one car length for each 10 km/h increment of speed is but one such car-following rule. The q-u and q-k relationships are backward bending as before, with maximum flow occurring at an intermediate speed um and concentration km. Typically given the u-k relationship, it is possible to estimate the other two relationships by following the procedure that was applied earlier for the case of uniformly scheduled transit.

14

Figure 10

6. Road capacity
Even with ideal roadway conditions, traffic volume tends to reach a maximum point at a relatively low speed. This phenomenon, which is puzzling to the casual observer, results from the fact that spacing allowed by the average driver when trailing another vehicle increases nonlinearly with increases in speed. Increasing speed tends to increase volume, but that effect is offset by concomitant decreases in density. Another factor limiting the number of vehicles passing a point is interference between vehicles in the traffic stream. This effect is especially noticeable along two-lane roads. With a low traffic volume, the vehicle operator has wide latitude in selecting the speed at which the driver wishes to travel. As traffic volume increases, the speed of each vehicle is influenced in a large measure by the speed of the slower vehicles. As traffic density increases, a point is finally reached where all vehicles are traveling at the speed of the slowest vehicles. This condition indicates that the ultimate capacity has been reached. The capacity of a road is therefore measured by its ability to accommodate traffic and is usually expressed as the number of vehicles that can pass a given point in a certain period of time at a given speed. Roadways, of course, are not ideal, and in road capacity estimates allowances must be made for prevailing roadway and traffic conditions that inhibit the ability of a road to accommodate traffic. The capacity of a given section of roadway, either in one direction or in both directions for a two-lane or three-lane roadway, may be defined as the maximum hourly rate at which 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. Although the maximum number of vehicles that can be accommodated remains fixed under similar roadway and traffic conditions, there is a range of lesser volumes that can be handled under differing operating
15

conditions. Operation at capacity provides the maximum volume, but as both volume and congestion decrease there is an improvement in the level of service. Level of service is a qualitative measure that describes operational conditions within a traffic stream and their perception by drivers and/or passenger. The full range of driving conditions from best to worst, in that order is mostly divided into several levels of service. These levels of service qualitatively measure the effect of such factors as travel time, speed, cost and freedom to maneuver, which, in combination with other factors, determine the type of service that any given facility provides to the user under the stated conditions. With each level of service, a service flow rate is defined. It is the maximum volume that can pass over a given section of lane or roadway while operating conditions are maintained at the specified level of service. The levels of service have traditionally been based on relationships involving operating and volume-tocapacity ratio. As the operating speed on a facility increases, the higher levels of service are attained, on the other hand, as the volume-to-capacity ratio increases, the facility tends to operate at a lower level of service.
Measures of effectiveness for level of service definition

Type of facility Uninterrupted flow facilities Freeways - basic freeway segments - weaving areas - ramp junctions Multilane highways Two-lane roads Interrupted flow facilities Signalized intersections Unsignalized intersections Arterials Transit Pedestrians

Measure of effectiveness

Units

Density Average travel speed Flow rate Density Percent time delay Average travel speed

Passenger cars per km per lane Km.h-1 Passenger cars per hour Passenger cars per km per lane Percent Km.h-1

Average individual stopped Seconds per vehicle delay Reserve capacity Average travel speed Load factor Space Passenger cars per hour Km.h-1 Person per seat Square meter per pedestrian

Factors affecting capacity, service flow rate and level of service Since prevailing conditions are seldom ideal, computation of capacity, service flow rate, or level of service must be adjusted to account for departures from the ideal. Prevailing conditions may be grouped into three categories: Roadway factors. a) The type of facility and its environment. b) Lane width. c) Shoulder widths and/or lateral clearances. d) Design speed. e) Horizontal and vertical alignments.
16

Traffic conditions refer to the types of vehicles using the facility and how the traffic flow is distributed by lane use and direction. It is well known that larger and heavier vehicles have an adverse effect on traffic flow in a number of ways. In addition to the distribution of vehicle types, the effects of two other traffic characteristic on capacity, service flow rates and level of service must be considered. Directional distribution has a significant impact on the directional split of two-lane rural highways. Capacity generally declines as the directional split becomes unbalanced. For multilane motorways and expressways, capacity analysis procedures focus on a single direction, that of the peak rate of flow. Lane distribution must also be considered for multilane facilities. Control conditions. For interrupted flow facilities, the control of the time available for movement of each traffic flow is of utmost importance in determining capacity, service flow rates and level of service. Of greatest importance is the control of such facilities by the traffic signal. Signalized operations are affected by the type of signal in use, the phasing, the allocation of green time and the length of cycle.

17