Chapter 6

Highway Capacity an d
Level-of-Service Analysi s

6.1 INTRODUCTIO N
The underlying objective of traffic analysis is to quantify a roadway's performanc e
with regard to specified traffic volumes . This performance can be measured in term s
of travel delay (as the roadway becomes increasingly congested) as well as othe r
factors . The comparative performance of various roadway segments (which i s
determined from an analysis of traffic) is important because it can be used as a basi s
to allocate limited roadway construction and improvement funds . The purpose of thi s
chapter is to apply the elements of uninterrupted traffic flow theory covered i n
Chapter 5 to the practical field analysis of traffic flow and capacity on freeways ,
multilane highways, and two-lane highways .
The main challenge of such a process is to adapt the theoretical formulations t o
the wide range of conditions that occur in the field . These diverse field condition s
must be taken into account in a traffic analysis methodology, yet the methodolog y
must remain theoretically consistent . For example, in Chapter 5, capacity (geap) is
simply defined as the highest traffic flow rate that the roadway is capable o f
supporting . For applied traffic analysis, a consistent and reasonably precise metho d
of determining capacity must be developed within this definition . Because it can
readily be shown that the capacity of a roadway segment is a function of factors suc h
as roadway type (freeway, multilane highway, or two-lane highway), free-flo w
speed, number of lanes, and widths of lanes and shoulders, the method of capacit y
determination clearly must account for a wide variety of physical and operationa l
roadway characteristics .
Additionally, recall that Chapter 5 defines traffic flow on the basis of units o f
vehicles per hour . Two practical issues arise concerning this unit of measure . First, in
many cases vehicular traffic consists of a variety of vehicle types with substantiall y
different performance characteristics . These performance differentials are likely to b e
magnified by changing roadway geometries, such as upgrades or downgrades, whic h
have a differential effect on the acceleration and deceleration capabilities of th e
various types of vehicles ; for example, grades have a greater impact on th e
performance of large trucks than automobiles . As a result, traffic must be defined no t
only in terms of vehicles per unit time but also in terms of vehicle composition ,
because it is clear that a 1500-veh/h traffic flow consisting of 100% automobiles wil l

175
176 Chapter 6 Highway Capacity and Level-of-Service Analysi s

differ significantly with regard to operating speed and traffic density from a 1500 -
veh/h traffic flow that consists of 50% automobiles and 50% heavy trucks .
The other flow-related concern is the temporal distribution of traffic . In practice,
the analysis of roadway traffic usually focuses on the most critical condition, whic h
is the most congested hour within a 24-hour daily period (the temporal distribution o f
traffic will be discussed in more detail in Section 6 .7) . However, within this mos t
congested peak hour, traffic flow is likely to be nonuniform . It is therefore necessary
to arrive at some method of defining and measuring the nonuniformity of flow withi n
the peak hour .
To summarize, the objective of applied traffic analysis is to provide a practica l
method of quantifying the degree of traffic congestion and to relate this to the overal l
traffic-related performance of the roadway . The following sections of this chapte r
discuss and demonstrate accepted standards for applied traffic analysis for the thre e
major types of uninterrupted-flow roadways : freeways, multilane highways, and two -
lane highways (one lane in each direction) .

6 .2 LEVEL-OF-SERVICE CONCEP T
The Highway Capacity Manual (RCM), produced by the Transportation Researc h
Board [2000], is a synthesis of the state of the art in methodologies for quantifyin g
traffic operational performance and capacity utilization (congestion level) for a
variety of transportation facilities . One of the foundations of the HCM is the concep t
of level of service (LOS) . The level of service represents a qualitative ranking of the
traffic operational conditions experienced by users of a facility under specifie d
roadway, traffic, and traffic control (if present) conditions . Current practic e
designates six levels of service ranging from A to F, with level of service A
representing the best operating conditions and level of service F the worst .
A number of operational performance measures, such as speed, flow, an d
density, can be measured or calculated for any transportation facility . To apply the
level-of-service concept to traffic analysis, it is necessary to select a performanc e
measure that is representative of how motorists actually perceive the quality o f
service they are receiving on a facility . Motorists tend to evaluate their receive d
quality of service in terms of factors such as speed and travel time, freedom t o
maneuver, traffic interruptions, and comfort and convenience . Thus, it is important t o
select a measure that encompasses some or all of these factors . The performanc e
measure that is selected for level-of-service (LOS) analysis for a particula r
transportation facility is referred to as the service measure .
The HCM [Transportation Research Board 2000] defines the LOS categories fo r
freeways and multilane highways as follows :

Level of service A . LOS A represents free-flow conditions (traffic operating a t

free-flow speeds, as defined in Chapter 5) . Individual users are virtually
unaffected by the presence of others in the traffic stream . Freedom to selec t
speeds and to maneuver within the traffic stream is extremely high . The genera l
level of comfort and convenience provided to drivers is excellent .

Level of service B . LOS B also allows speeds at or near free-flow speeds, but th e
presence of other users in the traffic stream begins to be noticeable . Freedom to
 6 .2 Level-of-Service Concept 17 7

select speeds is relatively unaffected, but there is a slight decline in the freedo m
to maneuver within the traffic stream relative to LOS A .

Level of service C . LOS C has speeds at or near free-flow speeds, but th e

freedom to maneuver is noticeably restricted (lane changes require carefu l
attention on the part of drivers) . The general level of comfort and convenienc e
declines significantly at this level . Disruptions in the traffic stream, such as a n
incident (for example, vehicular accident or disablement), can result i n
significant queue formation and vehicular delay . In contrast, the effects o f
incidents at LOS A or LOS B are minimal, with only minor delay in th e
immediate vicinity of the event .

Level of service D . LOS D represents the conditions where speeds begin t o

decline slightly with increasing flow . The freedom to maneuver becomes more
restricted, and drivers experience reductions in physical and psychological
comfort . Incidents can generate lengthy queues because the higher densit y
associated with this LOS provides little space to absorb disruptions in the traffi c
flow .

Level of service E . LOS E represents operating conditions at or near th e

roadway's capacity . Even minor disruptions to the traffic stream, such as vehicle s
entering from a ramp or vehicles changing lanes, can cause delays as othe r
vehicles give way to allow such maneuvers . In general, maneuverability i s
extremely limited, and drivers experience considerable physical and
psychological discomfort .

Level of service F . LOS F describes a breakdown in vehicular flow . Queue s

form quickly behind points in the roadway where the arrival flow rat e
temporarily exceeds the departure rate, as determined by the roadway's capacit y
(see Chapter 5) . Such points occur at incidents and on- and off-ramps, where
incoming traffic results in capacity being exceeded . Vehicles typically operate a t
low speeds under these conditions and are often required to come to a complete
stop, usually in a cyclic fashion . The cyclic formation and dissipation of queues
is a key characterization of LOS F .

A visual perspective of the level-of-service definitions for freeways is provide d

in Fig . 6 .1 . In dealing with level of service it is important to remember that when th e
traffic volume is at or near the roadway capacity (which will be shown as a functio n
of the prevailing traffic and physical characteristics of the roadway), the roadway i s
operating at LOS E . This, however, is not a desirable condition because under LOS E
conditions there is considerable driver discomfort, which could increase th e
likelihood of vehicular crashes and overall delay . In roadway design, the possibilit y
of degradation of level of service to LOS E should be avoided, although this is no t
always possible due to financial and environmental constraints that may limit th e
design speed, number of lanes, and other factors affecting roadway capacity .


178 Chapter 6 Highway Capacity and Level-of-Service Analysi s

LOS A LOS B

LOS C LOS D

LOSE LOS F
Figure 6 .1 Illustration of freeway level of service (A to F) .
Reproduced with permission of the Transportation Research Board, Highway Capacity Manual 2000, Copyright,
National Academy of Sciences, Washington, D .C. Illustrations 13-5 to 13-10, pp . 13-8 to 13-10 .
 6 .3 Level-of-Service Determination 17 9

6.3 LEVEL-OF-SERVICE DETERMINATIO N

There are several steps in a basic level-of-service determination for an uninterrupted -
flow facility . The remainder of this section describes the general details of each step ,
as applicable to uninterrupted-flow facility analyses . Facility-specific details of thes e
steps are described in the sections that follow .

6.3.1 Base Conditions and Capacity

The determination of a roadway's level of service begins with the specification o f
base roadway conditions . Recall that in the introduction to this chapter, the effects o f
vehicle performance and roadway design characteristics on traffic flow were
discussed qualitatively . In practice, the effects of such factors on traffic flow are
measured quantitatively, relative to the base traffic and roadway design conditions .
For uninterrupted-flow roadways, base conditions can be categorized as thos e
relating to roadway conditions, such as lane widths, lateral clearances, acces s
frequency, and terrain ; and traffic stream conditions such as the effects of heav y
vehicles (large trucks, buses, and RVs) and driver population characteristics . Base
conditions are defined as those conditions that represent unrestrictive geometric an d
traffic conditions. Additionally, base conditions are assumed to consist of favorabl e
environmental conditions (such as dry roadways) .
The capacity of a particular roadway segment will be greatest when all roadway
and traffic conditions meet or exceed their base values . Empirical studies hav e
identified the values of these base conditions for which the capacity of a roadway
segment is maximized . Values in excess of the base conditions will not increase th e
capacity of the roadway, but values more restrictive than the base conditions wil l
result in a lower capacity . For example, studies have identified a base lane width of
12 ft (3 .6 m). That is, lane widths in excess of 12 ft (3 .6 m) will not result i n
increased capacity ; however, lane widths less than 12 ft (3 .6 m) will result in a
reduction in capacity . Capacity values for base conditions have been determined fo r
all uninterrupted-flow facility types from field studies . It should be noted that fo r
purposes of level-of-service analysis, capacity is defined not as the absolut e
maximum flow rate ever observed for a particular facility type, but rather as th e
maximum flow rate that can be reasonably expected on a recurring basis .
Because all base conditions for a particular roadway type are seldom realized i n
practice, methods for converting the measured flow rate into an equivalent analysi s
flow rate in terms of passenger cars for the given traffic conditions and estimating th e
actual free-flow speed for the given roadway conditions are needed . The following
sections describe the procedures for arriving at flow and speed values for given
roadway and traffic conditions .

6.3.2 Determine Free-Flow Spee d

Free-flow speed (FFS) is a term that was introduced in Chapter 5 as the speed o f
traffic as the traffic density approaches zero . In practice, FFS is governed by roadwa y
design characteristics (horizontal and vertical curves, lane and shoulder widths, an d
median design), the frequency of access points, the complexity of the drivin g
environment (possible distractions from roadway signs and the like), and poste d
speed limits.
180 Chapter 6 Highway Capacity and Level-of-Service Analysi s

The free-flow speed must be determined given the characteristics of the roadway
segment. FFS is the mean speed of traffic as measured when flow rates are low t o
moderate (specific values are given under the individual sections for each roadway
type) . Ideally, FFS should be measured directly in the field at the site of interest .
However, if this is not possible or feasible, an alternative method can be employed t o
arrive at an estimate of FFS under the prevailing conditions . This method make s
adjustments to a base FFS (BFFS) depending on the physical characteristics of th e
roadway segment, such as lane width, shoulder width, and access frequency . Thi s
method has the same basic structure for the various roadway types, but contain s
adjustment factors and values appropriate for each roadway type .

6 .3 .3 Determine Analysis Flow Rat e

One of the fundamental inputs to a traffic analysis is the actual traffic volume on th e
roadway, in vehicles per hour, which is given the symbol V . Generally, the highes t
volume in a 24-hour period (the peak-hour volume) is used for V in traffic analysi s
computations . However, this hourly volume needs to be adjusted to reflect th e
temporal variation of traffic demand within the analysis hour, the impacts due t o
heavy vehicles, and, in the case of freeway and multilane roadways, th e
characteristics of the driving population . To account for these effects, the hourl y
volume is divided by adjustment factors to obtain an equivalent flow rate in terms o f
passenger cars per hour (pc/h) . Additionally, the flow rate is expressed on a per-lan e
basis (pc/h/ln) by dividing by the number of lanes in the analysis segment .

6 .3 .4 Calculate Service Measure(s) and Determine LO S

Once the previous steps have been completed, all that remains is to calculate th e
value of the service measure and then determine the LOS from the service measur e
value . For freeways and multilane highways, this is a relatively straightforward task .
However, for two-lane highways, there are actually two service measures, and th e
calculation of these and the subsequent LOS determination are more involved .

6 .4 BASIC FREEWAY SEGMENTS

A basic freeway segment is defined as a section of a divided roadway having two o r
more lanes in each direction, full access control, and traffic that is unaffected b y
merging or diverging movements near ramps . It is important to note that capacit y
analysis for divided roadways focuses on the traffic flow in one direction only . Thi s
is reasonable because the objective is to measure the highest level of congestion . Du e
to directional imbalance of traffic flows — for example, morning rush hours havin g
higher volumes going toward the central city and evening rush hours having highe r
volumes going away from the central city — consideration of traffic volumes in bot h
directions is likely to seriously understate the true level of traffic congestion .
Table 6 .1 provides the level-of-service criteria corresponding to traffic density ,
speed, volume-to-capacity ratio, and maximum service flow rate . A graphica l
representation of this table is provided in Fig . 6 .2 . The maximum service flow rate i s
simply the maximum flow rate, under base conditions, that can be sustained for a
given level of service . This value is related to speed and density as discussed i n
Chapter 5 . This speed-flow-density relationship is central to the analysis of basi c
freeway segments, as will be outlined in the remainder of this section .