Chapter 26 - Freeway and Highway Segments Supplemental - 700

HIGHWAY CAPACITY
MANUAL 7th Edition

A Guide for Multimodal Mobility Analysis
Transportation Research Board publications are available by ordering individual publications
directly from the TRB Business Office, through the Internet at www.TRB.org or
nationalacademies.org/trb, or by annual subscription through organizational or individual
affiliation with TRB. Affiliates and library subscribers are eligible for substantial discounts. For
further information, contact the Transportation Research Board Business Office, 500 Fifth
Street, NW, Washington, DC 20001 (telephone 202-334-3213; fax 202-334-2519; or e-mail
TRBsales@nas.edu).
Copyright 2022 by the National Academy of Sciences. All rights reserved.
Printed in the United States of America
Hardcover International Standard Book Number-13: 978-0-309-08766-7

Volume 1 International Standard Book Number-13: 978-0-309-27566-8
eBook International Standard Book Number-13: 978-0-309-27562-0
Digital Object Identifier: https://doi.org/10.17226/26432
Library of Congress Control Number: 2022930290
Suggested citation: National Academies of Sciences, Engineering, and Medicine. 2022. Highway
Capacity Manual 7th Edition: A Guide for Multimodal Mobility Analysis. Washington, DC: The
National Academies Press. https://doi.org/10.17226/26432.
The National Academy of Sciences was established in 1863 by an Act of Congress,
signed by President Lincoln, as a private, nongovernmental institution to advise the
nation on issues related to science and technology. Members are elected by their peers
for outstanding contributions to research. Dr. Marcia McNutt is president.
The National Academy of Engineering was established in 1964 under the charter of the
National Academy of Sciences to bring the practices of engineering to advising the
nation. Members are elected by their peers for extraordinary contributions to
engineering. Dr. John L. Anderson is president.
The National Academy of Medicine (formerly the Institute of Medicine) was established
in 1970 under the charter of the National Academy of Sciences to advise the nation on
medical and health issues. Members are elected by their peers for distinguished
contributions to medicine and health. Dr. Victor J. Dzau is president.
The three Academies work together as the National Academies of Sciences,

Engineering, and Medicine to provide independent, objective analysis and advice to the
nation and conduct other activities to solve complex problems and inform public policy
decisions. The Academies also encourage education and research, recognize
outstanding contributions to knowledge, and increase public understanding in matters
of science, engineering, and medicine.
Learn more about the National Academies of Sciences, Engineering, and Medicine at
www.nationalacademies.org.
The Transportation Research Board is one of seven major programs of the National
Academies of Sciences, Engineering, and Medicine. The mission of the Transportation
Research Board is to provide leadership in transportation improvements and innovation
through trusted, timely, impartial, and evidence-based information exchange,
research, and advice regarding all modes of transportation. The Board’s varied
activities annually engage about 8,000 engineers, scientists, and other transportation
researchers and practitioners from the public and private sectors and academia, all of
whom contribute their expertise in the public interest. The program is supported by
state departments of transportation, federal agencies including the component
administrations of the U.S. Department of Transportation, and other organizations and
individuals interested in the development of transportation.
Learn more about the Transportation Research Board at www.TRB.org.

Highway Capacity Manual: A Guide for Multimodal Mobility Analysis
CHAPTER 26
FREEWAY AND HIGHWAY SEGMENTS: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 26-1
2. STATE-SPECIFIC HEAVY-VEHICLE DEFAULT VALUES .......................... 26-2
3. TRUCK ANALYSIS USING THE MIXED-FLOW MODEL .......................... 26-4

Introduction .......................................................................................................... 26-4
Overview of the Methodology ........................................................................... 26-4
4. ADJUSTMENTS FOR DRIVER POPULATION EFFECTS ......................... 26-14
5. GUIDANCE FOR FREEWAY CAPACITY ESTIMATION .......................... 26-15

Freeway Capacity Definitions .......................................................................... 26-15
Capacity Measurement Locations ................................................................... 26-16
Capacity Estimation from Field Data .............................................................. 26-18
6. CONNECTED AND AUTOMATED VEHICLES .......................................... 26-22

Introduction ........................................................................................................ 26-22
Concepts ............................................................................................................. 26-22
Capacity Adjustment Factors ........................................................................... 26-27
Service Volume Tables ...................................................................................... 26-28
7. FREEWAY AND MULTILANE HIGHWAY EXAMPLE PROBLEMS ....... 26-30

Example Problem 1: Four-Lane Freeway LOS ............................................... 26-30
Example Problem 2: Number of Lanes Required for Target LOS ............... 26-33
Example Problem 3: Six-Lane Freeway LOS and Capacity ......................... 26-35
Example Problem 4: LOS on a Five-Lane Highway with a Two-Way
Left-Turn Lane ............................................................................................ 26-38
Example Problem 5: Mixed-Flow Freeway Operations ............................... 26-40
Example Problem 6: Severe Weather Effects on a Basic Freeway
Segment ....................................................................................................... 26-47
Example Problem 7: Basic Managed Lane Segment ..................................... 26-49
Chapter 26/Freeway and Highway Segments: Supplemental Contents

Version 7.0 Page 26-i
8. TWO-LANE HIGHWAY EXAMPLE PROBLEMS .........................................26-54

Example Problem 1: Level, Straight, Passing Constrained Segment ..........26-54
Example Problem 2: Passing Constrained Segment with Horizontal
Curves ..........................................................................................................26-60
Example Problem 3: Facility Analysis—Level Terrain .................................26-63
Example Problem 4: Facility Analysis – Mountain Road .............................26-85
Example Problem 5: Two-Lane Highway Bicycle LOS ................................26-98
9. REFERENCES .....................................................................................................26-100
APPENDIX A: TRUCK PERFORMANCE CURVES .......................................26-102
APPENDIX B: WORK ZONES ON TWO-LANE HIGHWAYS ....................26-107

Concepts ............................................................................................................26-107
Work Zone Capacity .......................................................................................26-107
Queuing and Delay Analysis .........................................................................26-114
Example Calculation .......................................................................................26-116
Reference ...........................................................................................................26-120
Contents Chapter 26/Freeway and Highway Segments: Supplemental

Page 26-ii Version 7.0
LIST OF EXHIBITS
Exhibit 26-1 State-Specific Default Values for Percentage of Heavy

Vehicles on Freeways .......................................................................................... 26-2
Exhibit 26-2 State-Specific Default Values for Percentage of Heavy
Vehicles on Multilane and Two-Lane Highways ............................................ 26-3
Exhibit 26-3 Overview of Operational Analysis Methodology for Mixed-
Flow Model........................................................................................................... 26-5
Exhibit 26-4 Speed–Flow Models for 70-mi/h Auto-Only Flow and a
Representative Mixed Flow................................................................................ 26-5
Exhibit 26-5 SUT Travel Time Versus Distance Curves for 70-mi/h FFS ............ 26-9
Exhibit 26-6 TT Travel Time Versus Distance Curves for 70-mi/h FFS ............... 26-9
Exhibit 26-7 δ Values for SUTs ................................................................................ 26-10
Exhibit 26-8 δ Values for TTs................................................................................... 26-10
Exhibit 26-9 Recommended CAF and SAF Adjustments for Driver
Population Impacts ........................................................................................... 26-14
Exhibit 26-10 Recommended Capacity Measurement Location for Merge
Bottlenecks .......................................................................................................... 26-17
Exhibit 26-11 Recommended Capacity Measurement Location for
Diverge Bottlenecks ........................................................................................... 26-17
Exhibit 26-12 Recommended Capacity Measurement Location for
Weaving Bottlenecks ......................................................................................... 26-17
Exhibit 26-13 Illustrative Example of the Capacity Estimation Procedure ....... 26-20
Exhibit 26-14 Capacity Estimation Using the 15% Acceptable Breakdown
Rate Method ....................................................................................................... 26-21
Exhibit 26-15 Capacity Adjustment Factors for CAVs for Basic Freeway
and Freeway Diverge Segments ...................................................................... 26-27
Exhibit 26-16 Capacity Adjustment Factors for CAVs for Freeway Merge
Segments ............................................................................................................. 26-28
Exhibit 26-17 Capacity Adjustment Factors for CAVs for Freeway
Weaving Segments ............................................................................................ 26-28
Exhibit 26-18 Daily Maximum Service Volumes for Basic Freeway
Segments with CAV Presence (2-way veh/day/ln) ....................................... 26-29
Exhibit 26-19 Hourly Maximum Service Volumes for Basic Freeway
Segments with CAV Presence (veh/h/ln) ....................................................... 26-29
Exhibit 26-20 List of Freeway and Multilane Highway Example Problems ..... 26-30
Exhibit 26-21 Example Problem 1: Graphical Solution ......................................... 26-32
Exhibit 26-22 List of Two-Lane Highway Example Problems ............................ 26-54
Exhibit 26-23 Example Problem 2: Horizontal Curve Inputs ............................... 26-60
Exhibit 26-24 Example Problem 2: Horizontal Curve Average Speed Results ... 26-62
Exhibit 26-25 Example Problem 2: Average Speeds by Subsegment ................... 26-62
Exhibit 26-26 Example Problem 3: Input Data ....................................................... 26-63
Chapter 26/Freeway and Highway Segments: Supplemental Contents

Version 7.0 Page 26-iii
Exhibit 26-27 Example Problem 3: LOS Results ....................................................26-85

Exhibit 26-28 Example Problem 4: Facility Diagram ............................................26-86
Exhibit 26-29 Example Problem 4: Facility Volume and Speed Data ................26-86
Exhibit 26-30 Example Problem 4: Facility Grade and Horizontal Curve
Data ......................................................................................................................26-86
Exhibit 26-31 Example Problem 4: Vertical Alignment Classifications by
Segment ...............................................................................................................26-88
Exhibit 26-32 Example Problem 4: Free-Flow Speed Results ..............................26-89
Exhibit 26-33 Example Problem 4: Unadjusted Average Speed Results ...........26-91
Exhibit 26-34 Example Problem 4: Adjusted Average Speed Results ................26-92
Exhibit 26-35 Example Problem 4: Percent Follower and Unadjusted
Follower Density Results ..................................................................................26-95
Exhibit 26-36 Example Problem 4: LOS Results ....................................................26-97
Exhibit 26-A1 SUT Travel Time Versus Distance Curves for 50-mi/h FFS .....26-102
Exhibit 26-A6 TT Travel Time Versus Distance Curves for 50-mi/h FFS ........26-104
Exhibit 26-A10 TT Travel Time Versus Distance Curves for 75-mi/h FFS ......26-106
Exhibit 26-B1 Traffic Control for a Two-Lane Highway Work Zone
Involving a Lane Closure ................................................................................26-108
Exhibit 26-B2 Two-Lane Highway Work Zone Grade Adjustment Factor
(fg) for Level Terrain, Rolling Terrain, and Specific Downgrades .............26-109
Exhibit 26-B3 Two-Lane Highway Work Zone Grade Adjustment Factor
(fg) for Specific Upgrades ................................................................................26-110
Exhibit 26-B4 Two-Lane Highway Work Zone Passenger Car Equivalents
for Trucks (ET) and RVs (ER) for Level Terrain, Rolling Terrain, and
Specific Downgrades .......................................................................................26-111
for Trucks (ET) on Specific Upgrades ............................................................26-112
for RVs (ER) on Specific Upgrades .................................................................26-112
Exhibit 26-B7 Directional Queueing Diagram for a Two-Lane Highway
Lane-Closure Work Zone ...............................................................................26-115
Exhibit 26-B8 Example Calculation: Work Zone Roadway Parameters ..........26-117
Exhibit 26-B9 Example Calculation: Work Zone Traffic Parameters ...............26-117
Contents Chapter 26/Freeway and Highway Segments: Supplemental

Page 26-iv Version 7.0
1. INTRODUCTION
Chapter 26 is the supplemental chapter for Chapter 12, Basic Freeway and VOLUME 4: APPLICATIONS
GUIDE
Multilane Highway Segments, and Chapter 15, Two-Lane Highways, which are 25. Freeway Facilities:
found in Volume 2 of the Highway Capacity Manual (HCM). Supplemental
26. Freeway and Highway
Section 2 provides state-specific heavy-vehicle default values that can be Segments:
Supplemental
applied to freeway, multilane highway, and two-lane highway analysis. 27. Freeway Weaving:
Supplemental
Section 3 presents a supplemental procedure for basic freeway segments that 28. Freeway Merges and
can be used to assess their operating performance under mixed-flow conditions Diverges: Supplemental
29. Urban Street Facilities:
when significant truck presence, a prolonged single upgrade, or both exist. Supplemental
Appendix A provides travel time versus distance curves for single-unit trucks 30. Urban Street Segments:
Supplemental
(SUTs) and tractor-trailers (TTs) for a range of free-flow speeds (FFS) for use with 31. Signalized Intersections:
this procedure. Chapter 25, Freeway Facilities: Supplemental, presents an Supplemental
32. STOP-Controlled
extension of this method for composite grades on freeway facilities. Intersections:
Supplemental
Section 4 provides suggested capacity and FFS adjustments to account for the 33. Roundabouts:
effects of different proportions of motorists on a freeway or multilane highway Supplemental
34. Interchange Ramp
who are not regular users of the facility. Terminals: Supplemental
35. Pedestrians and Bicycles:
Section 5 presents freeway capacity definitions, guidance on locating sensors
Supplemental
for use in measuring freeway capacity, and guidance on estimating capacity from 36. Concepts: Supplemental
37. ATDM: Supplemental
the collected sensor data.
38. Network Analysis
Section 6 provides guidance on incorporating the potential freeway capacity
benefits of connected and automated vehicles into an HCM analysis.
Section 7 provides seven example problems demonstrating the basic freeway
and multilane highway segment procedure presented in Chapter 12.
Section 8 provides five example problems demonstrating the motorized
vehicle and bicycle methodologies for two-lane highways presented in Chapter 15.
Appendix B describes a methodology for calculating capacity and related
performance measures for work zones along two-lane highways that involve the
closure of a single lane.
Chapter 26/Freeway and Highway Segments: Supplemental Introduction

Version 7.0 Page 26-1
2. STATE-SPECIFIC HEAVY-VEHICLE DEFAULT VALUES
Research into the percentage of heavy vehicles on uninterrupted-flow

facilities (1) found such a wide range of average values from state to state that
not even regional default values could be developed. Exhibit 26-1 presents
default values for the percentage of heavy vehicles on freeways by state and area
population based on data from the 2004 Highway Performance Monitoring
System. Exhibit 26-2 presents similar default values for multilane and two-lane
highways. In cases in which states or local jurisdictions have developed their
own default values, those values should be used in lieu of the values presented
here. Analysts may also wish to develop their own default values based on local
or more recent data.
Exhibit 26-1 Small Medium Large Small Medium Large

State-Specific Default Values State Rural Urban Urban Urban State Rural Urban Urban Urban
for Percentage of Heavy AL* 14a 7 7 7a MT 22c 16c 12c NA
Vehicles on Freeways AK 4 5b 5 3b NC* 19b 12b 12 10a
AR 30 24 13 14 ND 21c 22c 10c NA
AZ 21 19 18 11 NE 36 37 11 8
CA 16 10 7 6 NH 15b 12b 6b 7b
CO 12 10 8 7 NJ 8 6 6 9
CT 13 6 6 5 NM 26 12 21 12
DC NA NA NA 4b NV 34b 26 18b 11b
DE ⎯ ⎯ 9b 8b NY 18 11 11 7
FL* 11 7 12 6 OH 24 13 10 8
GA* 19b 7b 12 8b OK 28 27 12 10
HI 5 19b 2 3 OR 26 19 10 7
IA 20c 24c 11c 10c PA 16 13 9 8
ID 29c 28b 12b 7b PR* 6 7b 7 4b
IL 21 23 16 9 RI 3 ⎯ NA 4
IN 26 25 23 14 SC* 19b 7b 7 8b
KS 21c 17c 8c 9b SD 20c 14c 9c NA
KY* 20a 16 12 10a TN* 19 12 12 8
LA* 12c 7b 12 10c TX 16 28c 8 5
MA 7a 5 4a 4 UT 34c ⎯ 18 13
MD 18 14 17 8 VA* 9 7 7 4
ME 5 5 5 NA VT 15 12 6 NA
MI 18 12 13 8 WA 11 10 7 6
MN 11 10 6 4 WI 6 6 6 6
MO 29b 23b 13b 10b WV 16b 13b 9b NA
MS* 9b 7b 7 6b WY 33c 36a 28c,d NA
Source: Zegeer et al. (1 ).
Notes: Populations are as follows: rural: <5,000; small urban: 5,000–50,000; medium urban: 50,000–250,000;
large urban: >250,000.
Values shown represent mean values for the state for each population type except as otherwise noted.
NA = population group does not exist within the state; ⎯ = data not available.
* Because of limited data, small urban values were combined for two groups of states: AL, MS, PR, SC, and
VA and FL, GA, KY, LA, NC, and TN. Medium urban values were combined for AL, FL, and VA.
a
Reported values appeared to be a mix of field observations and statewide values. The latter were
discounted, such that the averages shown are based primarily on values deemed to be field observations,
with some consideration given to nearby states and the value state personnel thought was statewide.
b
The default value was estimated from field observations from nearby states because of insufficient field
data, a lack of data for this road type, or too-heavy reliance on statewide values.
c
The peak period percentage is identical to the daily average percentage for nearly all observations in the
2004 Highway Performance Monitoring System data set. Default values were estimated primarily from the
daily average value but took into account the results from nearby states, particularly the difference
between peak and daily values in those states.
d
This distribution was bimodal, with one group centered on 19% and the other on 44%.
State-Specific Heavy-Vehicle Default Values Chapter 26/Freeway and Highway Segments: Supplemental
Page 26-2 Version 7.0
Two-Lane Multilane Two-Lane Multilane Exhibit 26-2

Highways Highways Highways Highways State-Specific Default Values
Small Small Small Small for Percentage of Heavy
State Rural Urban Rural Urban State Rural Urban Rural Urban Vehicles on Multilane and
AL 6a 6a 4a 6a MT 10c 4c 6c 3c Two-Lane Highways
AK 10 2 6 3 NC 8b 4b 6b 6b
AR 14 7 11 12 ND 14c 3c 12c 7c
AZ 9 11 9 9 NE 10 3 12 5
CA 9 5 9 6 NH 6b 6a 6b 6b
CO 11 4 5 5 NJ 8 7 8 6b
CT 3 3 2 6b NM 17 7 23 12
DC NA NA NA NA NV 17b 5c 10c 6c
DE 7 6 9 8 NY 8 5 8 5
FL 8 4 7 7 OH 11 4 14 9
GA 8b 5b 6b 6b OK 14a 5 17 11
HI 3 3 2 2 OR 12 5 6 9
IA 4c 5c 5c 4c PA 6 3 5 4
ID 12c 7c 16c 9c PR 5 5b 5 6
IL 8 5 8 6 RI 2 1 2 6b
IN 10 6a 12 10 SC 8b 5b 6b 6b
KS 15a 3 12c 6c SD 13c 4c 12 c
7c
KY 16a 6a 9a 6a TN 5 4a 6 4
LA 16c 10c 6b 16 TX 13 9 12 9
MA 3a 3a 7b 6b UT 20c 9c 22c 14c
MD 10 6 12 8 VA 4 2 5 2
ME 5 3 4 3 VT 8 5a 7 6b
MI 9 7a 8 4 WA 15 8a 10 7
MN 9 8a 8 6 WI 4 5a 4 5a
MO 9c 6c 12b 10c WV 6b 6b 5b 6b
MS 14a 5a 6b 6a WY 15c 6c 10c 9c
Source: Zegeer et al. (1 ).
Notes: Populations are as follows: rural: <5,000; small urban: 5,000–50,000.
Values shown represent mean values for the state for each population type except as otherwise noted.
NA = population group does not exist within the state.
a
Reported values appeared to be a mix of field observations and statewide values. The latter were
discounted, such that the averages shown are based primarily on values deemed to be field observations,
with some consideration given to nearby states and the value state personnel thought was statewide.
b
Either there are insufficient field data, such that regional averages were used, or there are no usable field
data, either because there are no data in the state for this road type or because there is a too-heavy
reliance on statewide values for both the peak period and the daily average. In these cases, the default
value was estimated from field observations for nearby states.
c
The peak period percentage is identical to the daily average percentage for all or almost all observations
in the 2004 Highway Performance Monitoring System data set for this cell. Default values were estimated
primarily from the daily average value for this cell, taking into account the results for other similar states in
the same region, and in particular the difference between peak and daily average values in those states.
Chapter 26/Freeway and Highway Segments: Supplemental State-Specific Heavy-Vehicle Default Values
3. TRUCK ANALYSIS USING THE MIXED-FLOW MODEL
INTRODUCTION
This section presents a supplemental procedure that can be used to assess the
operating performance of freeway segments under mixed-flow conditions when
significant truck presence, a prolonged single upgrade, or both exist. This
procedure must be used if the analyst is interested in estimating space mean
speeds and densities for cars and trucks separately or for the mixed-traffic stream.
Chapter 12, Basic Freeway and Multilane Highway Segments, describes a
methodology drawn from this procedure that can be used to assess a segment’s
level of service (LOS) by converting heavy vehicles into passenger cars by using
passenger car equivalent (PCE) values. However, users are cautioned that the
auto-only speeds and densities estimated by the PCE-based procedure are likely
to be an approximation of reality at high truck percentages and on steep
upgrades. For these situations, the mixed-flow model described here is
recommended.
Analysts can also use the mixed-flow model for analyzing downgrades and
both types of general terrain (level and rolling). When the truck percentage is
low or the upgrade is not steep, both the mixed-flow model and the Chapter 12
PCE-based method provide similar results. Chapter 25, Freeway Facilities:
Supplemental, extends the mixed-flow model to freeway facilities with multiple,
composite grades. National research (2) shows that when the truck presence is
low or the upgrade is not steep, both the mixed-flow model and the procedure
applying PCE values provide similar results.
OVERVIEW OF THE METHODOLOGY

The process flow for applying the mixed-flow model is depicted in Exhibit
26-3. Selected parameters referenced in the methodology are indicated in Exhibit
26-4 for a 70-mi/h auto-only traffic stream and a representative mixed-traffic
stream.
Truck Analysis Using the Mixed-Flow Model Chapter 26/Freeway and Highway Segments: Supplemental
Exhibit 26-3
Overview of Operational
Analysis Methodology for
Mixed-Flow Model
Notes: SUT = single-unit truck; TT = tractor-trailer; FFS = free-flow speed; MFM = mixed-flow model.
Exhibit 26-4
Speed–Flow Models for 70-
mi/h Auto-Only Flow and a
Representative Mixed Flow
Notes: BP = breakpoint; FFS = free-flow speed; c = capacity.
Chapter 26/Freeway and Highway Segments: Supplemental Truck Analysis Using the Mixed-Flow Model
Step 1: Input Data

For a typical operational analysis, the analyst must specify the flow rate of
the mixed-traffic stream vmix, grade g, grade length d, SUT percentage PSUT, and
TT percentage PTT for the traffic stream.
Step 2: Compute Mixed-Flow Capacity Adjustment Factor and Capacity

The capacity adjustment factor (CAF) for mixed-flow CAFmix converts auto-
only capacities into mixed-traffic capacities. It is computed with Equation 26-1.
Equation 26-1 𝐶𝐴𝐹mix = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix
where
CAFmix = mixed-flow capacity adjustment factor for the basic freeway segment
(decimal);
CAFao = capacity adjustment factor for the auto-only case (decimal);
CAFT,mix = capacity adjustment factor for percentage of trucks for the mixed-
flow case (decimal); and
CAFg,mix = capacity adjustment factor for grade for the mixed-flow case
(decimal).
CAF for the Auto-Only Case

Because CAFao is used to convert auto-only capacities into mixed-traffic
capacities, it defaults to a value of 1.0 unless other capacity adjustments are in
effect (e.g., weather, incidents, driver population factor).
CAF for Truck Percentage

The CAF for truck percentage CAFT,mix is computed with Equation 26-2.
Equation 26-2 𝐶𝐴𝐹𝑇,mix = 0.53 × 𝑃𝑇 0.72
where PT is the total percentage of SUTs and TTs in the traffic stream (decimal).
CAF for Grade Effect

The CAF for grade effect CAFg,mix accounts for the grade severity, grade length,
and truck presence. It is computed by using Equation 26-3 with Equation 26-3.
Equation 26-3 𝐶𝐴𝐹𝑔,𝑚𝑖𝑥 = 𝜌𝑔,mix × max[0, 0.69 × (𝑒 12.9𝑔 − 1)]
× max[0, 1.72 × (1 − 1.71𝑒 −3.16𝑑 )]
with
8 × 𝑃𝑇 𝑃𝑇 < 0.01
Equation 26-4 𝜌𝑔,mix = {
0.126 − 0.03𝑃𝑇 otherwise
where
ρg,mix = coefficient for grade term in the mixed-flow CAF equation (decimal),
PT = total truck percentage (decimal),
g = grade (decimal), and
d = grade length (mi).
Once CAFmix is computed, the mixed-flow capacity can be computed with

Equation 26-5.
𝐶mix = 𝐶𝑎𝑜 × 𝐶𝐴𝐹mix Equation 26-5
where
Cmix = mixed-flow capacity (veh/h/ln);
Cao = auto-only capacity for the given FFS, from Exhibit 12-6 (pc/h/ln); and
(decimal).
If the input flow rate of the mixed-traffic stream vmix exceeds the mixed-flow
capacity computed in Equation 26-5, then LOS F prevails, and the segment
procedure stops. A facility analysis is recommended under these conditions.
Step 3: Compute Mixed-Flow FFS and FFS Adjustment Factor

Equation 26-6 through Equation 26-8 compute the free-flow travel rates (in
seconds per mile) for SUTs, TTs, and automobiles, respectively, for a specific
segment with a steep grade, high truck percentage, or both. For the purposes of
calculating the automobile free-flow travel rate, the flow rate of the mixed-traffic
stream vmix is assumed to be 1 veh/h/ln when Equation 26-8 is used.
𝜏𝑆𝑈𝑇 = 𝜏𝑆𝑈𝑇,𝑘𝑖𝑛 + 𝛥𝜏 𝑇𝐼 Equation 26-6
𝜏𝑇𝑇 = 𝜏𝑇𝑇,𝑘𝑖𝑛 + 𝛥𝜏𝑇𝐼 Equation 26-7
3,600 Equation 26-8

𝜏𝑎 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆
2.76
𝑣mix 0.46 𝜏𝑆𝑈𝑇,𝑘𝑖𝑛 3,600
+100.42 × ( ) × 𝑃𝑆𝑈𝑇 0.68 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
1.81
𝑣mix 1.36 𝜏𝑇𝑇,𝑘𝑖𝑛 3,600
+110.64 × ( ) × 𝑃𝑇𝑇 0.62 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
where
τa = automobile free-flow travel rate (s/mi),
τSUT = SUT free-flow travel rate (s/mi),
τTT = TT free-flow travel rate (s/mi),
τSUT,kin = kinematic travel rate of SUTs (s/mi),
τTT,kin = kinematic travel rate of TTs (s/mi),
ΔτTI = traffic interaction term (s/mi),
vmix = flow rate of mixed traffic (veh/h/ln),
FFS = base free-flow speed of the basic freeway segment (mi/h),
PSUT = SUT percentage (decimal),
PTT = TT percentage (decimal), and
3,600 = number of seconds in 1 h.
Traffic Interaction Term

The traffic interaction term computed by Equation 26-9 is the contribution of
traffic interactions to mixed-flow FFS. For the purposes of calculating the
automobile free-flow travel rate, the traffic interaction term ΔτTI is set to 0 when
Equation 26-8 is used.
3,600 3,600 1
Equation 26-9 𝛥𝜏𝑇𝐼 = ( − ) × (1 + 3 [ − 1])
𝑆𝑎𝑜 𝐹𝐹𝑆 𝐶𝐴𝐹mix
where
ΔτTI = traffic interaction term (s/mi);
Sao = auto-only speed for the given flow rate, from Equation 26-10 (mi/h);
FFS = base free-flow speed of the basic freeway segment (mi/h); and
from Equation 26-1 (decimal).
Auto-Only Speed for the Given Flow Rate

The auto-only travel rate for the given flow rate is computed with Equation
26-10.
𝑣mix
𝐹𝐹𝑆 ≤ 𝐵𝑃𝑎𝑜
𝐶𝐴𝐹mix
2
Equation 26-10 𝑆𝑎𝑜 = 𝑐 𝑣
(𝐹𝐹𝑆 − ) ( mix − 𝐵𝑃𝑎𝑜) 𝑣mix
𝐷𝑐 𝐶𝐴𝐹mix
𝐹𝐹𝑆 − > 𝐵𝑃𝑎𝑜
{ (𝑐 − 𝐵𝑃𝑎𝑜)2 𝐶𝐴𝐹mix
where
FFS = base free-flow speed of the basic freeway segment (mi/h);
c = base segment capacity, from Exhibit 12-6 (pc/h/ln);
BPao = breakpoint for the auto-only flow condition, from Exhibit 12-6
(pc/h/ln);
Dc = density at capacity = 45 pc/mi/ln; and
CAFmix = mixed-flow capacity adjustment factor for the basic freeway
segment, from Equation 26-1 (decimal).
Kinematic Travel Rates for SUTs and TTs

The kinematic travel rates for SUTs and TTs are obtained from truck travel
time versus distance performance curves on the basis of the truck weight-to-
horsepower ratio, grade, and grade length. Exhibit 26-5 shows truck travel time
versus distance curves for a representative SUT starting from a speed of 70 mi/h.
Alternate representations of how the propulsive and resistive forces vary with
speed can produce slightly different results (e.g., 3, 4).
Exhibit 26-6 shows the corresponding curves for TTs for a base FFS of 70
mi/h. These curves can be used when the base FFS is within 2.5 mi/h of 70 mi/h.
Appendix A provides additional curves for SUTs and TTs for FFS values of 50,
55, 60, 65, and 75 mi/h.
On downgrades, trucks are able to maintain their FFS, and their kinematic
performance is the same as passenger cars. The analyst could use the Chapter 12
PCE-based method instead of the mixed-flow model in those cases.
Exhibit 26-5
SUT Travel Time Versus
Distance Curves for 70-mi/h
FFS
Note: Curves in this graph assume a weight-to-horsepower ratio of 100.
Exhibit 26-6
TT Travel Time Versus
FFS
The x-axis in Exhibit 26-5 and Exhibit 26-6 represents the distance d traveled
by the truck, and the y-axis represents the travel time T to cover the grade length
d. Different curves provide the travel times for different upgrades. The kinematic
space mean travel rate can be computed with Equation 26-11.
𝜏𝑘𝑖𝑛 = 𝑇/𝑑 Equation 26-11
where
τkin = kinematic travel rate (s/mi),
T = travel time (s), and
The maximum grade length shown in Exhibit 26-5 and Exhibit 26-6 is 10,000
ft. When the grade is longer than 10,000 ft, the kinematic travel rate can be
computed with Equation 26-12.
𝑇10000 10,000
Equation 26-12 𝜏𝑘𝑖𝑛 = + 𝛿 (1 − ) × 5,280
𝑑 5,280𝑑
where
τkin = kinematic travel rate (s/mi),
T10000 = travel time at 10,000 ft (s),
δ = slope of the travel time versus distance curve (s/ft),
d = grade length (mi), and
5,280 = number of feet in 1 mi.
The δ value for SUTs and TTs is shown in Exhibit 26-7 and Exhibit 26-8,
respectively, for different combinations of grade and FFS.
Exhibit 26-7 Free-Flow Speed (mi/h)

δ Values for SUTs Grade 50 55 60 65 70 75
–5% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
0% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
2% 0.0136 0.0124 0.0114 0.0105 0.0100 0.0099
3% 0.0136 0.0124 0.0114 0.0113 0.0112 0.0112
4% 0.0136 0.0129 0.0128 0.0128 0.0128 0.0127
5% 0.0146 0.0146 0.0146 0.0146 0.0145 0.0145
6% 0.0165 0.0165 0.0165 0.0165 0.0165 0.0165
7% 0.0186 0.0186 0.0186 0.0186 0.0186 0.0186
8% 0.0208 0.0208 0.0208 0.0208 0.0208 0.0208
Exhibit 26-8 Free-Flow Speed (mi/h)

δ Values for TTs Grade 50 55 60 65 70 75
–5% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
0% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
2% 0.0136 0.0124 0.0119 0.0118 0.0116 0.0115
3% 0.0143 0.0143 0.0142 0.0141 0.0140 0.0138
4% 0.0171 0.0171 0.0171 0.0170 0.0169 0.0168
5% 0.0202 0.0202 0.0202 0.0202 0.0202 0.0202
6% 0.0236 0.0236 0.0236 0.0236 0.0236 0.0236
7% 0.0272 0.0272 0.0272 0.0272 0.0272 0.0272
8% 0.0310 0.0310 0.0310 0.0310 0.0310 0.0310
Once τSUT,kin and τTT,kin are obtained, Equation 26-6 and Equation 26-7 can be
used to add the traffic interaction term to obtain the truck free-flow travel rates
τSUT and τTT. Equation 26-8 can then be used to compute the automobile free-flow
travel rate τa. Again, the mixed-flow rate vmix is assumed to be 1 veh/h/ln when
Equation 26-8 is used to estimate the automobile free-flow travel rate.
Mixed-Flow FFS
Equation 26-13 converts individual free-flow travel rates by mode into a
mixed-flow free-flow travel rate, and Equation 26-14 then converts the mixed-
flow free-flow travel rate into a mixed-flow FFS.
𝜏 = 𝑃𝑎 𝜏𝑎 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇 + 𝑃𝑇𝑇 𝜏𝑇𝑇 Equation 26-13
3,600 3,600
𝐹𝐹𝑆mix = = Equation 26-14
𝜏 𝑃𝑎 𝜏𝑎 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇 + 𝑃𝑇𝑇 𝜏𝑇𝑇
where
τ = mixed-flow free-flow travel rate (s/mi),
τa = automobile free-flow travel rate (s/mi),
τSUT = SUT free-flow travel rate (s/mi),
τTT = TT free-flow travel rate (s/mi),
Pa = automobile percentage (decimal),
PSUT = SUT percentage (decimal),
PTT = TT percentage (decimal), and
FFSmix = mixed-flow free-flow speed (mi/h).
FFS Adjustment Factor

The segment’s speed adjustment factor (SAF) is estimated with Equation
26-15.
𝑆𝐴𝐹mix = 𝐹𝐹𝑆mix /𝐹𝐹𝑆 Equation 26-15
where
SAFmix = mixed-flow speed adjustment factor for the basic freeway segment
(decimal),
FFSmix = mixed-flow free-flow speed (mi/h), and
FFS = base free-flow speed of the basic freeway segment (mi/h).
Step 4: Compute the Speed–Flow Relationship Breakpoint for the

Mixed-Flow Model
The breakpoint is the maximum flow rate up to which speed is maintained at
the adjusted FFS level. It is computed by Equation 26-16 and is depicted in
Exhibit 26-4.
𝐵𝑃mix = max[0, 𝐵𝑃𝑎𝑜 (1 − 0.4𝑃𝑇 0.1 × max[0, 𝑒 30𝑔 + 1] × 𝑑 0.01 )] Equation 26-16
where
BPmix = breakpoint for mixed flow (veh/h/ln);
BPao = breakpoint for the auto-only flow condition, from Exhibit 12-6 (pc/h/ln);
PT = total truck percentage (decimal);
g = grade (decimal); and
Step 5: Compute Mixed-Flow Speeds at Capacity and 90% of Capacity

To determine the mixed-flow speeds for the given mixed-flow rate, mixed-
flow speeds at capacity and 90% of capacity are computed for calibration
purposes. This computation, in turn, requires applying Equation 26-6 through
Equation 26-8 to calculate individual speeds for SUTs, TTs, and automobiles,
respectively. The equations are applied twice, first applying the value of Cmix as
vmix to calculate speed at capacity, and then applying the value of 0.9Cmix as vmix to
calculate speed at 90% of capacity.
The resulting modal travel time rates are converted to modal speeds Sm by
using Equation 26-17.
3,600
Equation 26-17 𝑆𝑚 =
𝜏𝑚
where Sm is the speed (mi/h) for mode m (SUT, TT, or automobile), and τm is the
travel time rate (s/mi) for mode m.
Next, densities for individual modes are computed with Equation 26-18.
Equation 26-18 𝐷𝑚 = 𝑣𝑚 /𝑆𝑚
where Dm is the density (SUT/mi, TT/mi, or pc/mi, depending on the mode) for
mode m, vm is the flow rate (SUT/h, TT/h, or pc/h) for mode m, and Sm is the
speed (mi/h) for mode m.
Finally, the mixed-flow speed used for calibration Scalib is calculated with
Equation 26-19.
3,600
Equation 26-19 𝑆𝑐𝑎𝑙𝑖𝑏 =
𝑃𝑎 𝜏𝑎 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇 + 𝑃𝑇𝑇 𝜏𝑇𝑇
Equation 26-19 is applied twice (i.e., two calibration points are needed), once
using τ values at capacity and again using τ values for 90% of capacity.
Mixed-flow travel rates and mixed-flow speeds are calculated with
Equations 26-13 and 26-14 twice (i.e., two calibration points are needed), once at
capacity and once at 90% capacity.
Step 6: Compute the Exponent for the Mixed-Flow Model Speed–Flow

Curve
The exponent for the speed–flow curve, which describes the rate at which
speed drops as the flow rate increases in the nonlinear portion of the mixed-flow
speed–flow curve (see Exhibit 26-4), is computed with Equation 26-20.
𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,90𝑐𝑎𝑝
ln ( )
𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝
Equation 26-20 𝜙mix = 1.195 ×
0.9𝐶 − 𝐵𝑃
ln ( 𝐶 mix− 𝐵𝑃 mix )
mix mix
where
φmix = exponent for the speed–flow curve (decimal),
FFSmix = mixed-flow free-flow speed (mi/h),
Scalib,90cap = mixed-flow speed at 90% of capacity (mi/h),
Scalib,cap = mixed-flow speed at capacity (mi/h),

Cmix = mixed-flow capacity (veh/h/ln), and
BPmix = breakpoint for mixed flow (veh/h/ln).
Step 7: Compute the Mixed-Flow Speed Under Mixed-Flow Conditions

The mixed-flow speed for mixed-flow conditions is computed by using the
generic form of the basic freeway segment speed–flow model, as shown in
Equation 26-21.
𝐹𝐹𝑆mix 𝑣mix ≤ 𝐵𝑃mix
𝑆mix = Equation 26-21
𝑣mix − 𝐵𝑃mix 𝜙mix
𝐹𝐹𝑆mix − (𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝 ) ( ) 𝑣mix > 𝐵𝑃mix
{ 𝐶mix − 𝐵𝑃mix
where
Smix = mixed-flow speed (mi/h),
FFSmix = mixed-flow free-flow speed (mi/h),
Scalib,cap = mixed-flow speed at capacity (mi/h),
vmix = flow rate of mixed traffic (veh/h/ln),
BPmix = breakpoint for mixed flow (veh/h/ln),
Cmix = mixed-flow capacity (veh/h/ln), and
φmix = exponent for the speed–flow curve (decimal).
Step 8: Compute the Mixed-Flow Density Under Mixed-Flow Conditions

The mixed-flow density is computed by Equation 26-22.
𝐷mix = 𝑣mix /𝑆mix Equation 26-22
where
Dmix = mixed-flow density (veh/mi/ln),
vmix = flow rate of mixed traffic (veh/h/ln), and
Smix = mixed-flow speed (mi/h).
4. ADJUSTMENTS FOR DRIVER POPULATION EFFECTS
The base traffic stream characteristics for basic freeway and multilane
highway segments are representative of traffic streams composed primarily of
commuters or drivers who are familiar with the facility. It is generally accepted
that traffic streams with different characteristics (e.g., recreational trips) use
freeways less efficiently. Although data are sparse and reported results vary
substantially, significantly lower capacities have been reported on weekends,
particularly in recreational areas. Thus, it may generally be assumed the
reduction in capacity extends to service flow rates and service volumes for other
levels of service as well. In addition, it is expected that a reduction in FFS would
be observed when large numbers of unfamiliar drivers are present in a freeway
or multilane highway traffic stream.
The driver population adjustment factor fp has previously been used in the
HCM to reflect the effects of unfamiliar drivers in the traffic stream; it was
applied as an increase in demand volume. The values of fp ranged from 0.85 to
1.00 in most cases, although lower values have been observed in isolated cases.
The HCM recommended the analyst use a value of 1.00 for this factor (reflecting
a traffic stream composed of commuters or other regular drivers), unless there
was sufficient evidence that a lower value should be used. When greater
accuracy was needed, comparative field studies of commuter and noncommuter
traffic flow and speeds were recommended.
With the addition of a unified speed–flow equation in Chapter 12, Basic
Freeway and Multilane Highway Segments, and the ability to adjust both the
base FFS and capacity in all freeway segment chapters (Chapters 12, 13, and 14)
to account for incidents and weather events, the driver population factor is no
longer used. Instead, FFS and capacity adjustment factors SAFpop and CAFpop are
applied in combination with other applicable SAFs and CAFs.
In the absence of new research on driver population effects, recommended
values of SAFpop and CAFpop have been developed that produce similar density
results as those predicted using the former driver population factor approach.
This conversion was performed by using the unified equation of Chapter 12 and
therefore represents a slight approximation in the cases of weaving, merge, and
diverge segments.
Judgment is still required when the analyst applies these adjustments and, in
the absence of information to the contrary, the default value for SAFpop and CAFpop
is always 1.0. Should the analyst expect a significant presence of unfamiliar
drivers, the values shown in Exhibit 26-9 can serve as a guide for the analysis.
Exhibit 26-9 Level of Driver Familiarity CAFpop SAFpop

Recommended CAF and SAF All familiar drivers, regular commuters 1.000 1.000
Adjustments for Driver Mostly familiar drivers 0.968 0.975
Population Impacts Balanced mix of familiar and unfamiliar drivers 0.939 0.950
Mostly unfamiliar drivers 0.898 0.913
All or overwhelmingly unfamiliar drivers 0.852 0.863
Adjustments for Driver Population Effects Chapter 26/Freeway and Highway Segments: Supplemental
5. GUIDANCE FOR FREEWAY CAPACITY ESTIMATION
This section presents guidance for field measuring and estimating freeway
capacity. The section is organized as follows: overall definitions of freeway
capacity, guidance for field data collection using sensors, and guidance for
estimating capacity from the collected data.
FREEWAY CAPACITY DEFINITIONS

Freeway segment capacity is commonly understood to be a maximum flow
rate that is associated with the occurrence of some type of breakdown that in
turn results in lower speeds and higher densities after the breakdown event.
When oversaturation begins, queues develop and vehicles discharge from the
bottleneck at a queue discharge rate that is usually lower than the throughput
rate before the breakdown. This lower discharge rate after a breakdown is also
known as the capacity drop phenomenon. Several key terms related to freeway
capacity are defined below as they apply to this chapter.
Freeway Breakdown
A flow breakdown on a freeway represents the transition from uncongested
to congested conditions, as evidenced by the formation of queues upstream of
the bottleneck and reduced prevailing speeds.
In the HCM freeway methodology, the breakdown event on a freeway
bottleneck is defined as a sudden drop in speed at least 25% below the FFS for a
sustained period of at least 15 min that results in queuing upstream of the
bottleneck.
Recovery
A freeway segment is considered to have recovered from the breakdown
event and the resulting oversaturated conditions when the average speed (or
density) reaches prebreakdown conditions for a minimum duration of 15 min.
The definition of recovery is therefore the inverse of the definition of breakdown,
requiring a recovery to be near prebreakdown conditions (operations above the
speed threshold) for at least 15 min.
The HCM defines the breakdown recovery on a freeway bottleneck as a
return of the prevailing speed to within 10% of the FFS for a sustained period of
at least 15 min, without the presence of queuing upstream of the bottleneck.
Prebreakdown Flow Rate

The prebreakdown flow rate is the flow rate that immediately precedes the
occurrence of a breakdown event. The literature suggests this flow rate does not
have a fixed value, as evidence shows breakdowns are stochastic in nature and
can occur following a range of flow rates. The prebreakdown flow rate is
typically expressed in units of passenger cars per hour per lane. To achieve a
uniform expression, trucks and other heavy vehicles are converted into an
equivalent passenger car traffic stream.
Chapter 26/Freeway and Highway Segments: Supplemental Guidance for Freeway Capacity Estimation
In the HCM, the prebreakdown flow rate is defined as the 15-min average
flow rate that occurs immediately prior to the breakdown event. For the
purposes of this chapter, the prebreakdown flow rate is equivalent to the
segment capacity.
Postbreakdown Flow Rate or Queue Discharge Flow Rate

The postbreakdown flow rate is also referred to as the queue discharge flow
rate or the average discharge flow rate. This flow rate is usually lower than the
prebreakdown flow rate, resulting in a significant loss of freeway throughput
during congestion. Cases in which the postbreakdown flow rate exceeds the
prebreakdown flow rate have been observed, mostly when the prebreakdown
flow rate is low. Studies (5) have indicated the average difference between
postbreakdown and prebreakdown flow rates varies widely, from as little as 2%
to as much as 20%. In the absence of local information, a default value of 7% is
recommended.
In the HCM, the queue discharge flow rate is defined as the average flow
rate during oversaturated conditions (i.e., during the time interval after
breakdown and prior to recovery).
CAPACITY MEASUREMENT LOCATIONS

Research at freeway merging segments (6) has found a breakdown may first
be observed either upstream or downstream of the actual bottleneck. Some
research has indicated a breakdown may first be observed upstream of the
bottleneck, slowly spreading downstream as vehicles accelerate past the start of
the bottleneck. Other research has found the breakdown initially occurs
downstream of the merge point and then moves upstream as a shock wave.
To identify the breakdown event from field data, the following process
should be followed:
• Data are obtained at three sensors: (a) a bottleneck location (e.g., just
downstream of the end of the acceleration lane), (b) at a nearby sensor
location downstream of the bottleneck, and (c) at a nearby sensor location
upstream of the bottleneck.
• Upstream and downstream sensors should be within 0.5 mi of the
bottleneck, and the freeway ideally should have no entry or exit points
between the three sensors (other than, for example, a bottleneck on-ramp).
• The bottleneck detector should be upstream of the beginning of the
deceleration lane or downstream of the end of the acceleration lane to
avoid missing flow in those lanes.
• The analyst evaluates data from the bottleneck sensor to identify a
breakdown by using the definitions provided above.
• The analyst evaluates data from the downstream sensor for the same time
period to ensure no breakdown exists, which indicates congestion at the
bottleneck sensor is unlikely due to spillback from downstream
congestion.
Guidance for Freeway Capacity Estimation Chapter 26/Freeway and Highway Segments: Supplemental
• The analyst evaluates data from the upstream sensor to verify queues are
forming as a result of breakdown at the bottleneck. This check ensures
observed drops in speeds and increases in density at the bottleneck sensor
are indeed due to breakdown.
It is important that the measurements of flows, speeds, and densities used to
estimate capacity are carried out at the correct locations, especially if the data
will be generated from existing fixed freeway sensors, which may or may not be
at the optimal locations to detect breakdown events. Capacity should always be
measured at the bottleneck location. At merge bottlenecks or lane drops, this
location is downstream of the merge point (Exhibit 26-10). At diverge
bottlenecks, this location is upstream of the diverge point (Exhibit 26-11). At
weaving bottlenecks, this location is within the weaving area (Exhibit 26-12).
Exhibit 26-10
Recommended Capacity
Measurement Location for
Merge Bottlenecks
Source: Elefteriadou, Kondyli, and St. George (6).
Upstream Bottleneck Downstream Exhibit 26-11

traffic Recommended Capacity
detector detector detector
Measurement Location for
Diverge Bottlenecks
Location of
capacity
measurement Diverge point
Upstream Bottleneck Downstream Exhibit 26-12

traffic Recommended Capacity
detector detector detector Measurement Location for
Weaving Bottlenecks
Location of
capacity
measurement
Regardless of the bottleneck type, the analyst will be able to identify and
measure capacity only if a breakdown occurs. As discussed below, the
breakdown event is associated with the development of queues that form
upstream of the bottleneck location (i.e., merge point, diverge point, weaving
section) and propagate further upstream, but queues also propagate downstream
as vehicles accelerate past the start of the bottleneck. Once breakdown events are
identified, the analyst will be able to identify the prebreakdown and
postbreakdown flow rates and estimate segment capacity based on the method
discussed in the next section.
CAPACITY ESTIMATION FROM FIELD DATA

To estimate the capacity of the various freeway segments it is important to
analyze data obtained under recurring congestion and under similar operational
and weather conditions. Observations in which adverse weather, incidents, work
zones, or special events were present must be analyzed separately to obtain
capacities under various prevailing conditions. To obtain a reasonable capacity
estimate, it is important to analyze a considerable amount of data over a period
of several months to an entire year.
The recommended method for capacity estimation from sensor data takes
into account that capacity is stochastic. That is, the same flow rate may or may
not be followed by a breakdown. Therefore, during an observation period, both
prebreakdown flow rates and flow rates that are not followed by breakdown
events (uncongested flow rates) are considered. From these flow rates, the
method develops a capacity distribution and then selects a capacity value based
on an acceptable rate of breakdown. Two plausible (and equivalent) freeway
segment capacity definitions are offered:
1. Definition A: Freeway segment capacity is the maximum 15-min flow rate
(in passenger cars per hour per lane) that produces an acceptable (λ%)
rate of breakdown.
2. Definition B: Freeway segment capacity is the maximum 15-min flow rate
(in passenger cars per hour per lane) that ensures stable flow (100 – λ%)
of the time.
The rate of breakdown λ is the ratio of the total number of periods observed
under prebreakdown conditions, divided by the total number of 15-min
uncongested observations under the same flow rate. A default acceptable rate of
breakdown λ of 15% is recommended.
The capacity estimation process follows a series of eight steps and assumes
sensors are placed at the appropriate locations (as discussed above) and are
available to measure prebreakdown flows and ensure the absence of
downstream congestion, which may bias the results described below.
1. Record the distribution of 15-min flow rates (in passenger cars per hour
per lane) during the observation period (preferably a long period). For
example, sampling from the sensor 24 h per day on weekdays over a year
gives approximately 24 × 4 × 250 = 24,000 flow rate observations.
2. Exclude the 15-min analysis periods when the freeway is in breakdown

mode, as defined earlier, which will result in a distribution of
uncongested 15-min flow rates. It is recommended to filter breakdowns
due to nonrecurring sources of congestion, such as severe weather events
or incidents, as the focus is on estimating the bottleneck’s capacity under
recurring congestion conditions.
3. Bin the uncongested flow rates into 100- or 200-pc/h/ln bins.
4. Compute the average flow rate in each bin.
5. For each bin, count the number of times the flow rates in the bin were
immediately followed by the occurrence of a breakdown. In other words,
bin the prebreakdown 15-min flow rates.
6. Calculate the actual probability of breakdown P(BF) in each bin, defined
as the number of times a flow rate bin was in a prebreakdown condition
n(B), divided by the number of times that bin was observed to have
occurred, or n(F). The probability of breakdown P(BF) in each bin is
simply P(BF) = n(B)/n(F).
7. Fit a Weibull distribution (7) to the empirical probability of breakdown
computed in Step 6.
8. Based on the selected threshold breakdown (or stable flow) rate λ or (1 – λ),
determine the resulting capacity value from the Weibull distribution
developed in Step 6 by using Equation 26-23. A value of λ of 15% is
recommended.
𝛽
Capacity = 𝛾 × √−ln (1 − 𝜆) Equation 26-23
where β and γ, respectively, are the shape and scale parameters of the
fitted Weibull distribution, and λ is as defined previously. When λ = 0.15,
the equation simplifies to c = γ (0.163)1/β.
The following example is based on actual data and involves estimating the
capacity of a bottleneck on southbound I-440 in Raleigh, North Carolina. In this
example, sensor data in the vicinity of an on-ramp bottleneck were collected for
260 weekdays from June 2014 to May 2015. The average percentage of trucks
observed in the traffic stream was less than 1%; therefore, the conversion of
trucks into PCEs is ignored for the purposes of this example.
The theoretical number of 15-min observations is 260 days × 96 observations
per day = 24,960 observations. After outliers were removed (observations from
incident and weather events and congested-flow periods), there remained 22,984
periods when flow was deemed uncongested and that represented similar
operational and weather conditions. Within these periods, 192 breakdowns were
identified that met the criteria described above.
Exhibit 26-13 summarizes the computations for this example, using the eight
steps given above. The example illustrates how the process yields a capacity
value based on the recommended 15% breakdown rate.
Exhibit 26-13 [1] [2] [3] [4] [5] [6] [7]

Illustrative Example of the
No. of No. of Observed
Capacity Estimation Procedure
Average Observed 15-min Periods
Flow Rate in Flow Rate 15-min at a Probability of Cumulative
Bins (pc/h/ln) in Bin Uncongested Prebreakdown Breakdown Probability of
From To (pc/h/ln) Periods Flow Rate in Bin Breakdown
0 99 50 4,570 0 0.0% 0.0%
100 199 150 1,657 1 0.1% 0.5%
200 299 250 1,009 3 0.3% 2.1%
300 399 350 765 2 0.3% 3.1%
400 499 450 889 2 0.2% 4.2%
500 599 550 913 0 0.0% 4.2%
600 699 650 746 0 0.0% 4.2%
700 799 750 657 0 0.0% 4.2%
800 899 850 534 0 0.0% 4.2%
900 999 950 458 0 0.0% 4.2%
1,000 1,099 1,050 798 0 0.0% 4.2%
1,100 1,199 1,150 1,801 1 0.1% 4.7%
1,200 1,299 1,250 2,171 2 0.1% 5.7%
1,300 1,399 1,350 1,662 5 0.3% 8.3%
1,400 1,499 1,450 1,185 8 0.7% 12.5%
1,500 1,599 1,550 866 10 1.2% 17.7%
1,600 1,699 1,650 618 13 2.1% 24.5%
1,700 1,799 1,750 495 22 4.4% 35.9%
1,800 1,899 1,850 322 6 1.9% 39.1%
1,900 1,999 1,950 258 16 6.2% 47.4%
2,000 2,099 2,050 301 45 15.0% 70.8%
2,100 2,199 2,150 227 37 16.3% 90.1%
2,200 2,299 2,250 79 18 22.8% 99.5%
2,300 2,399 2,350 3 1 33.3% 100.0%
2,400 2,499 2,450 0 0 NA 100.0%
Sum 22,984 192
Notes: Numbers in brackets indicate column numbers. NA = not applicable.
The exhibit shows 22,984 15-min flow rate observations in Column 4,

equivalent to 5,746 h of observations. Column 5 shows 192 breakdown events.
The probability of breakdown in a bin is computed in Column 6, which is used to
estimate capacity based on the defined λ threshold. Finally, Column 7 shows the
cumulative distribution of prebreakdown flow rates, based on the data in
Column 5.
The information in Exhibit 26-13 is shown graphically in Exhibit 26-14. The
solid black curve to the right shows the Weibull distribution fitted to the data in
Column 6; the actual data are also plotted. The distribution parameters were β =
9.13 and γ = 2,569. Substituting these values into Equation 26-23 and using λ =
0.15 yields a capacity value of 2,105 pc/h/ln. The gray dashed curve to the left in
the exhibit represents the cumulative distribution of prebreakdown flow rates
(i.e., Column 7). In this case, the calculated capacity value corresponded to
approximately the 85th percentile of the prebreakdown flow rate distribution, as
represented by the dotted lines.
Exhibit 26-14
Capacity Estimation Using the
15% Acceptable Breakdown
Rate Method
In summary, the capacity estimation method considers the fact that flow
rates preceding breakdown can also occur at other times without being followed
by a breakdown. The definition of capacity is clear and unambiguous and can be
explained to the HCM user or practitioner without much difficulty. However, the
analyst needs to ensure there are a sufficient number of breakdown observations to be
confident in the calculated capacity value.
6. CONNECTED AND AUTOMATED VEHICLES
INTRODUCTION
This section provides capacity adjustment factors (CAFs) for freeway system
elements to account for the presence of connected and automated vehicles
(CAVs) in the traffic stream. It also provides daily and hourly maximum service
volumes for basic freeway segments for different proportions of CAVs in the
traffic stream. Although CAVs are still a developing technology, transportation
agencies have an immediate need as part of their long-range planning efforts to
account for CAVs’ potential ability to increase existing roadways’ throughput.
At the time of writing, CAVs capable of fully controlling the vehicle for an
entire trip without the possible need for human intervention, either under
specified operated conditions or under any operating condition [i.e., Society of
Automotive Engineers automation levels 4 and 5 (8)], were not yet in production
for consumer use. Although other HCM methodologies are based on empirical
observations of actual vehicles using actual roadway facilities, calibrated simulation,
or both, these approaches are currently infeasible given the absence of level 4 and
5 CAVs in the traffic stream. Instead, uncalibrated simulation modeling was
conducted using CAV logic developed for the Federal Highway Administration.
Details about this modeling are available in a paper (9) available online in HCM
Volume 4 (hcmvolume4.org) in the Technical Reference Library section for
Chapter 26.
CONCEPTS
CAV Technology
CAVs integrate two separate types of technology, communications and
automation. The combination of these technologies is required to achieve
roadway capacity increases, as described below:
• Connected vehicles transmit data about their status to their surroundings
(e.g., roadside infrastructure, other road users). They also receive
information about their surroundings (e.g., traffic conditions, weather
conditions, presence of potential conflicting vehicles, traffic signal timing)
that motorists can use to adjust their driving behavior in response to
conditions present at a given time and location. This exchange of
information offers potential safety, fuel economy, and environmental
benefits. However, it is not clear how connectivity affects car following
and driver behavior and subsequently freeway capacity.
• Automated vehicles take over all or a portion of the driving task. Depending
on the level of automation, a human may still need to take over under
certain conditions. In the absence of connectivity, the information
available to automated vehicles is limited to that which can be gathered
by on-board sensors, which is typically constrained by a sensor’s line of
sight and the rate at which the sensor takes measurements (e.g., 10 times
per second). As a result, for both safety and passenger comfort reasons,
current adaptive cruise control systems offer minimum time gaps that are
Connected and Automated Vehicles Chapter 26/Freeway and Highway Segments: Supplemental
similar to, or longer than, the gaps used by human drivers, and thus may
decrease roadway capacity when in widespread use (10).
• Connected and automated vehicles communicate with each other and with
roadside infrastructure. The connectivity element provides automated
driving systems with more complete information about a vehicle’s
surroundings and enables cooperative vehicle maneuvers that improve
roadway operations. The vehicle’s enhanced detection capabilities, as well
as redundancy in detection, enable an automated driving system to
operate more efficiently and more safely than with only an on-board
system (11). In particular, the cooperative adaptive cruise control (CACC)
feature enabled by vehicle-to-vehicle communication allows CAVs to
safely operate in platoons at shorter headways than possible by either
human-driven vehicles or automated vehicles using adaptive cruise
control only.
Factors Influencing Freeway Capacity with CAVs in the Traffic Stream

Basic Freeway Segments
The HCM’s base capacity for basic freeway segments of 2,400 pc/h/ln
represents an average headway (i.e., back bumper to back bumper) of 1.5 s
between successive vehicles in a lane. Depending on vehicle length and speed, it
represents an intervehicle gap (i.e., back bumper to front bumper) of around 1.1 s.
To the extent that CAVs can (a) safely reduce this average intervehicle gap and
(b) reduce oscillations in the size of the gap between successive vehicles, shorter
headways and thus greater capacities are feasible on basic freeway segments.
Basic freeway segment capacity is also affected by less-than-ideal roadway
geometric conditions (e.g., lane widths less than 12 ft, right-side lateral clearance
less than 6 ft), because human motorists drive more cautiously under these
conditions. However, lane-guidance technology is expected to allow CAVs to
travel along more-constrained freeway sections without changing their speed or
separation from another vehicle.
Drivers’ familiarity with a freeway can also affect capacity, with those less
familiar with a freeway’s geometry and typical traffic patterns likely to drive
more cautiously. CAVs, on the other hand, can be expected to have current
information about roadway geometry and conditions available to them and
therefore would be considered “familiar drivers” wherever they operate.
The proportion of the traffic stream that is composed of CAVs will also
influence the achievable capacity increase. The greater the proportion of CAVs in
the traffic stream, the more frequently the benefits of connectivity can be
realized, because it becomes more likely that one or more CAVs will trail another
CAV and can form a short-headway platoon. In contrast, a CAV following a
human-driven vehicle or an automated vehicle without connectivity will rely on
adaptive cruise control and therefore travel at headways similar to or longer than
those used at present by human drivers. Thus, under base conditions, a much
greater increase in capacity is seen when the proportion of CAVs on a freeway
increases from 80% to 100% than when it increases from 0% to 20% (9). However,
Chapter 26/Freeway and Highway Segments: Supplemental Connected and Automated Vehicles
when a freeway has a lower initial capacity (e.g., due to a lower design speed),
capacity tends to increase more linearly with increasing proportions of CAVs (9).
Freeway Merge, Diverge, and Weaving Segments

The HCM’s capacity-estimation methods for freeway merge, diverge, and
weaving segments use the basic freeway segment capacity as a starting point;
therefore, all of the CAV-related factors that affect basic segment capacity also
affect the capacity of merge, diverge, and weaving segments. CACC is used to
insert a merging CAV into a platoon of CAVs, assuming the platoon’s maximum
length has not yet been reached, or as the new leader of a trailing platoon
otherwise.
Freeway Managed Lane Segments

No specific research on CAV effects on managed lane segments is available.
However, the results for a market penetration of 100% may be used to approximate
the capacity effects of a CAV-only managed lane.
Oversaturated Conditions
No specific research on CAV effects on oversaturated freeways is available.
As such, this section’s CAFs are primarily intended to be used for planning-level
estimates of freeway capacities, and not for detailed operational analyses.
Assumptions Affecting CAV Ability to Provide Higher Capacities

Intervehicle Gap
Given that CAV technology and regulation is still in development,
assumptions necessarily have to be made when estimating CAVs’ potential
capacity benefit. A key assumption used in developing this section’s CAFs was
the minimum achievable intervehicle gap. Factors that could affect the eventual
intervehicle gap include:
• Legal or regulatory requirements dictating a minimum gap.
• Liability concerns on the part of vehicle manufacturers that cause them to
use a more conservative gap length than strictly needed for safety.
• Passenger comfort concerns on the part of vehicle manufacturers to
minimize the amount and magnitude of acceleration and deceleration
needed to maintain intervehicle gaps and facilitate cooperative merging
and lane changing.
• Passenger lack of trust concerns on the part of vehicle owners related to
traveling at high speed close behind another vehicle.
• Need for sufficient gaps to accommodate lane-changing and merging.
• Mechanical differences between vehicles that affect their operational
characteristics, such as braking and acceleration.
The simulation modeling that developed this section’s CAFs incorporated

the following assumptions related to intervehicle gaps (9):
• CAV capability. The modeled CAVs had vehicle-to-vehicle communication
abilities and a working CACC system. CAVs acting as platoon leaders
reverted to adaptive cruise control (i.e., relying on on-board sensors only).
• Human-driven vehicle capability. The operation of human-driven vehicles
was calibrated for three scenarios: 2,400 pc/h/ln (matching the HCM’s
base capacity for basic freeway segments with a 70 mi/h free-flow speed),
2,100 pc/h/ln, and 1,800 pc/h/ln. The latter two scenarios represent
freeway segments with some combination of lower base free-flow speeds,
narrow lanes, limited or no lateral clearance, high ramp density, and
unfamiliar drivers.
• Platooning behavior. CAVs formed platoons. A CAV became the leader of a
platoon when the vehicle in front of it was either a non-CAV or a CAV
that was the last vehicle in a platoon that had reached its maximum
length. Otherwise, a CAV that followed another CAV joined the
preceding CAV’s platoon. One-vehicle platoons were possible, and
relatively common when the proportion of CAVs in the traffic stream was
low. A CAV’s status could change from leader to follower and back,
depending on lane-changing and merging activity.
• Intraplatoon gaps. A distribution of intervehicle gaps within platoons was
used, based on a field study of driver preferences for gap size that used
vehicles whose adaptive cruise control systems were set to allow short
gaps. The distribution consisted of 0.6 s for 57% of vehicles, 0.7 s for 24%,
0.9 s for 7%, and 1.1 s for 12%, resulting in an average intervehicle gap
within platoons of 0.71 s (12).
• Interplatoon gaps. A CAV that was the leader of a platoon operated in
adaptive cruise control mode, with a gap to the next vehicle of 2.0 s, based
on adaptive cruise control settings in commercial use (13).
• Maximum platoon size. The maximum platoon size was 10 passenger cars,
constrained by the need to accommodate lane-changes and merges at
ramps and the need to maintain reliable communication between the
platoon leader and the vehicles toward the rear of the platoon.
System Reliability
The ability of CAVs to safely operate with short intervehicle gaps requires,
among other things, low communications latency (i.e., information can be
quickly exchanged between vehicles and acted upon), vehicle manufacturers to
build vehicles with reliable components, vehicle owners to promptly repair
components if they do break, and regulatory agencies to provide adequate
bandwidth for vehicle-to-vehicle communication. Consistent with other base-
condition assumptions in the freeway methodology (e.g., standard lane widths,
good weather), a base assumption for CAV analysis is that all necessary
communication elements are in place and working with a high degree of
reliability.
Traffic Stream Composition

A key assumption an analyst will need to make when performing a CAV
analysis is the percentage of CAVs that will be in the traffic stream during the
analysis year(s). Once CAVs become available to consumers, it may take many
years for the vehicle fleet to transition to an all-CAV fleet. In 2018, the average
age of light cars and trucks in the United States was just under 12 years (14), and
it takes even longer for the national fleet to turn over. Furthermore, based on past
adoption rates of new automotive technologies such as automatic transmissions,
airbags, and hybrid vehicles, many people may not choose a CAV the first or
even the second time they replace their vehicle (15). On the other hand, if many
urban dwellers decide not to replace their car and rely instead on mobility
services employing CAVs, adoption of CAVs could occur more rapidly than with
prior automotive technologies. The possibility also exists that high-occupancy
vehicle (HOV) lanes could be converted to CAV-only lanes (i.e., 100% CAVs) even
though overall CAV market penetration might be considerably less than 100%.
The simulation modeling that developed this section’s CAFs assumed a
traffic stream consisting of 100% passenger cars. The percentage of CAVs in the
traffic stream was varied from 0% to 100% in 20% increments. Analysts should
consider the latest available information about CAV adoption rates and the
effects of CAV usage on travel demand when performing an analysis of CAV
effects on freeway capacity.
Addressing Uncertain Assumptions in a CAV Analysis

Any evaluation of future conditions requires assumptions about future
population growth, mode choice, travel demand, and travel patterns, among
other factors, none of which are known with great certainty. Adding assumptions
related to CAVs, particularly when based on simulation that cannot yet be
calibrated to actual operating conditions, only increases the uncertainty in the
analysis inputs.
Because of this uncertainty, it is recommended that the CAV CAFs and service
volume tables presented in this section be applied to the evaluation of “what if”
scenarios, rather than being taken as the final word on what will happen once
CAVs become widespread. In particular, the analyst should consider:
• What if the minimum headway permitted by technology, regulation, or policy, or
the average headway produced by different vehicles’ user settings, is longer than
the modeling assumed? In this case, the capacity increase would be less than
predicted by the CAV CAFs or service volume tables.
• How reliable will the necessary communications and automation technology be?
To the extent that individual CAV-capable vehicles must be driven by a
human at any given time due to equipment malfunction, the proportion
of operating CAVs in the traffic stream will be less than the proportion of
CAV-capable vehicles. (Alternatively, the demand will be lower, in the
situation where only vehicles with functioning systems are allowed on the
facility.)
• How quickly will CAV technology become available and adopted, and how will
CAVs affect travel demand? The assumptions made related to these
questions will determine the assumed volume and proportion of CAVs in
the traffic stream, along with the assumed CAF.
CAPACITY ADJUSTMENT FACTORS

Basic Freeway Segments
Exhibit 26-15 provides CAFs for basic freeway segments where CAVs are
present in the traffic stream, CAFCAV. To determine the CAF value, first calculate
the segment’s initial adjusted capacity cadj using Equation 12-8, applying all other
applicable CAFs. Next, determine the CAFCAV value from Exhibit 26-15 based on
the proportion of CAVs in the traffic stream and the initial adjusted capacity,
interpolating as needed. Finally, determine the segment’s adjusted capacity with
CAVs by multiplying the initial adjusted capacity by CAFCAV.
Proportion of CAVs Adjusted Segment Capacity Exhibit 26-15

in Traffic Stream 2,400 pc/h/ln 2,100 pc/h/ln 1,800 pc/h/ln Capacity Adjustment Factors
0 1.00 1.00 1.00 for CAVs for Basic Freeway
20 1.02 1.02 1.15 and Freeway Diverge
40 1.07 1.10 1.27 Segments
60 1.13 1.25 1.40
80 1.22 1.37 1.60
100 1.33 1.52 1.78
Notes: CAV = connected and automated vehicle, defined here as a vehicle with an operating cooperative adaptive
cruise control system.
Interpolate for other CAV proportions and adjusted segment capacities.
Assumptions: Average intervehicle gap within CAV platoons = 0.71 s based on a distribution (see text),
CAV interplatoon gap = 2.0 s, maximum CAV platoon size = 10 pc, human-driven vehicles operate with
average gaps calibrated to the given adjusted segment capacity.
Freeway Diverge Segments

CAFCAV values for freeway diverge segments are determined similarly to
basic freeway segments. First, calculate the segment’s initial adjusted capacity
cmda using Equation 14-21, applying all other applicable CAFs. Next, determine
the CAFCAV value from Exhibit 26-15 based on the percentage of CAVs in the
traffic stream and the initial adjusted capacity, interpolating as needed. Finally,
determine the segment’s adjusted capacity with CAVs by multiplying the initial
adjusted capacity by CAFCAV.
Freeway Merge Segments

CAFCAV values for freeway merge segments where CAVs are present in the
traffic stream are given in Exhibit 26-16, based on the proportion of CAVs in the
traffic stream and interpolating as needed. The segment’s adjusted capacity is
then determined using Equation 14-21 by applying all applicable CAFs,
including CAFCAV.
Exhibit 26-16 Proportion of CAVs in Traffic Stream CAFCAV

Capacity Adjustment Factors 0 1.00
for CAVs for Freeway Merge 20 1.02
Segments 40 1.07
60 1.16
80 1.33
100 1.45
Interpolate for other CAV proportions and adjusted segment capacities.
average gaps calibrated to 2,200 pc/h/ln.
Freeway Weaving Segments

CAFCAV values for freeway weaving segments where CAVs are present in the
traffic stream are given in Exhibit 26-17. The CAF value is determined from the
proportion of CAVs in the traffic stream and the volume ratio (i.e., the weaving
demand flow rate divided by the total demand flow rate in the weaving
segment), interpolating as needed. The weaving segment method in Chapter 13
offers two definitions of capacity, one based on weaving flows and one based on
density. The CAFs in Exhibit 26-17 are intended to be applied to the latter,
density-based measure. The segment’s adjusted capacity is then determined
using Equation 13-9 by applying all applicable CAFs, including CAFCAV.
Exhibit 26-17 Proportion of CAVs Volume Ratio

Capacity Adjustment Factors in Traffic Stream 0.2 0.3 0.4
for CAVs for Freeway Weaving 0 1.00 1.00 1.00
Segments 20 1.03 1.04 1.05
40 1.08 1.08 1.09
60 1.15 1.15 1.13
80 1.23 1.22 1.20
100 1.37 1.37 1.34
Interpolate for other CAV proportions and volume ratios.
The volume ratio is the weaving demand flow rate divided by the total demand flow rate in the segment.
average gaps calibrated to 2,200 pc/h/ln.
SERVICE VOLUME TABLES

Exhibit 26-18 presents daily maximum service volumes at LOS E for basic
freeway segments with CAVs present in the traffic stream. Values in the exhibit
represent 2-way average annual daily traffic (AADT) divided by the total
number of lanes in both directions. Exhibit 26-19 presents hourly maximum
volumes per lane at LOS E for basic freeway segments with CAVs present in the
traffic stream. Assumptions used in creating these exhibits are listed below the
exhibit; see Chapter 12 for definitions of these terms.
Area Proportion of CAVs in Traffic Stream Exhibit 26-18

Type Terrain 0% 20% 40% 60% 80% 100% Daily Maximum Service
Urban Level 19,900 20,500 21,800 24,600 26,800 29,700 Volumes for Basic Freeway
Urban Rolling 19,000 19,900 21,400 24,500 26,800 29,700 Segments with CAV Presence
(2-way veh/day/ln)
Rural Level 16,800 17,900 19,300 22,000 24,400 26,800
Rural Rolling 15,200 17,200 19,100 21,600 24,400 26,800
Values represent the maximum annual average daily traffic per lane at LOS E.
Urban assumptions: Free-flow speed = 70 mph, 5% trucks, PHF = 0.94, K-factor = 0.09, D-factor = 0.60.
Rural assumptions: Free-flow speed = 70 mph, 12% trucks, PHF = 0.94, K-factor = 0.10, D-factor = 0.60.
CAV assumptions: Average intervehicle gap within CAV platoons = 0.71 s based on a distribution (see
text), CAV interplatoon gap = 2.0 s, maximum CAV platoon size = 10 pc, human-driven vehicles operate
with average gaps calibrated to 2,400 pc/h/ln.
Area Proportion of CAVs in Traffic Stream Exhibit 26-19

Type Terrain 0% 20% 40% 60% 80% 100% Hourly Maximum Service
Urban Level 2,150 2,210 2,350 2,660 2,900 3,200 Volumes for Basic Freeway
Urban Rolling 2,050 2,150 2,310 2,640 2,900 3,200 Segments with CAV Presence
(veh/h/ln)
Rural Level 2,010 2,140 2,310 2,640 2,900 3,200
Rural Rolling 1,820 2,060 2,290 2,580 2,900 3,200
Values represent the maximum analysis hour volume per lane at LOS E.
Urban assumptions: Free-flow speed = 70 mph, 5% trucks, PHF = 0.94, K-factor = 0.09, D-factor = 0.60.
Rural assumptions: Free-flow speed = 70 mph, 12% trucks, PHF = 0.94, K-factor = 0.10, D-factor = 0.60.
CAV assumptions: Average intervehicle gap within CAV platoons = 0.71 s based on a distribution (see
text), CAV interplatoon gap = 2.0 s, maximum CAV platoon size = 10 pc, human-driven vehicles operate
with average gaps calibrated to 2,400 pc/h/ln.
7. FREEWAY AND MULTILANE HIGHWAY

EXAMPLE PROBLEMS
Exhibit 26-20 lists the seven example problems provided in this section. The
problems demonstrate the computational steps involved in applying the
automobile methodology to basic freeway and multilane highway segments. All
the freeway example problems address urban freeway situations.
Exhibit 26-20 Example

List of Freeway and Multilane Problem Description Application
Highway Example Problems 1 Four-lane freeway LOS Operational analysis
2 Number of lanes required for target LOS Design analysis
3 Six-lane freeway LOS and capacity Operational and planning analysis
4 LOS on a five-lane highway with a two-way
Operational analysis
left-turn lane
5 Mixed-flow operational performance Operational analysis
6 Severe weather effects on a basic freeway
Operational analysis
segment
7 Basic managed lane segment Operational analysis
EXAMPLE PROBLEM 1: FOUR-LANE FREEWAY LOS

The Facts
• Four-lane freeway (two lanes in each direction)
• Lane width = 11 ft
• Right-side lateral clearance = 2 ft
• Commuter traffic (regular users)
• Peak hour, peak direction demand volume = 2,000 veh/h
• Traffic composition: 5% trucks
• Peak hour factor (PHF) = 0.92
• One cloverleaf interchange per mile
• Level terrain
• Facility operates under ideal conditions (no incidents, work zones, or
weather events).
Comments
The task is to find the expected LOS for this freeway during the worst 15 min
of the peak hour. With one cloverleaf interchange per mile, the total ramp
density will be 4 ramps/mi.
Step 1: Input Data

All input data are specified above.
Freeway and Multilane Highway Example Problems Chapter 26/Freeway and Highway Segments: Supplemental
Step 2: Estimate and Adjust FFS

The FFS of the freeway is estimated from Equation 12-2 as follows:
𝐹𝐹𝑆 = 75.4 − 𝑓𝐿𝑊 − 𝑓𝑅𝐿𝐶 − 3.22 × 𝑇𝑅𝐷 0.84
The adjustment for lane width is selected from Exhibit 12-20 for 11-ft lanes
(1.9 mi/h). The adjustment for right-side lateral clearance is selected from Exhibit
12-21 for a 2-ft clearance on a freeway with two lanes in one direction (2.4 mi/h).
The total ramp density is 4 ramps/mi. Then
𝐹𝐹𝑆 = 75.4 − 1.9 − 2.4 − 3.22 × 40.84 = 60.8 mi/h
Because the facility is operating under ideal conditions, the SAF used in
Equation 12-5 is 1, and FFSadj = FFS.
Step 3: Estimate and Adjust Capacity

The capacity of the freeway is estimated from Equation 12-6 as follows:
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (60.8 − 50) = 2,308 pc/h/ln
Because the facility is operating under ideal conditions, the CAF used in
Equation 12-8 is 1, and cadj = c.
Step 4: Adjust Demand Volume

The demand volume must be adjusted to a flow rate that reflects passenger
cars per hour per lane under equivalent base conditions by using Equation 12-9.
𝑉
𝑣𝑝 =
𝑃𝐻𝐹 × 𝑁 × 𝑓𝐻𝑉
The demand volume is given as 2,000 veh/h. The PHF is specified to be 0.92,
and there are two lanes in each direction. The driver population consists of
regular users (commuters). Trucks make up 5% of the traffic stream, so a heavy-
vehicle adjustment factor must be determined.
From Exhibit 12-25, the PCE for trucks is 2.0 for level terrain. The heavy-
vehicle adjustment factor is then computed with Equation 12-10.
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1)
1
𝑓𝐻𝑉 = = 0.952
1 + 0.05(2.0 − 1)
then
2,000
𝑣𝑝 = = 1,142 pc/h/ln
0.92 × 2 × 0.952
Because this value is less than the base capacity of 2,308 pc/h/ln for a freeway
with FFS = 60.8 mi/h, LOS F does not exist, and the analysis continues to Step 5.
Chapter 26/Freeway and Highway Segments: Supplemental Freeway and Multilane Highway Example Problems
Step 5: Estimate Speed and Density

The FFS of the basic freeway segment is now estimated along with the
demand flow rate (in passenger cars per hour per lane) under equivalent base
conditions. Using the equations provided in Exhibit 12-6, the breakpoint for a
60.8-mi/h FFS speed–flow curve is
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 𝐹𝐹𝑆𝑎𝑑𝑗 )] × 𝐶𝐴𝐹 2 = 1,568 pc/h/ln
As the flow rate of 1,142 pc/h/ln is less than the breakpoint value of 1,568
pc/h/ln, the freeway operates within the constant-speed portion of the speed–
flow curve, so S = 60.8 mi/h. The density of the traffic stream may now be
computed from Equation 12-11.
𝑣𝑝 1,142
𝐷= = = 18.8 pc/mi/ln
𝑆 60.8
Step 6: Determine LOS

From Exhibit 12-15, a density of 18.8 pc/mi/ln corresponds to LOS C, but it is
close to the boundary for LOS B, which is a maximum of 18 pc/mi/ln. This
solution could also be calculated graphically from Exhibit 12-16, as illustrated in
Exhibit 26-21.
Exhibit 26-21
Example Problem 1: Graphical
Solution
Discussion
This basic freeway segment of a four-lane freeway is expected to operate at
LOS C during the worst 15 min of the peak hour. It is important to note that the
operation, although at LOS C, is close to the LOS B boundary. In most
jurisdictions, this operation would be considered to be quite acceptable.
EXAMPLE PROBLEM 2: NUMBER OF LANES REQUIRED FOR TARGET LOS

The Facts
• Demand volume = 4,000 veh/h (one direction)
• Level terrain
• Traffic composition: 8% SUTs and buses
• Provision of 12-ft lanes
• Provision of 6-ft right-side lateral clearance
• PHF = 0.85
• Ramp density = 3 ramps/mi
• Target LOS = D
weather events).
Comments
This example problem is a classic design application of the methodology.
The number of lanes needed to provide LOS D during the worst 15 min of the
peak hour is to be determined.
Step 1: Input Data


FFS is estimated by using Equation 12-2. Because the lane width and lateral
clearance to be provided on the new freeway will be 12 ft and 6 ft, respectively,
there are no adjustments for these features from Exhibit 12-20 or Exhibit 12-21.
The total ramp density is given as 3 ramps/mi. Then
𝐹𝐹𝑆 = 75.4 − 𝑓𝐿𝑊 − 𝑓𝑅𝐿𝐶 − 3.22 × 𝑇𝑅𝐷 0.84
𝐹𝐹𝑆 = 75.4 − 0 − 0 − 3.22 × 30.84 = 67.3 mi/h
Because the facility is operating under ideal conditions, the SAF used in
Equation 12-5 is 1, and FFSadj = FFS.

The capacity of the freeway is estimated from Equation 12-6.
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (67.3 − 50) = 2,373 pc/h/ln
Step 4: Estimate Number of Lanes Needed

Because this is a design analysis, Step 4 of the operational analysis
methodology is modified. Equation 12-23 may be used directly to determine the
number of lanes needed to provide at least LOS D.
𝑉
𝑁=
𝑀𝑆𝐹𝑖 × 𝑃𝐻𝐹 × 𝑓𝐻𝑉
A value of the maximum service flow rate must be selected from Exhibit 12-
37 for an FFS of 65 mi/h and LOS D. Note that this exhibit only provides these
values in 5-mi/h increments; therefore, FFS is rounded to 65 mi/h. The
corresponding maximum service flow rate is 2,060 pc/h/ln.
The PHF is given as 0.85. A heavy-vehicle factor for 8% trucks must be
determined by using Exhibit 12-25 for level terrain. The PCE of trucks on level
terrain is 2.0, so the heavy-vehicle adjustment based on Equation 12-10 is
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1)
1
𝑓𝐻𝑉 = = 0.926
1 + 0.08(2 − 1)
and
4,000
𝑁= = 2.5 ln
2,060 × 0.85 × 0.926
It is not possible to build 2.5 lanes. To provide a minimum of LOS D, it will
be necessary to provide three lanes in each direction, or a six-lane freeway.
At this point, the design application ends. It is possible, however, to consider
what speed, density, and LOS will prevail when three lanes are actually
provided. Therefore, the example problem continues with Steps 5 and 6.

In pursuing additional information, the problem now reverts to an
operational analysis of a three-lane basic freeway segment with a demand
volume of 4,000 pc/h.
Equation 12-9 is used to compute the actual demand flow rate per lane under
equivalent base conditions.
𝑉
𝑣𝑝 =
4,000
𝑣𝑃 = = 1,694 pc/h/ln
0.85 × 3 × 0.926
From Exhibit 12-6, the breakpoint for a speed–flow curve with FFS equal to
67.3 is
In this case, the demand flow rate of 1,694 pc/h/ln exceeds the breakpoint
value of 1,308 pc/h/ln, and the average speed will be less than the FFS.
The expected speed of the traffic stream may be estimated by using either
Exhibit 12-7 (for a graphical solution) or Equation 12-1 as follows:
𝑐𝑎𝑑𝑗 𝑎
(𝐹𝐹𝑆𝑎𝑑𝑗 − 𝐷 ) (𝑣𝑝 − 𝐵𝑃)
𝑐
𝑆 = 𝐹𝐹𝑆𝑎𝑑𝑗 − 𝑎
(𝑐𝑎𝑑𝑗 − 𝐵𝑃)
2,373
(67.3 − 45 ) (1,694 − 1,308)2
𝑆 = 67.3 − = 65.4 mi/h
(2,373 − 1,308)2
The density may now be computed from Equation 12-11.
𝑣𝑝 1,694
𝐷= = = 25.9 pc/mi/ln
𝑆 65.4

Entering Exhibit 12-15 with a density of 25.9 pc/mi/ln, the LOS is C, but that
density is very close to the boundary of LOS D, which is 26 pc/mi/ln.
Discussion
The resulting LOS is C, which represents a better performance than the target
design. Although the minimum number of lanes needed was 2.5, which would
have produced a minimal LOS D, providing three lanes yields a density that is
close to the LOS C boundary. In any event, the target LOS of the design will be
met by providing a six-lane basic freeway segment.
EXAMPLE PROBLEM 3: SIX-LANE FREEWAY LOS AND CAPACITY

The Facts
• Volume of 5,000 veh/h (one direction, existing)
• Volume of 5,788 veh/h (one direction, in 3 years)
• Rolling terrain
• Three lanes in each direction
• FFS = 70 mi/h (measured)
• PHF = 0.96
• Traffic growth = 5% per year
weather events).
Comments
This example consists of two operational analyses, one for the present
demand volume of 5,000 pc/h and one for the demand volume of 5,788 pc/h
expected in 3 years. In addition, a planning element is introduced: Assuming
traffic grows as expected, when will the capacity of the roadway be exceeded?
This analysis requires that capacity be determined in addition to the normal
output of operational analyses.
Step 1: Input Data


Step 2 is not needed, as the FFS was directly measured (70 mi/h). Because the
facility is operating under ideal conditions, the SAF used in Equation 12-5 is 1,
and FFSadj = FFS.

The capacity of the freeway is estimated from Equation 12-6.
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (70 − 50) = 2,400 pc/h/ln

In this case, two demand volumes will be adjusted by using Equation 12-9.
𝑉
𝑣𝑝 =
The PHF is given as 0.96, and there are three lanes in each direction. The
heavy-vehicle factor will reflect 4% trucks in rolling terrain. From Exhibit 12-25,
the PCE for trucks in rolling terrain is 3.0. Equation 12-10 then gives
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1)
1
𝑓𝐻𝑉 = = 0.926
1 + 0.04(3.0 − 1)
Two values of vp are computed: one for present conditions and one for
conditions in 3 years.
5,000
𝑣𝑝 (present) = = 1,875 pc/h
0.96 × 3 × 0.926
5,788
𝑣𝑝 (future) = = 2,171 pc/h
0.96 × 3 × 0.926

Two values of speed and density will be estimated, one each for the present
and future conditions. Equation 12-1 will be used to estimate speeds. First, the
breakpoint for the speed–flow curve is computed from Exhibit 12-6.
One equation applies to both cases; a 70-mi/h FFS with a flow rate over 1,200
pc/h/ln is used.
𝑐
2,400
(70 − 45 ) (1,875 − 1,200)2
𝑆(present) = 70 − = 64.7 mi/h
(2,400 − 1,200)2
2,400
(70 − 45 ) (2,171 − 1,200)2
𝑆(future) = 70 − = 59.1 mi/h
(2,400 − 1,200)2
The corresponding densities may now be estimated from Equation 12-11.
𝑣𝑃
𝐷=
𝑆
1,875
𝐷(present) = = 29.0 pc/mi/ln
64.7
2,171
𝐷(future) = = 36.7 pc/mi/ln
59.1

From Exhibit 12-15, the LOS for the present situation is D, and the LOS for
the future scenario (in 3 years) is E due to the increase in density.
Step 7: Determine When Capacity Will Be Reached

Step 7 is an additional step for this problem. To determine when capacity
will be reached, the capacity of the basic freeway segment must be estimated.
From Exhibit 12-37, the maximum service flow rate for LOS E on a basic freeway
segment with a 70-mi/h FFS is 2,400 pc/h/ln. This flow rate is synonymous with
capacity.
The analyst must be sure the capacity and demand flow rates compared in
Step 7 are measured on the same basis. The 2,400 pc/h/ln is a flow rate under
equivalent base conditions. The demand flow rate in 3 years was estimated to be
2,171 pc/h/ln on this basis. These two values, therefore, may be compared. As an
alternative, the capacity could be computed for prevailing conditions with
Equation 12-24.
𝑆𝐹𝐸 = 𝑀𝑆𝐹𝐸 × 𝑁 × 𝑓𝐻𝑉
𝑆𝐹𝐸 = 2,400 × 3 × 0.926 = 6,667 veh/h
This capacity, however, is stated as a flow rate. The demand volume is stated
as an hourly volume. Thus, a service volume for LOS E is needed as estimated
from Equation 12-25.
𝑆𝑉𝐸 = 𝑆𝐹𝐸 × 𝑃𝐻𝐹 = 6,667 × 0.96 = 6,400 veh/h
The problem may be solved either by comparing the demand volume of
5,788 veh/h (in 3 years) with the hourly capacity of 6,400 veh/h or by comparing
the demand flow rate under equivalent base conditions of 2,171 pc/h/ln with the
base capacity of 2,400 pc/h/ln. With the hourly demand volume and capacity,
6,400 = 5,788 × (1.05)𝑛
𝑛 = 2.06 years
On the basis of the forecasts of traffic growth, the basic freeway segment
described will reach capacity within 5 years. The demand value of 5,788 veh/h
occurs 3 years from the present per the problem description, and the calculation
above shows capacity is reached after an additional 2 years. If this result is added
to the 3-year planning horizon, capacity will be reached within 5 years of the
time of the analysis.
Discussion
The LOS on this segment will reach LOS E within 3 years due to the increase
in density. The demand is expected to exceed capacity within 5 years. Given the
normal lead times for planning, design, and approvals before the start of
construction, it is probable that planning and preliminary design for an
improvement should be started immediately.
EXAMPLE PROBLEM 4: LOS ON A FIVE-LANE HIGHWAY WITH

A TWO-WAY LEFT-TURN LANE
The Facts
• Lane width: 12 ft
• Lateral clearance, both sides of the roadway: 12 ft
• Traffic composition: 6% trucks, with default truck mix (30% SUTs, 70%
TTs)
• Access points per mile: eastbound = 10; westbound = 0
• PHF = 0.90
• Median type: two-way left-turn lane
• Peak hour demand: 1,500 veh/h
• The upgrade occurs in the westbound direction
• Posted speed limit = 45 mi/h
Comments
A 6,600-ft segment of a five-lane highway (two travel lanes in each direction
plus a two-way left-turn lane) is on a 3.5% grade. At what LOS is the facility
expected to operate in each direction?
There is one segment in each direction. The upgrade and downgrade
segments on the 3.5% grade must be analyzed separately. This example is more
complex than the previous examples because the segment characteristics are not
all the same, particularly the number of access points. Because no base FFS is
given, it will be estimated as the speed limit plus 7 mi/h, or 45 + 7 = 52 mi/h.
Step 1: Input Data

All input data are given above.

FFS is estimated by using Equation 12-3.
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑓𝐿𝑊 − 𝑓𝑇𝐿𝐶 − 𝑓𝑀 − 𝑓𝐴
In this case, the base FFS is estimated to be 52 mi/h. The lane width is 12 ft, which
is the base condition; therefore, fLW = 0.0 mi/h (Exhibit 12-20). The lateral
clearance is 12 ft at each roadside, but a maximum value of 6 ft may be used. A
two-way left-turn lane is considered to have a median lateral clearance of 6 ft.
Thus, the total lateral clearance is 6 + 6 = 12 ft, which is also a base condition.
Therefore, fTLC = 0.0 mi/h (Exhibit 12-22). The median-type adjustment fM is also
0.0 mi/h (Exhibit 12-23).
For this example problem, only the access-point density produces a nonzero
adjustment to the base FFS. The eastbound (EB) segment (3.5% downgrade) has
10 access points/mi. From Exhibit 12-24, the corresponding FFS adjustment is 2.5
mi/h. The westbound (WB) segment (3.5% upgrade) has 0 access points/mi and a
corresponding FFS adjustment of 0.0 mi/h. Therefore,
𝐹𝐹𝑆𝐸𝐵 = 52.0 − 0.0 − 0.0 − 0.0 − 2.5 = 49.5 mi/h
𝐹𝐹𝑆𝑊𝐵 = 52.0 − 0.0 − 0.0 − 0.0 − 0.0 = 52.0 mi/h

The capacity of the multilane highway segment is estimated as follows from
Equation 12-7.
𝑐 = 1,900 + 20 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 45)
𝑐𝐸𝐵 = 1,900 + 20 × (49.5 − 45) = 1,990 pc/h/ln
𝑐𝑊𝐵 = 1,900 + 20 × (52.0 − 45) = 2,040 pc/h/ln

Demand volume is adjusted by using Equation 12-9.
𝑉
𝑣𝑝 =
To compute the heavy-vehicle adjustment factor fHV, PCEs for trucks are
needed for (a) the 3.5%, 6,600-ft upgrade and (b) the 3.5%, 6,600-ft downgrade.
The segment is 1.25 mi (6,600/5,280 ft) long. The following values are obtained
from Exhibit 12-26:
• Eastbound: 2.24 (using 6% trucks, a 2% downgrade, and 1.25-mi grade
length). Note that all downgrades exceeding 2% use the PCE values for a
2% downgrade.
• Westbound: 3.97 (using 6% trucks, a 3.5% upgrade, and a 1.25-mi grade
length).
The heavy-vehicle adjustment factors fHV for each segment are calculated
from Equation 12-10.
1
𝑓𝐻𝑉,𝐸𝐵 = = 0.93
1 + 0.06 × (2.24 − 1)
1
𝑓𝐻𝑉,𝑊𝐵 = = 0.85
1 + 0.06 × (3.97 − 1)
The segments’ flow rates are then calculated as
1,500
𝑣𝑝,𝐸𝐵 = = 896 pc/h/ln
0.90 × 2 × 0.93
1,500
𝑣𝑝,𝑊𝐵 = = 980 pc/h/ln
0.90 × 2 × 0.85

Speed is estimated with Equation 12-1 or the graph in Exhibit 12-7. With
Equation 12-1, both demand flow rates are less than the multilane highway
breakpoint value of 1,400 pc/h/ln. Therefore, the speeds S are equal to FFS. The
densities are computed from Equation 12-11.
𝑣𝑝,𝐸𝐵 896
𝐷𝐸𝐵 = = = 18.1 pc/mi/ln
𝑆𝐸𝐵 49.5
𝑣𝑝,𝑊𝐵 980
𝐷𝑊𝐵 = = = 18.8 pc/mi/ln
𝑆𝑊𝐵 52

LOS is found by comparing the densities of the segments with the criteria in
Exhibit 12-15. As both densities are greater than 18 pc/mi/ln, both upgrade and
downgrade segments operate at LOS C.
Discussion
Even though the upgrade and downgrade segments operate at LOS C, they
are very close to the LOS B boundary (18.0 pc/mi/ln). Both directions of the
multilane highway on this grade operate well.
EXAMPLE PROBLEM 5: MIXED-FLOW FREEWAY OPERATIONS

This example illustrates the application of the mixed-flow model for an
extended single grade on a six-lane rural freeway.
The Facts
• 2-mi basic segment on a 5% upgrade
• Traffic composition: 5% SUTs and 10% TTs
• FFS = 65 mi/h
• Mixed-traffic flow rate = 1,500 veh/h/ln
Comments
The task is to estimate the segment’s speed and density. Given the significant
truck presence (15%) and the 5%, 2-mi grade, the mixed-flow model should be
applied.
Step 1: Input Data

Step 2: Compute Mixed-Flow Capacity Adjustment Factor

Capacity is computed with Equation 26-1.
𝐶𝐴𝐹mix = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix
There are three terms in the equation. The CAF for auto-only CAFao is 1.00, as
no driver population, weather, incident, or work zone adjustments are specified
in the problem statement.
The truck effect term is computed with Equation 26-2.
𝐶𝐴𝐹𝑇,mix = 0.53 × 𝑃𝑇 0.72 = 0.53 × 0.150.72 = 0.135

The grade effect term is computed with Equation 26-3 and Equation 26-4.
𝐶𝐴𝐹𝑔,mix = 𝜌𝑔,mix × max[0, 0.69 × (𝑒 12.9𝑔 − 1)]
× max [0, 1.72 × (1 − 1.71𝑒 −3.16𝑑 )]
𝜌𝑔,mix = 0.126 − 0.03𝑃𝑇 = 0.126 − (0.03)(0.15) = 0.1215
𝐶𝐴𝐹𝑔,mix = 0.1215 × max[0, 0.69 × (𝑒 (12.9)(0.05) − 1)]
× max[0, 1.72 × (1 − 1.71𝑒 (−3.16)(2) )] = 0.131
then
𝐶𝐴𝐹mix = 1 − 0.135 − 0.131 = 0.734
The mixed-flow capacity is then computed from Equation 26-5.
𝐶mix = 𝐶𝑎𝑜 × 𝐶𝐴𝐹mix
The auto-only capacity 𝐶𝑎𝑜 is computed from Exhibit 12-6.
𝐶𝑎𝑜 = 2,200 + 10(𝐹𝐹𝑆 − 50) = 2,200 + 10 × (65 − 50) = 2,350 pc/h/ln
then
𝐶mix = 2,350 × 0.734 = 1,725 veh/h/ln
As the mixed-traffic flow rate of 1,500 veh/h/ln is less than the mixed-flow
capacity of 1,725 veh/h/ln, the analysis can proceed.
Step 3: Compute Mixed-Flow FFS and FFS Adjustment Factor

Equation 26-6 through Equation 26-8 compute the free-flow travel rates for
SUTs, TTs, and automobiles, respectively. The FFS of this basic freeway segment
is 65 mi/h. Truck performance curves for free-flow speeds other than 70 ± 2.5
mi/h are provided in Appendix A. The 65-mi/h curves for SUTs and TTs are
found in Exhibit 26-A4 and Exhibit 26-A9, respectively.
The travel time for a SUT TSUT at a point 10,000 ft along the upgrade can be
read directly from Exhibit 26-A4 by observing where the 5% upgrade curve
intersects 10,000 ft: 134 s. Similarly, the travel time for a TT TTT is 173 s.
As the grade is 2 mi (10,560 ft) long and the performance curves only provide
values up to 10,000 ft, Equation 26-12 is used to determine the travel time rates
for the upgrade as a whole. The slope of the travel time versus distance curve δ,
which is used in Equation 26-12, can be determined from Exhibit 26-7 for SUTs
and Exhibit 26-8 for TTs. The δ values are 0.0146 and 0.0202, respectively.
Then
𝑇𝑆𝑈𝑇,10000𝑓𝑡 10,000
𝜏𝑘𝑖𝑛,𝑆𝑈𝑇 = + 𝛿 (1 − ) × 5,280
𝑑 5280𝑑
134 10,000
𝜏𝑘𝑖𝑛,𝑆𝑈𝑇 = + 0.0146 (1 − ) × 5,280 = 71.1 s/mi
2 10,560
173 10,000
𝜏𝑘𝑖𝑛,𝑇𝑇 = + 0.0202 (1 − ) × 5,280 = 92.2 s/mi
2 10,560
As this step’s objective is to compute the FFS of the mixed-traffic stream, the
traffic interaction term ΔτTI is zero, and the mixed-flow rate is set to 1 veh/h/ln.
The SUT, TT, and auto travel time rates are then computed using Equation 26-6
through Equation 26-8.
𝜏𝑆𝑈𝑇,𝐹𝐹𝑆 = 71.1 + 0 = 71.1 s/mi
𝜏𝑇𝑇,𝐹𝐹𝑆 = 92.2 + 0 = 92.2 s/mi
3,600
𝜏𝑎,𝐹𝐹𝑆 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆
2.76
𝑣mix 0.46 𝜏𝑘𝑖𝑛,𝑆𝑈𝑇 3,600
+100.42 × ( ) × 𝑃𝑆𝑈𝑇 0.68 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
1.81
𝑣mix 1.36 𝜏𝑘𝑖𝑛,𝑇𝑇 3,600
+110.64 × ( ) × 𝑃𝑇𝑇 0.62 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
3,600
𝜏𝑎,𝐹𝐹𝑆 = +0
65
1 0.46 71.1 3,600 2.76
+100.42 × ( ) × 0.050.68 × max [0, − ]
1,000 100 (65 × 100)
1 1.36 92.2 3,600 1.81
+110.64 × ( ) × 0.10.62 × max [0, − ]
1,000 100 (65 × 100)
𝜏𝑎,𝐹𝐹𝑆 = 55.4 s/mi

Mixed-flow travel rates and speeds are computed with Equation 26-13 and
Equation 26-14.
𝜏mix,𝐹𝐹𝑆 = 𝑃𝑎 𝜏𝑎,𝐹𝐹𝑆 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇,𝐹𝐹𝑆 + 𝑃𝑇𝑇 𝜏𝑇𝑇,𝐹𝐹𝑆
𝜏mix,𝐹𝐹𝑆 = (0.85)(55.4) + (0.05)(71.1) + (0.1)(92.2) = 59.87 s/mi
3,600 3,600
𝐹𝐹𝑆mix = = = 60.1 mi/h
𝜏mix,𝐹𝐹𝑆 59.87
Finally, the segment’s SAF is estimated with Equation 26-15.
𝐹𝐹𝑆mix 60.1
𝑆𝐴𝐹mix = = = 0.92
𝐹𝐹𝑆 65
Step 4: Compute the Mixed-Flow Rate at the Breakpoint

The breakpoint is calculated from Equation 26-16.
𝐵𝑃mix = max[0, 𝐵𝑃𝑎𝑜 (1 − 0.4𝑃𝑇 0.1 × max[0, 𝑒 30𝑔 + 1] × 𝑑 0.01 )]
where the auto-only breakpoint is calculated by using an equation given in
Exhibit 12-6.
𝐵𝑃𝑎𝑜 = [1,000 + 40 × (75 − 𝐹𝐹𝑆)] × 𝐶𝐴𝐹 2
𝐵𝑃𝑎𝑜 = [1,000 + 40 × (75 − 65)] × 12 = 1,400 veh/h/ln
then
𝐵𝑃mix = max[0, (1,400)(1 − 0.4(0.15)0.1 × max[0, 𝑒 30×0.05 + 1] × 20.01 )]
𝐵𝑃mix = 0 veh/h/ln
This result implies that speeds drop immediately at zero flow (i.e., the
mixed-flow FFS cannot be sustained even at low flows).
Step 5: Compute Modal and Mixed-Flow Speeds at Capacity and 90%

of Capacity
The speeds and densities for each mode at capacity and 90% of capacity are
calculated in this step. Equation 26-6 through Equation 26-8 are applied twice
more, once for a flow rate equal to the mixed-flow capacity of 1,725 veh/h/ln
calculated in Step 2, and again for a flow rate equal to 90% of capacity. Applying
these equations requires determining the traffic interaction term ΔτTI, which in
turn requires determining the equivalent auto-only speed Sao.
The calculation process will be demonstrated for conditions at capacity. The
value of Cmix determined in Step 2 (1,725 veh/h/ln) will be used as vmix in the
calculations.
The auto-only speed at capacity is computed by Equation 26-10.
𝑣mix
𝐹𝐹𝑆 ≤ 𝐵𝑃𝑎𝑜
𝐶𝐴𝐹mix
2
𝑆𝑎𝑜 = 𝑐 𝑣
(𝐹𝐹𝑆 − ) ( mix − 𝐵𝑃𝑎𝑜) 𝑣mix
𝐷𝑐 𝐶𝐴𝐹mix
𝐹𝐹𝑆 − > 𝐵𝑃𝑎𝑜
{ (𝑐 − 𝐵𝑃𝑎𝑜)2 𝐶𝐴𝐹mix
The value of vmix/CAFmix is 1,725/0.734 = 2,350 veh/h/ln, which is greater than
the auto-only breakpoint of 1,400 veh/h/ln calculated in Step 4. Therefore, the
second of the two equations is applied.
2
2,350 1,725
(65 − ) (0.734 − 1,400)
𝑆𝑎𝑜,𝑐𝑎𝑝 = 65 − 45 = 52.2 mi/h
(2,350 − 1,400)2
The traffic interaction term can now be computed with Equation 26-9.
3,600 3,600 1
𝛥𝜏𝑇𝐼,𝑐𝑎𝑝 = ( − ) × (1 + 3 [ − 1])
𝑆𝑎𝑜,𝑐𝑎𝑝 𝐹𝐹𝑆 𝐶𝐴𝐹mix
3,600 3,600 1
𝛥𝜏𝑇𝐼,𝑐𝑎𝑝 = ( − ) × (1 + 3 [ − 1]) = 28.3 s/mi
52.2 65 0.734
Equation 26-6 through Equation 26-8 are now applied to find the modal
travel time rates at capacity.
𝜏𝑆𝑈𝑇,𝑐𝑎𝑝 = 𝜏𝑆𝑈𝑇,𝑘𝑖𝑛 + 𝛥𝜏𝑇𝐼 = 71.1 + 28.3 = 99.4 s/mi
𝜏𝑇𝑇,𝑐𝑎𝑝 = 𝜏𝑇𝑇,𝑘𝑖𝑛 + 𝛥𝜏𝑇𝐼 = 92.2 + 28.3 = 120.5 s/mi
3,600
𝜏𝑎,𝑐𝑎𝑝 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆
2.76
𝑣mix 0.46 𝜏𝑆𝑈𝑇,𝑘𝑖𝑛 3,600
+100.42 × ( ) × 𝑃𝑆𝑈𝑇 0.68 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
1.81
𝑣mix 1.36 𝜏𝑇𝑇,𝑘𝑖𝑛 3,600
+110.64 × ( ) × 𝑃𝑇𝑇 0.62 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
3,600
𝜏𝑎,𝑐𝑎𝑝 = + 28.3
65
1,725 0.46 71.1 3,600 2.76
+100.42 × ( ) × 0.050.68 × max [0, − ]
1,000 100 (65 × 100)
1,725 1.36 92.2 3,600 1.81
+110.64 × ( ) × 0.10.62 × max [0, − ]
1,000 100 (65 × 100)
𝜏𝑎,𝑐𝑎𝑝 = 92.9 s/mi

Based on these travel rates, the overall mixed-traffic space mean speed at
capacity can be calculated with Equation 26-19.
3,600
𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝 =
𝑃𝑎 𝜏𝑎 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇 + 𝑃𝑇𝑇 𝜏𝑇𝑇
3,600
𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝 = = 37.5 mi/h
(0.85)(92.9) + (0.05)(99.4) + (0.1)(120.5)
The same process is used to calculate the mixed-traffic speed at 90% of
capacity (vmix = 0.9 × 1,725 = 1,553 veh/h/ln). The resulting calculation results are
2
2,350 1,553
(65 − ) (0.734 − 1,400)
𝑆𝑎𝑜,90𝑐𝑎𝑝 = 65 − 45 = 57.7 mi/h
(2,350 − 1,400)2
3,600 3,600 1
𝛥𝜏𝑇𝐼,90𝑐𝑎𝑝 = ( − ) × (1 + 3 [ − 1]) = 14.6 s/mi
57.7 65 0.734
𝜏𝑆𝑈𝑇,90𝑐𝑎𝑝 = 71.1 + 14.6 = 85.7 s/mi
𝜏𝑇𝑇,90𝑐𝑎𝑝 = 92.2 + 14.6 = 106.8 s/mi
3,600
𝜏𝑎,90𝑐𝑎𝑝 = + 14.6
65
1,553 0.46 71.1 3,600 2.76
+100.42 × ( ) × 0.050.68 × max [0, − ]
1,000 100 (65 × 100)
1,553 1.36 92.2 3,600 1.81
+110.64 × ( ) × 0.10.62 × max [0, − ]
1,000 100 (65 × 100)
𝜏𝑎,90𝑐𝑎𝑝 = 78.0 s/mi

3,600
𝑆𝑐𝑎𝑙𝑖𝑏,90𝑐𝑎𝑝 = = 44.3 mi/h
(0.85)(78.0) + (0.05)(85.7) + (0.1)(106.8)
Step 6: Compute the Exponent for the Speed–Flow Curve

The exponent for the speed–flow curve is computed from Equation 26-20.
𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,90𝑐𝑎𝑝
ln ( 𝐹𝐹𝑆 − 𝑆 )
mix 𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝
𝜙mix = 1.195 ×
0.9𝐶 − 𝐵𝑃
ln ( 𝐶 mix− 𝐵𝑃 mix )
mix mix
60.1 − 44.3
ln (
𝜙mix = 1.195 × 60.1 − 37.5) = 4.07
1,553 − 0
ln (
1,725 − 0)
Step 7: Compute the Mixed-Flow Speed Under Mixed-Flow Conditions

The mixed-flow speed under mixed-flow conditions is computed by
Equation 26-21.
𝐹𝐹𝑆mix 𝑣mix ≤ 𝐵𝑃mix
𝑆mix =
𝑣mix − 𝐵𝑃mix 𝜙mix
𝐹𝐹𝑆mix − (𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝 ) ( ) 𝑣mix > 𝐵𝑃mix
{ 𝐶mix − 𝐵𝑃mix
The mixed-flow rate is 1,500 veh/h/ln, which is greater than the breakpoint.
Therefore,
1,500 − 0 4.07
𝑆mix = 60.1 − (60.1 − 37.5) ( ) = 47.3 mi/h
1,725 − 0
Step 8: Compute the Mixed-Flow Density Under Mixed-Flow Conditions

The final step is to compute the mixed-flow density by using Equation 26-22.
𝑣mix 1,500
𝐷mix = = = 31.7 veh/mi/ln
𝑆mix 47.3
Comparison with the PCE-Based Approach

For comparison purposes, the following procedure show the results for this
case if the PCE-based approach explained in Chapter 12 is applied.
Step 1: Input Data


For basic freeway segments, Equation 12-2 can be used to estimate FFS.
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑓𝐿𝑊 − 𝑓𝑅𝐿𝐶 − 3.22 × 𝑇𝑅𝐷 0.84
For the purposes of comparing the two methods with respect to truck effects
on FFS, the lane width, lateral clearance, and ramp density adjustment factors
can be neglected. Then,
𝐹𝐹𝑆 = 65 − 0 − 0 − 3.22 × 00.84 = 65 mi/h
The adjusted FFS is computed from Equation 12-5, assuming no weather or
incident effects.
𝐹𝐹𝑆𝑎𝑑𝑗 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹
𝐹𝐹𝑆𝑎𝑑𝑗 = 65 × 1 = 65 mi/h

Equation 12-6 is used to compute the capacity of a basic freeway segment.
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (65 − 50) = 2,350 pc/h/ln
Assuming no adverse weather conditions or incidents, the adjusted capacity
from Equation 12-8 is then
𝑐𝑎𝑑𝑗 = 𝑐 × 𝐶𝐴𝐹 = 2,350 × 1 = 2,350 pc/h/ln

This basic freeway segment is in a rural area with more TTs than SUTs.
Therefore, the PCE table for 30% SUTs and 70% TTs (Exhibit 12-26) will be used.
As stated in the Facts section of the example problem, the grade is 5% for 2 mi.
There are no values specifically for a 5% grade in Exhibit 12-26; therefore, PCE
values will be interpolated from the values for 4.5% and 5.5%. As the maximum
grade length provided in the exhibit is 1 mi for these two grades, values for a 1-
mi grade will also apply to longer grades. For a 1-mi, 4.5% grade, the PCE value
for 15% trucks is 3.11; and the PCE value for a 1-mi, 5.5% grade with 15% trucks
is 3.51. Interpolating between these two values for a 5% grade results in a PCE of
3.31.
The heavy-vehicle factor can be computed with Equation 12-10.
1 1
𝑓𝐻𝑉 = = = 0.743
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.15 × (3.31 − 1)
Equation 12-9 is used to adjust the demand volume to account for truck
presence. The freeway is a three-lane facility and the driver population is
assumed to be all local drivers.
𝑉 1,500 × 3
𝑣𝑝 = = = 2,019 pc/h/ln
𝑃𝐻𝐹 × 𝑁 × 𝑓𝐻𝑉 1 × 3 × 0.743

The speed can be read directly from Exhibit 12-7 for a demand flow rate of
2,019 pc/h/ln. Under base conditions, the mean speed of the traffic stream is 59.6
mi/h as calculated from Equation 26-1.
Equation 12-11 is used to compute density.
𝑣𝑃 2,019
𝐷= = = 33.9 pc/mi/ln
𝑆 59.6
If the density above is multiplied by the heavy-vehicle factor, then the
mixed-flow density Dmix can be estimated as follows:
𝐷mix = 𝐷 × 𝑓𝐻𝑉 = 33.9 × 0.743 = 25.2 veh/mi/ln
The PCE-based density of 25.2 veh/mi/ln is about 22% lower than 32.6
veh/mi/ln, which is the density predicted in Step 8 of the mixed-flow model.
Dmix is the mixed-flow density, not an auto-only flow density. As such, it cannot
be used to derive LOS.
EXAMPLE PROBLEM 6: SEVERE WEATHER EFFECTS ON A BASIC

FREEWAY SEGMENT
The Facts
• Four-lane freeway (two lanes in each direction)
• Peak hour, peak direction demand volume = 2,000 veh/h
• PHF = 0.92
• Rolling terrain
• Facility operates under heavy snow conditions (CAF = 0.78; SAF = 0.86).
Comments
The task is to find the expected LOS for this freeway during the worst 15 min
of the peak hour under heavy snow conditions. With one cloverleaf interchange
per mile, the total ramp density will be 4 ramps/mi. This example problem is
similar to Example Problem 1, with the only change being the presence of heavy
snow.
Step 1: Input Data


The FFS of the freeway is estimated from Equation 12-2 as follows:
𝐹𝐹𝑆 = 75.4 − 𝑓𝐿𝑊 − 𝑓𝑅𝐿𝐶 − 3.22 × 𝑇𝑅𝐷 0.84
The adjustment for lane width is selected from Exhibit 12-20 for 11-ft lanes
(1.9 mi/h). The adjustment for right-side lateral clearance is selected from Exhibit
12-21 for a 2-ft clearance on a freeway with two lanes in one direction (2.4 mi/h).
The total ramp density is 4 ramps/mi. Then
𝐹𝐹𝑆 = 75.4 − 1.9 − 2.4 − 3.22 × 40.84 = 60.8 mi/h
A free-flow speed adjustment factor (SAF) for heavy snow conditions can be
obtained from Exhibit 11-5 in Chapter 11, Freeway Reliability Analysis, by
interpolating between the values for 60 and 65 mi/h (0.86 and 0.85, respectively),
resulting in a SAF of 0.86. No other speed adjustments are made, as no incidents
were specified in the problem statement and because the driver population was
specified to be commuters. The SAF is applied through Equation 12-5.
𝐹𝐹𝑆𝑎𝑑𝑗 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 = 60.8 × 0.86 = 52.3 mi/h

Exhibit 11-5 also provides a CAF of 0.78 for heavy snow conditions,
applicable to all FFS values. As with the SAF in Step 2, no other capacity
adjustments apply in this situation. The freeway’s capacity is then estimated
𝑐 = 𝐶𝐴𝐹(2,200 + 10 × [𝐹𝐹𝑆𝑎𝑑𝑗 − 50])
𝑐 = 0.78 × (2,200 + 10 × [52.3 − 50]) = 1,734 pc/h/ln

The demand volume is adjusted by using Equation 12-9 to a flow rate that
reflects passenger cars per hour per lane under equivalent base conditions.
𝑉
𝑣𝑝 =
The demand volume is given as 2,000 veh/h. The PHF is specified to be 0.92,
and there are two lanes in each direction. Trucks make up 5% of the traffic
stream, so a heavy-vehicle adjustment factor must be determined.
From Exhibit 12-25, the PCE for trucks is 3.0 for rolling terrain. The heavy-
vehicle adjustment factor is then computed by using Equation 12-10.
1 1
𝑓𝐻𝑉 = = = 0.909
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.05(3 − 1)
then
2,000
𝑣𝑝 = = 1,195 pc/h/ln
0.92 × 2 × 0.91
Because this value is less than the base capacity of 1,743 pc/h/ln for a freeway
with an FFS of 52.3 mi/h, LOS F conditions do not exist, and the analysis
continues to Step 5.

The FFS of the basic freeway segment is now estimated along with the
conditions. Using the equations provided in Exhibit 12-6, the breakpoint for a
53.5-mi/h FFS speed–flow curve is
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 𝐹𝐹𝑆𝑎𝑑𝑗 )] × (𝐶𝐴𝐹)2
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 52.3)] × (0.78)2 = 1,161 pc/h/ln
Because the flow rate is greater than the breakpoint value, the operating
speed of the segment is estimated from Equation 12-1, by using a value of 2 for
the exponent calibration parameter a from Exhibit 12-6.
𝑐
1,734
(52.3 − 45 ) (1,195 − 1,161)2
𝑆 = 52.3 − = 52.3 mi/h
(1,734 − 1,161)2
The density may now be computed from Equation 12-11.

𝑣𝑝 1,195
𝐷= = = 22.8 pc/mi/ln
𝑆 52.3

From Exhibit 12-15, a density of 22.8 pc/mi/ln corresponds to LOS C.
Discussion
This basic freeway segment of a four-lane freeway is expected to operate at
LOS C during the worst 15 min of the peak hour under heavy snow conditions,
with an average speed of 52.3 mi/h and a density of 22.8 pc/mi/ln. By contrast,
the same facility under no adverse weather conditions would be expected to
operate at an FFS of 60.8 mi/h and a density of 19.7 pc/mi/ln, but still at LOS C.
Although the segment’s performance is affected by the snow, the overall LOS is
unchanged.
However, the segment’s capacity is reduced from 2,308 to 1,734 pc/h/ln,
which means the snow effect would be more severe at elevated volume-to-
capacity ratios, particularly as the segment approached capacity. For elevated
flow rates, the snow condition is expected to result in further deterioration of
speed and breakdown at lower flow rates.
EXAMPLE PROBLEM 7: BASIC MANAGED LANE SEGMENT

The Facts
• Six-lane freeway with two general purpose lanes and one managed lane
in each direction
• Peak hour, peak direction demand volume in the general purpose lanes =
2,000 veh/h (Case 1) or 3,800 veh/h (Case 2)
• Peak hour, peak direction demand volume in the managed lane (both
cases) = 1,300 veh/h
• Continuous access separation between the managed and general purpose
lanes
• FFS = 60 mi/h for both the managed and general purpose lanes
• Traffic composition: 7.5% trucks, using the default truck mix for both the
managed and general purpose lanes
• PHF = 0.92
• Level terrain
weather events).
Comments
The task is to find the expected LOS for this freeway for both the managed
and general purpose lanes during the worst 15 min of the peak hour for the two
described cases. With one cloverleaf interchange per mile, the total ramp density
will be 4 ramps/mi.
Step 1: Input Data


The facility’s FFS is given as 60 mi/h for both the managed and general
purpose lanes. Because the facility is operating under ideal conditions, the SAF
used in Equation 12-5 is 1.

The capacity of the freeway general purpose lanes is estimated from
Equation 12-6 as follows:
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (60 − 50) = 2,300 pc/h/ln
As the freeway is operating under ideal conditions, no capacity adjustment is
made for the general purpose lanes (i.e., CAF = 1 in Equation 12-8).
The capacity of the managed lane is calculated with Equation 12-14.
𝑐𝑎𝑑𝑗 = 𝐶𝐴𝐹 × (𝑐75 − 𝜆𝑐 × [75 − 𝐹𝐹𝑆𝑎𝑑𝑗 ])
As with the general purpose lanes, CAF = 1 for the managed lane. The values
of the parameters C75 and λc are obtained from Exhibit 12-30, and are 1,800 and
10, respectively, for continuous access separation. Then
𝑐𝑎𝑑𝑗 = 1.00 × (1,800 − 10 × [75 − 60]) = 1,650 pc/h/ln

The demand volume is adjusted by using Equation 12-9 to a flow rate that
reflects passenger cars per hour per lane under equivalent base conditions.
𝑉
𝑣𝑝 =
The demand volume is given as 2,000 veh/h and 3,800 veh/h for Cases 1 and
2, respectively. The PHF is specified to be 0.92, and there are two lanes in each
direction. Trucks make up 5% of the traffic stream, so a heavy-vehicle adjustment
factor must be determined.
From Exhibit 12-25, the PCE for trucks is 2.0 for level terrain. The heavy-
vehicle adjustment factor is then computed using Equation 12-10.
1 1
𝑓𝐻𝑉 = = 0.93
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.075(2.0 − 1)
Then for Case 1,
2,000
𝑣𝑝,𝐺𝑃,Case1 = = 1,169 pc/h/ln
0.92 × 2 × 0.93
and for Case 2,

3,800
𝑣𝑝,𝐺𝑃,Case2 = = 2,221 pc/h/ln
0.92 × 2 × 0.93
The flow rate on the managed lane is
1,300
𝑣𝑝,𝑀𝐿 = = 1,519 pc/h/ln
0.92 × 1 × 0.93
Because all the flow rates are less than their corresponding capacities, LOS F
conditions do not exist, and the analysis continues to Step 5.

The FFS of the basic freeway segment is now estimated, along with the
conditions. Based on the equations provided in Exhibit 12-6, the breakpoint for a
60-mi/h FFS speed–flow curve is
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 𝐹𝐹𝑆𝑎𝑑𝑗 )] × (𝐶𝐴𝐹)2
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 60)] × (1.00)2 = 1,600 pc/h/ln
In Case 1, the flow rate is less than the breakpoint value of 1,600 pc/h/ln. As
this flow rate is in the constant-speed portion of the curve, SGP,Case1 = 60 mi/h. The
density of the traffic stream is computed from Equation 12-11.
𝑣𝑝 1,169
𝐷𝐺𝑃,Case1 = = = 19.5 pc/mi/ln
𝑆 60
In Case 2, the flow rate is higher than the breakpoint. Therefore, the speed is
computed with Equation 12-1, by using a value of 2 for the exponent calibration
parameter a from Exhibit 12-6, as follows:
(𝐹𝐹𝑆𝑎𝑑𝑗 − ) (𝑣𝑝 − 𝐵𝑃)
𝐷𝑐
2,300
(60 − 45 ) (2,221 − 1,600)2
𝑆𝐺𝑃,Case2 = 60 − = 53.0 mi/h
(2,300 − 1,600)2
Density is computed with Equation 12-11.
𝑣𝑝 2,221
𝐷𝐺𝑃,Case2 = = = 41.9 pc/mi/ln
𝑆 53
To compute the managed lane speed, the breakpoint first needs to be
computed by using Equation 12-13 and values for the parameters BP75 and λBP
from Exhibit 12-30.
𝐵𝑃𝑀𝐿 = [𝐵𝑃75 + 𝜆𝐵𝑃 × (75 − 𝐹𝐹𝑆𝑎𝑑𝑗 )] × 𝐶𝐴𝐹2
𝐵𝑃𝑀𝐿 = [500 + 0 × (75 − 60)] × (1.00)2 = 500 pc/h/ln
Because the managed lane flow rate is higher than the breakpoint, three
speeds, S1, S2, and S3, need to be computed by using Equations 12-15, 12-17, and
12-19, respectively (with parameters from Exhibit 12-30), as follows:
𝑆1 = 𝐹𝐹𝑆𝑎𝑑𝑗 − 𝐴1 × min(𝑣𝑝 , 𝐵𝑃) = 60 − 0 × min(1,519, 500) = 60 mi/h

𝑐𝑎𝑑𝑗
(𝑆1,𝐵𝑃 − 𝑛𝑓 )
𝐾𝑐 𝐴2
𝑆2 = 𝐴2 (𝑣𝑝 − 𝐵𝑃)
1,650
(60 − 30 )
𝑆2 = (1,519 − 500)2.5 = 3.7 mi/h
(1,650 − 500)2.5
𝑐𝑎𝑑𝑗 𝑐𝑎𝑑𝑗
( 𝑛𝑓 ) − ( 𝑓 )
𝐾𝑐 𝐾𝑐 2
𝑆3 = 2 (𝑣𝑝 − 𝐵𝑃)
1,650 1,650
( 30 ) − ( )
𝑆3 = 45 (1,519 − 500)2 = 14.4 mi/h
(1,650 − 500) 2
The space mean speed of the managed lane is given by Equation 12-12.
𝑆1 𝑣𝑝 ≤ 𝐵𝑃
𝑆𝑀𝐿 = {
𝑆1 − 𝑆2 − 𝐼𝑐 × 𝑆3 𝐵𝑃 < 𝑣𝑝 ≤ 𝑐
Because the managed lane’s demand flow of 1,519 pc/h/ln is greater than the
breakpoint value of 500 pc/h/ln calculated in Step 4, the second of the two
equations applies. To apply this equation, the value of the indicator variable Ic
must first be determined from Equation 12-18.
0 𝐾𝐺𝑃 ≤ 35 pc/mi/ln
𝐼𝑐 = { or segment type is Buffer 2, Barrier 1, or Barrier 2
1 otherwise
In Case 1, the density of the adjacent general purpose lane is less than 35
pc/mi/ln, as determined in Step 5. As a result, the indicator variable 𝐼𝑐 will have a
value of zero. Thus, the managed lane speed in Case 1 will be
𝑆𝑀𝐿,Case1 = 60 − 3.7 − (0 × 14.4) = 56.3 mi/h
In Case 2, the density of the adjacent general purpose lane is greater than 35
pc/ln/mi, and therefore the indicator variable 𝐼𝑐 will have a value of 1. The
managed lane speed in Case 2 will be
𝑆𝑀𝐿,Case2 = 60 − 3.7 − (1 × 14.4) = 41.9 mi/h
The managed lane density for the two cases is given by Equation 12-11.
𝑣𝑝 1,519
𝐷𝑀𝐿,Case1 = = = 27.0 pc/mi/ln
𝑆 56.3
𝑣𝑝 1,519
𝐷𝑀𝐿,Case2 = = = 36.3 pc/mi/ln
𝑆 41.9

The managed lane facility’s density of 27.0 pc/mi/ln under Case 1
corresponds to LOS D, but it is close to the LOS C boundary, which has a
maximum value of 26 pc/mi/ln. In Case 2, the density of 36.3 pc/mi/ln
corresponds to LOS E.
Discussion
In this example, the managed lane’s operating speed and density have been
investigated for two operating conditions in the general purpose lanes. When
high-density conditions exist in the general purpose lanes, the managed lane’s
operational speed is reduced and, as a consequence, the managed lane operates
at a worse LOS than when lower-density conditions exist in the general purpose
lanes.
8. TWO-LANE HIGHWAY EXAMPLE PROBLEMS
Exhibit 26-22 lists the five example problems provided in this section. The
problems demonstrate the computational steps involved in applying the two-
lane highway automobile and bicycle methodologies.
Exhibit 26-22 Problem

List of Two-Lane Highway Number Description Type of Analysis
Example Problems 1 Level, straight, Passing Constrained segment Operational analysis
2 Passing Constrained segment with horizontal curves Operational analysis
3 Facility analysis in level terrain Operational analysis
4 Facility analysis in mountainous terrain Operational analysis
5 Bicycle LOS Planning analysis
EXAMPLE PROBLEM 1: LEVEL, STRAIGHT, PASSING CONSTRAINED

SEGMENT
This example problem illustrates the computation of the LOS in a single
direction on a straight, 0.75-mi-long Passing Constrained segment in level terrain.
This example problem follows the flowchart of analysis steps outlined in Exhibit
15-9.
The Facts
The segment has the following known characteristics:
• Segment length = 3,960 ft (0.75 mi);
• Segment type = Passing Constrained;
• No upstream passing lanes;
• Vehicle count in the analysis direction = 752 veh/h;
• PHF = 0.94;
• Posted speed limit: 50 mi/h;
• Percent heavy vehicles (%HV) = 5%;
• Percent grade = 0%;
• Horizontal curvature = none;
• Lane width = 12 ft;
• Shoulder width = 6 ft; and
• Access points = 0.
Objective
Estimate the LOS in the subject direction on the two-lane highway segment
as described.
Step 1: Identify Facility Study Boundaries and Segmentation

The limits of the segment were identified following the guidance given in
Step 1 on page 15-15. The characteristics of this segment were determined by
examination to be essentially homogenous. These characteristics included the
Two-Lane Highway Example Problems Chapter 26/Freeway and Highway Segments: Supplemental
ability to pass, lane geometry, grades, lane and shoulder widths, posted speed
limit, traffic demands, adjacent land uses, and driveways.
A field examination of the segment determined that it met the definition of a
Passing Constrained segment, being a segment in which “passing in the
oncoming lane is either prohibited or is effectively negligible due to geometric or
sight distance limitations.”
Step 2: Determine Demand Flow Rates, Capacity, and d/c Ratio

In this step, the hourly demand volume at the upstream entrance of the
directional segment being evaluated is converted to a peak 15-min flow rate by
applying the peak hour factor (PHF) using Equation 15-1:
𝑉𝑑 752
𝑣𝑑 = = = 800 veh/h
𝑃𝐻𝐹 0.94
The capacity of a Passing Constrained segment is 1,700 veh/h, as stated in the
description of Step 2 on page 15-18. The demand flow rate is less than capacity;
therefore the calculation process proceeds to Step 3. Note that it is only necessary
to compute the actual opposing flow rate for Passing Zone segments.
Step 3: Determine Vertical Alignment Classification

The segment is assigned a vertical alignment classification of 1, based on
Exhibit 15-11 for a level (0% grade), 0.75-mi-long segment. From Exhibit 15-10,
the segment length of 0.75 mi is between the minimum (0.25 mi) and maximum
(3.0 mi) lengths for a Passing Constrained segment of vertical class 1, and
therefore no adjustment is needed to the segment length.
Step 4: Determine the Free-Flow Speed

As stated on page 15-18, direct field measurement of FFS is preferred. In this
case it is not feasible to measure the FFS, so it will be estimated using the
procedure given in Step 4.
Because this is a Passing Constrained segment, the opposing flow rate vo is
set at 1,500 veh/h in Equation 15-4 for the purposes of computing FFS, regardless
of the actual opposing flow rate. First, the base free-flow speed BFFS is estimated
using Equation 15-2. Next, Equation 15-4 through Equation 15-6 are used to
determine factors relating to lane and shoulder width, access-point density, and
heavy-vehicle percentage, which are used in the estimation of FFS. Finally, the
FFS is estimated by Equation 15-3.
𝐵𝐹𝐹𝑆 = 1.14 × 𝑆𝑝𝑙 = 1.14 × 50 = 57.0 mi/h
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)

𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 6) = 0
𝐴𝑃𝐷 0
𝑓𝐴 = min ( , 10) = min ( , 10) = 0
4 4
𝑎 = max[0.0333, 0] = 0.0333 Coefficients a0 through a5 are
all 0 for vertical class 1 (Exhibit
15-12), so the right side of
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑎(𝐻𝑉%) − f LS − f A Equation 15-6 reduces to 0.
Chapter 26/Freeway and Highway Segments: Supplemental Two-Lane Highway Example Problems
𝐹𝐹𝑆 = 57.0 − (0.0333)(5) − 0 − 0 = 56.83 mi/h

There are no geometry-related FFS adjustments for 12-ft lanes and 6-ft
shoulders. There are also no adjustments for access points (driveways or streets),
because this segment has no access points.
Step 5: Estimate the Average Speed

Because the demand flow rate in the subject direction is greater than 100
veh/h, the equations given in Step 5 are used to estimate the average speed.
Step 5a: Calculate the Slope Coefficient

The slope coefficient m determines how rapidly the average speed is
estimated to decrease as a function of the entering flow rate. It is computed as a
function of six coefficients b0 to b5, which are obtained from Exhibit 15-13 for a
Passing Constrained segment. For a segment of vertical class 1, these coefficients
are 0.0558, 0.0542, 0.3278, 0.1029, 0, and 0, respectively. Equation 15-8 is then
used to determine the slope coefficient.
𝑣𝑜
𝑚 = max [𝑏5 , 𝑏0 + 𝑏1 × 𝐹𝐹𝑆 + 𝑏2 × √ + max(0, 𝑏3 ) × √𝐿
1,000
+ max(0, 𝑏4 ) × √𝐻𝑉% ]
1,500
𝑚 = max [0,0.0558 + 0.0542 × 56.83 + 0.3278 × √
1,000
+ max(0, 0.1029) × √0.75 + max(0, 0) × √5 ]
𝑚 = 3.626
Step 5b: Calculate the Power Coefficient

The power coefficient p is used to estimate how fast the average speed
decreases at higher flow rates. The equation uses nine coefficients f0 to f8, which
are obtained from Exhibit 15-19 for a Passing Constrained segment. For a
segment of vertical class 1, all of these coefficients take on values of 0, except for
f0 (0.67576), f3 (0.12060), and f4 (−0.35919). Equation 15-11 is then used to
determine the power coefficient.
𝑣𝑜 𝑣𝑜
𝑝 = max [𝑓8 , 𝑓0 + 𝑓1 × 𝐹𝐹𝑆 + 𝑓2 × 𝐿 + 𝑓3 × + 𝑓4 × √ + 𝑓5 × 𝐻𝑉%
1,000 1,000
+ 𝑓6 × √𝐻𝑉% + 𝑓7 × (𝐿 × 𝐻𝑉%)]
1,500
𝑝 = max [0,0.67576 + 0 × 56.83 + 0 × 0.75 + 0.12060 ×
1,000
1,500
− 0.35919 × √ + 0 × 5 + 0 × √5 + 0 × (0.75 × 5)]
1,000
𝑝 = 0.41676
Step 5c: Calculate Average Speed for the Segment

The average speed for the given entry flow rate is computed in Step 5-3
using Equation 15-7. The previously computed flow rate vd, slope coefficient m,
and power coefficient p are used in this equation.
𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.41676
800
𝑆 = 56.83 − 3.626 ( − 0.1)
1,000
𝑆 = 53.7 mi/h
Step 5d: Adjust Speed for Horizontal Alignment

Because this is a straight segment, no adjustment to the speed estimate is
required for horizontal alignment.
Assessment of Speed Results

The average speed at a directional flow rate of 800 veh/h is estimated to be
approximately 3 mi/h (about 5%) lower than the FFS, but about 4 mi/h (about
7%) higher than the posted speed limit.
Step 6: Estimate the Percent Followers

The service measure percent followers is estimated in Step 6 for the given flow
rate vd. under the prevailing geometric conditions and percent heavy vehicles.
First, the percent followers at 100% of capacity is estimated. Next, the percent
followers at 25% of capacity is estimated. Then, the slope and power coefficients
for an exponential curve fitting those two points (percent followers at 25% and
100% capacity) are estimated. Finally, the fitted curve is used to estimate the
percent followers for the given flow rate.
Step 6a: Compute Percent Followers at Capacity

Percent followers at capacity is calculated using Equation 15-18, applying
eight parameters b0 to b7 obtained from Exhibit 15-24.
𝑃𝐹𝑐𝑎𝑝 = 𝑏0 + 𝑏1 (𝐿) + 𝑏2 (√𝐿) + 𝑏3 (𝐹𝐹𝑆) + 𝑏4 (√𝐹𝐹𝑆) + 𝑏5 (𝐻𝑉%) +

𝑜𝑣 𝑣
0
𝑏6 (𝐹𝐹𝑆 × 1,000 ) + 𝑏7 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(0.75) − 7.90866(√0.75) − 0.94321(56.83) +

1,500 1,500
13.64266(√56.83) − 0.00050(5) − 0.05500 (56.83 × 1,000) + 7.1376 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 86.41%
Step 6b: Compute Percent Followers at 25% Capacity

Percent followers at 25 percent of capacity is calculated using Equation 15-20,
applying eight parameters c0 to c7 obtained from Exhibit 15-26.
𝑃𝐹25𝑐𝑎𝑝 = 𝑐0 + 𝑐1 (𝐿) + 𝑐2 (√𝐿) + 𝑐3 (𝐹𝐹𝑆) + 𝑐4 (√𝐹𝐹𝑆) + 𝑐5 (𝐻𝑉%)

𝑣𝑜 𝑣0
+ 𝑐6 (𝐹𝐹𝑆 × ) + 𝑐7 (√ )
1,000 1,000
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(0.75) − 21.60000(√0.75) − 0.97853(56.83)

1,500
+ 12.05214(√56.83) − 0.00750(5) − 0.06700 (56.83 × )
1,000
1,500
+ 11.6041 (√ )
1,000
𝑃𝐹25𝑐𝑎𝑝 = 50.52%
Step 6c: Calculate the Slope Coefficient

Equation 15-22 is used to compute the slope coefficient m for an exponential
curve fitted between percent following at capacity and percent following at 25%
capacity. It employs two parameters d1 and d2 obtained from Exhibit 15-28; the
parameters for Passing Constrained segments are used.
𝑃𝐹25𝑐𝑎𝑝 𝑃𝐹𝑐𝑎𝑝
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = 𝑑1 ( 𝑐𝑎𝑝 ) + 𝑑2 ( 𝑐𝑎𝑝 )
0.25 [1,000] [1,000]
50.52 86.41
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( )
1,700 1,700
0.25 [1,000] [1,000]
𝑚 = −1.337
Step 6d: Calculate the Power Coefficient

Equation 15-23 is used to compute the power coefficient p for an exponential
capacity. It employs five parameters e0 through e4 obtained from Exhibit 15-29;
the parameters for Passing Constrained segments are used.
0 − ln [1 − ] 0 − ln [1 − ]
100 100
𝑝 = 𝑒0 + 𝑒1 ( 𝑐𝑎𝑝 ) + 𝑒2 ( 𝑐𝑎𝑝 )
0.25 [ ] [ ]
1,000 1,000
0 − ln (1 − ) 0 − ln (1 − )
100 100
+ 𝑒3 √ 𝑐𝑎𝑝 + 𝑒4
√
𝑐𝑎𝑝
0.25 [ ] [ ]
1,000 1,000
50.52 86.41
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
50.52 86.41
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.7524
Step 6e: Calculate Percent Followers

Equation 15-17 is used to compute percent followers PF.
𝑣𝑑 𝑝
(𝑚 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
800 0.7524
(−1.337 ×{ } )
1,000
𝑃𝐹 = 100 × [1 − 𝑒 ]
𝑃𝐹 = 67.7%
Step 7: Calculate Additional Performance Measure Values for a Passing

Lane Segment
This step is only applicable to passing lane segments. Therefore, Step 7 is
skipped.
Step 8: Calculate Follower Density

Follower density FD is estimated using Equation 15-35.
𝑃𝐹 𝑣𝑑 67.7 800
𝐹𝐷 = × = × = 10.1 followers/mi/ln
100 𝑆 100 53.7
Step 9: Determine Potential Adjustment to Follower Density

There is no passing lane upstream of the analysis segment. Therefore, no
adjustment is needed to follower density and Step 9 is skipped.

The segment’s LOS is determined from Exhibit 15-6, using the column for a
higher-speed highway (posted speed limit equal to or greater than 50 mi/h). With
10.1 followers/mi, the subject direction of travel on the segment operates at
LOS D.
Discussion
The estimated FFS and average speed for a flow rate of 800 veh/h are both
above the posted speed limit. This result is reasonable for a flat, straight segment
in this volume range. However, the follower density produces LOS D operations.
This flow rate is large enough to produce fairly high levels of platooning, but not
so high as to cause significant reductions in speed. The combination of a
moderately high flow rate and moderately high level of platooning will result in
travelers perceiving a relatively poor level of service.
EXAMPLE PROBLEM 2: PASSING CONSTRAINED SEGMENT WITH

HORIZONTAL CURVES
This example problem illustrates the computation of LOS in a single direction
of a curving, 0.75-mi-long Passing Constrained segment in level terrain.
The Facts
The segment to be evaluated has the same general demand and geometric
characteristics as the segment evaluated in Example Problem 1. The difference is
that this segment has horizontal curvature instead of being straight; otherwise,
the same inputs are used as for Example Problem 1.
The segment is split into 11 subsegments, with each subsegment being either
straight (tangent) or curved. Horizontal curvature data for each subsegment is
provided in Exhibit 26-23.
Exhibit 26-23
Example Problem 2:
Horizontal Curve Inputs
Super- Central
Length elevation Radius Angle Horizontal
Subsegment Type (ft)a (%) (ft) (deg) Classb
1 Tangent 280 -- -- -- --
2 Horizontal curve 432 3 450 55 3
3 Tangent 260 -- -- -- --
4 Horizontal curve 366.5 2 300 70 4
5 Tangent 250 -- -- -- --
7 Tangent 275.6 -- -- -- --
9 Tangent 285 -- -- -- --
10 Horizontal curve 767.9 4 1,100 40 1
11 Tangent 369 -- -- -- --
Total 3,960
a
Length for horizontal curves = radius × central angle × π/180.
b
Determined from Exhibit 15-22, with radius and superelevation as inputs.
Objective
Estimate the average speed in the subject direction on the two-lane highway
segment, taking into account the effects of horizontal alignment on the average
speed.

This step was completed in Example Problem 1. The horizontal alignment
does not affect the selection of study boundaries and segmentation.

does not affect the computation of the demand flow rates.
Step 3: Determine the Vertical Alignment Classification

does not affect the determination of the vertical alignment classification.

This step was completed in Example Problem #1. The computed base FFS
applies to the tangent subsegments.

Steps 5a, 5b, and 5c were applied in Example Problem 1 to obtain the
following speed results for the tangent subsegments:
• Base free-flow speed BFFS = 57.0 mi/h,
• Tangent free-flow speed FFS = 56.8 mi/h, and
• Average speed for tangent subsegments S = 53.7 mi/h.

In this step, the average speed for each subsegment with a horizontal curve
is determined. There are three substeps: (a) identifying the horizontal alignment
classification for each subsegment with a horizontal curve, (b) calculating the
average speed for each subsegment with a horizontal curve, and (c) calculating
the adjusted average speed for the segment.
Step 5d.1: Identify all Horizontal Curves Within the Segment

In this step the tangent (straight) length, curve radius, and superelevation are
identified for each horizontal curve within the segment. Each curve is assigned a
horizontal alignment classification on the basis of its radius and percent
superelevation, using Exhibit 15-22. The resulting horizontal classes were
indicated previously in Exhibit 26-23. Note that in typical designs, the crown of
the roadway (designed to shed water from the paved way) will cause the
superelevation to vary by direction of travel. Therefore, a curve’s horizontal class
may also vary by direction of travel.
Step 5d.2: Calculate Average Speed for each Horizontal Curve Within the
Segment
The average speed for a subsegment with horizontal curvature is determined
using Equation 15-12 though Equation 15-15. The process is demonstrated for
Subsegment 2.
Subsegment 2 has a horizontal alignment class of 3 and the BFFS for the
preceding tangent section is 57.0 mi/h. Equation 15-14 is applied to compute the
base free-flow speed for Subsegment 2:
𝐵𝐹𝐹𝑆𝐻𝐶2 = min(𝐵𝐹𝐹𝑆𝑇 , 44.32 + 0.3728 × 𝐵𝐹𝐹𝑆𝑇 − 6.868 × 𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠2 )
𝐵𝐹𝐹𝑆𝐻𝐶2 = min(57.0, 44.32 + 0.3728 × 57.0 − 6.868 × 3)
𝐵𝐹𝐹𝑆𝐻𝐶2 = min(57.0, 44.9656) = 44.9656 mi/h

Subsegment 2’s FFS is computed using Equation 15-13:
𝐹𝐹𝑆𝐻𝐶2 = 𝐵𝐹𝐹𝑆𝐻𝐶2 − 0.0255 × 𝐻𝑉%
𝐹𝐹𝑆𝐻𝐶2 = 44.9656 − 0.0255 × 5
𝐹𝐹𝑆𝐻𝐶2 = 44.8381 𝑚𝑖/ℎ
The slope coefficient m used in the determination of average speed is
computed using Equation 15-15 as follows:
𝑚 = max(0.277, −25.8993 − 0.7756 × 𝐹𝐹𝑆𝐻𝐶2 + 10.6294 × √𝐹𝐹𝑆𝐻𝐶2
+ 2.4766 × 𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠2 − 9.8238 × √𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠2 )
𝑚 = max(0.277, −25.8993 − 0.7756 × 44.8381 + 10.6294 × √44.8381

+ 2.4766 × 3 − 9.8238 × √3)
𝑚 = max(0.277, 0.9145) = 0.9145
Finally, the average speed of Subsegment 2 is computed by Equation 15-12.
𝑣𝑑
𝑆𝐻𝐶2 = min (𝑆, 𝐹𝐹𝑆𝐻𝐶2 − 𝑚 × √ − 0.1)
1,000
800
𝑆𝐻𝐶2 = min (53.7, 44.8381 − 0.9145 × √ − 0.1)
1,000
𝑆𝐻𝐶2 = min(53.7, 44.0730) = 44.1 mi/h

Similar computations are performed for the other subsegments with
horizontal curves. The results are presented in Exhibit 26-24.
Exhibit 26-24 Subsegment BFFSHci (mi/h) FFSHci (mi/h) m SHci (mi/h)

Example Problem 2: 2 44.9656 44.8381 0.9145 44.1
Horizontal Curve Average 4 38.0976 37.9701 0.4081 37.6
Speed Results 6 31.2296 31.1021 0.2770 30.9
8 51.8336 51.7061 1.4905 50.5
10 57.0000 56.8725 2.8036 53.7
Step 5d.3: Calculate Adjusted Average Speed for the Segment

The speed results for all subsegments are summarized in Exhibit 26-25.
Exhibit 26-25 Subsegment Type Speed (mi/h) Length (ft)

Example Problem 2: Average 1 Tangent 53.7 280
Speeds by Subsegment 2 Horizontal curve 44.1 432
3 Tangent 53.7 260
4 Horizontal curve 37.6 366.5
5 Tangent 53.7 250
6 Horizontal curve 30.9 216
7 Tangent 53.7 275.6
8 Horizontal curve 50.5 458
9 Tangent 53.7 285
10 Horizontal curve 53.7 767.9
11 Tangent 53.7 369
Total 3,960
Equation 15-16 is used to calculate the segment’s adjusted average speed by

taking a length-weighted average of the subsegment speeds.
∑11
𝑖=1(𝑆𝑢𝑏𝑠𝑒𝑔𝑆𝑝𝑒𝑒𝑑𝑖 × 𝑆𝑢𝑏𝑠𝑒𝑔𝐿𝑒𝑛𝑔𝑡ℎ𝑖 )
𝑆=
𝐿
(53.7 × 280) + (44.1 × 432) + (53.7 × 260) + (37.6 × 366.5)
[+(53.7 × 250) + (30.9 × 216) + (53.7 × 275.6) + (50.5 × 458)]
+(53.7 × 285) + (53.7 × 767.9) + (53.7 × 369)
𝑆=
3,960
𝑆 = 49.5 mi/h
Discussion
Compared to the straight segment studied in Example Problem 1, the
horizontal curvature in the segment studied in Example Problem 2 reduces the
average speed from 53.7 mi/h to 49.5 mi/h, which is close to the segment’s posted
speed limit of 50 mi/h.
EXAMPLE PROBLEM 3: FACILITY ANALYSIS—LEVEL TERRAIN

This example problem illustrates the computation of LOS in the eastbound
direction of a straight, 5.5-mi-long two-lane highway in level terrain. The study
facility includes a 1.5-mi-long passing lane and a 0.5-mi-long passing zone.
The Facts
The input data for the eastbound direction of the facility are provided in
Exhibit 26-26.
Exhibit 26-26
Example Problem 3:
Input Data
Posted Directional Opposing Peak Heavy

Horizontal Vertical Speed Limit Volume Volume Hour Vehicles
Segment Class Class (mi/h) (veh/h) (veh/h) Factor (%)
1 1 1 55 850 * 0.94 8
2 1 1 55 825 * 0.95 8
3 1 1 55 820 * 0.95 8
4 1 1 55 800 500 0.94 7.5
5 1 1 55 795 * 0.935 8
Note: *Only required for Passing Zone segments.
The facility has the following additional characteristics:

• Facility length = 29,040 ft (5.5 mi);
• No upstream passing lanes;
• Percent grade = 0%;
• Horizontal curvature = none;
• Lane width = 12 ft in all segments;
• Paved shoulder width = 6 ft in all segments; and

• Access points = 0 in all segments (for simplification of this problem,
despite variation in segment volume).
Objective
Estimate the LOS in the subject direction on the two-lane highway segment,
taking into account the effects of the passing lane and the passing zone.

The facility was divided into homogeneous segments following the guidance
given in Step 1. The characteristics considered when segmenting the facility
included the ability to pass, lane geometry, grades, lane and shoulder widths,
posted speed limit, traffic demands, adjacent land uses, and driveways. Each
segment was designated as a Passing Constrained, Passing Zone, or Passing Lane
segment following the guidance on Segmentation given on page 15-4. The
resulting segment lengths and designations were shown in Exhibit 26-26.
Steps 2–9 of the two-lane highway analysis procedure are now followed for
each of the facility’s five segments, starting with the most upstream segment
(Segment 1) and proceeding in sequence to the downstream segments.
Segment 1: Passing Constrained Segment

Equation 15-1 is used to convert the segment’s hourly demand volume to a
peak 15-min flow rate:
𝑉𝑑 850
𝑣𝑑 = = = 904 veh/h
𝑃𝐻𝐹 0.94
description of Step 2 on page 15-18. The demand flow rate is less than capacity;
therefore the calculation process proceeds to Step 3.

According to Exhibit 15-11, a segment with a level grade is assigned a
vertical alignment classification of 1. From Exhibit 15-10, the segment length of
0.75 mi is between the minimum (0.25 mi) and maximum (3.0 mi) lengths for a
Passing Constrained segment of vertical class 1, and therefore no adjustment is
needed to the segment length.

Because Segment 1 is a Passing Constrained segment, the opposing flow rate
vo is set at 1,500 veh/h in Equation 15-4 for the purposes of computing FFS. First,
the base free-flow speed BFFS is estimated using Equation 15-2. Next, Equation
15-4 through Equation 15-6 are used to determine factors relating to lane and
shoulder width, access-point density, and heavy-vehicle percentage, which are
used in the estimation of FFS. Finally, the FFS is estimated by Equation 15-3.
𝐵𝐹𝐹𝑆 = 1.14 × 𝑆𝑝𝑙 = 1.14 × 55 = 62.7 mi/h
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)

𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 6) = 0
𝐴𝑃𝐷 0
𝑓𝐴 = min ( , 10) = min ( , 10) = 0
4 4
𝑎 = max[0.0333, 0] = 0.0333 Coefficients a0 through a5 are
all 0 for vertical class 1 (Exhibit
15-12), so the right side of
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑎(𝐻𝑉%) − f LS − f A Equation 15-6 reduces to 0.
𝐹𝐹𝑆 = 62.7 − (0.0333)(8) − 0 − 0 = 62.43 mi/h

veh/h, the equations given in Step 5 are used to estimate the average speed.

The slope coefficient m is computed using six coefficients b0 to b5, which are
obtained from Exhibit 15-13 for a Passing Constrained segment. For a segment of
vertical class 1, these coefficients are 0.0558, 0.0542, 0.3278, 0.1029, 0, and 0,
respectively. Equation 15-8 is then used to determine the slope coefficient.
𝑣𝑜
1,000
+ max(0, 𝑏4 ) × √𝐻𝑉% ]
1,500
𝑚 = max [0,0.0558 + 0.0542 × 62.43 + 0.3278 × √
1,000
+ max(0, 0.1029) × √0.75 + max(0, 0) × √8 ]
𝑚 = 3.930

The power coefficient p is computed using nine coefficients f0 to f8, which are
obtained from Exhibit 15-19 for a Passing Constrained segment. For a segment of
vertical class 1, all of these coefficients take on values of 0, except for f0 (0.67576),
f3 (0.12060), and f4 (−0.35919). Equation 15-11 is then used to calculate p.
𝑣𝑜 𝑣𝑜
1,000 1,000
+ 𝑓6 × √𝐻𝑉% + 𝑓7 × (𝐿 × 𝐻𝑉%)]
1,500
𝑝 = max [0,0.67576 + 0 × 62.43 + 0 × 0.75 + 0.12060 ×
1,000
1,500
− 0.35919 × √ + 0 × 8 + 0 × √8 + 0 × (0.75 × 8)]
1,000
𝑝 = 0.417

The average speed for Segment 1 is calculated using Equation 15-7.
𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.417
904
𝑆 = 62.43 − 3.930 ( − 0.1)
1,000
𝑆 = 58.8 mi/h

Because Segment 1 is straight, no adjustment to the speed estimate is

Percent followers at capacity for a Passing Constrained segment is calculated
using Equation 15-18, applying eight parameters b0 to b7 obtained from Exhibit
15-24.
𝑣
𝑜 𝑣
0
𝑏6 (𝐹𝐹𝑆 × 1,000 ) + 𝑏7 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(0.75) − 7.90866(√0.75) − 0.94321(62.43) +

1,500 1,500
13.64266(√62.43) − 0.00050(8) − 0.05500 (62.43 × 1,000) + 7.1376 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 85.62%

Percent followers at 25 percent of capacity for a Passing Constrained segment
is calculated using Equation 15-20, applying eight parameters c0 to c7 obtained
from Exhibit 15-26.
𝑣𝑜 𝑣0
+ 𝑐6 (𝐹𝐹𝑆 × ) + 𝑐7 (√ )
1,000 1,000
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(0.75) − 21.60000(√0.75) − 0.97853(62.43)

1,500
+ 12.05214(√62.43) − 0.00750(8) − 0.06700 (62.43 × )
1,000
1,500
+ 11.6041 (√ )
1,000
𝑃𝐹25𝑐𝑎𝑝 = 48.83%

0 − ln [1 − 0 − ln [1 −
100 ] 100 ]
𝑚 = 𝑑1 ( 𝑐𝑎𝑝 ) + 𝑑2 ( 𝑐𝑎𝑝 )
0.25 [1,000] [1,000]
48.83 85.62
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( )
1,700 1,700
0.25 [1,000] [1,000]
𝑚 = −1.289

0 − ln [1 − ] 0 − ln [1 − ]
100 100
𝑝 = 𝑒0 + 𝑒1 ( 𝑐𝑎𝑝 ) + 𝑒2 ( 𝑐𝑎𝑝 )
0.25 [ ] [ ]
1,000 1,000
0 − ln (1 − ) 0 − ln (1 − )
100 100
+ 𝑒3 √ 𝑐𝑎𝑝 + 𝑒4
√
𝑐𝑎𝑝
0.25 [ ] [ ]
1,000 1,000
48.83 85.62
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
48.83 85.62
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.767

𝑣𝑑 𝑝
(𝑚 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
904 0.767
(−1.289 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
𝑃𝐹 = 69.7%
Step 7: Calculate Additional Performance Measure Values for a Passing Lane

Segment
skipped.

𝑃𝐹 𝑣𝑑 69.7 904
100 𝑆 100 58.8

There is no passing lane upstream of Segment 1. Therefore, no adjustment is
needed to follower density and Step 9 is skipped.

The segment’s LOS is determined from Exhibit 15-6, using the column for a
higher-speed highway (posted speed limit equal to or greater than 50 mi/h). With
10.7 followers/mi/ln, Segment 1 operates at LOS D in the analysis direction of
travel.
Segment 2: Passing Lane Segment

𝑉𝑑 825
𝑣𝑑 = = = 868 veh/h
𝑃𝐻𝐹 0.95
From Exhibit 15-5, the capacity of a Passing Lane segment of vertical class 1
with 8% heavy vehicles is 1,500 veh/h. The demand flow rate is less than
capacity; therefore the calculation process proceeds to Step 3.

Passing Lane segment of vertical class 1, and therefore no adjustment is needed
to the segment length.

Because Segment 2 is a Passing Lane segment, the opposing flow rate vo is set
at 0 veh/h in Equation 15-4 for the purposes of computing FFS. Otherwise, the
computations are the same as for Segment 1.
𝐵𝐹𝐹𝑆 = 1.14 × 𝑆𝑝𝑙 = 1.14 × 55 = 62.7 mi/h
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)

𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 6) = 0
𝐴𝑃𝐷 0
𝑓𝐴 = min ( , 10) = min ( , 10) = 0
4 4
𝑎 = max[0.0333, 0] = 0.0333
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑎(𝐻𝑉%) − f LS − f A
𝐹𝐹𝑆 = 62.7 − (0.0333)(8) − 0 − 0 = 62.43 mi/h

The slope coefficient m is computed by Equation 15-8, applying six
coefficients b0 to b5, which are obtained from Exhibit 15-14 for a Passing Lane
segment. This exhibit references Equation 15-9 to calculate b3 and Equation 15-10
to calculate b4, and directly provides numerical values for the other coefficients.
The calculation of the b3 coefficient requires four additional coefficients, c0
through c3, which are obtained from Exhibit 15-16 for Passing Lane segments.
The segment has vertical class 1; therefore c1 takes a value of 0.2667, while the
other coefficients are 0.
𝑏3 = 𝑐0 + 𝑐1 × √𝐿 + 𝑐2 × 𝐹𝐹𝑆 + 𝑐3 × (𝐹𝐹𝑆 × √𝐿)
𝑏3 = 0 + 0.2667 × √1.5 + 0 × 62.43 + 0 × (62.43 × √1.5)
𝑏3 = 0.3266
The calculation of the b4 coefficient requires four additional coefficients, d0
through d3, which are obtained from Exhibit 15-18 for Passing Lane segments.
The segment has vertical class 1; therefore d1 takes a value of 0.1252, while the
other coefficients are 0.
𝑏4 = 𝑑0 + 𝑑1 × √𝐻𝑉% + 𝑑2 × 𝐹𝐹𝑆 + 𝑑3 × (𝐹𝐹𝑆 × √𝐻𝑉%)
𝑏4 = 0 + 0.1252 × √8 + 0 × 62.43 + 0 × (62.43 × √8)
𝑏4 = 0.354
𝑣𝑜
1,000
+ max(0, 𝑏4 ) × √𝐻𝑉% ]
0
𝑚 = max [0, −1.138 + 0.094 × 62.43 + 0.0000 × √
1,000
+ max(0, 0.3266) × √1.5 + max(0, 0.354) × √8 ]
𝑚 = 6.139

The power coefficient p is computed by Equation 15-11, applying nine
coefficients f0 to f8, which are obtained from Exhibit 15-20 for a Passing Lane
segment.
𝑣𝑜 𝑣𝑜
1,000 1,000
+ 𝑓6 × √𝐻𝑉% + 𝑓7 × (𝐿 × 𝐻𝑉%)]
0
𝑝 = max [0, 0.91793 − 0.00557 × 62.43 + 0.36862 × 1.5 + 0 ×
1,000
0
+0×√ + 0.00611 × 8 + 0 × √8 − 0.00419 × (1.5 × 8)]
1,000
𝑝 = 1.122

𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
1.122
868
𝑆 = 62.43 − 6.139 ( − 0.1)
1,000
𝑆 = 57.9 mi/h


Percent followers at capacity for a Passing Lane segment is calculated using
Equation 15-19, applying eight parameters b0 to b7 obtained from Exhibit 15-25.

𝑏6 (√𝐻𝑉%) + 𝑏7 (𝐹𝐹𝑆 × 𝐻𝑉%)
𝑃𝐹𝑐𝑎𝑝 = 61.73075 + 6.73922(1.5) − 23.68853(√1.5) − 0.84126(62.43) +

11.44533(√62.43) − 1.05124(8) + 1.5039(√8) + 0.00491(62.43 × 8)
𝑃𝐹𝑐𝑎𝑝 = 79.04%

Percent followers at 25 percent of capacity for a Passing Lane segment is
calculated using Equation 15-21, applying eight parameters c0 to c7 obtained from
Exhibit 15-27.
+ 𝑐6 (√𝐻𝑉% ) + 𝑐7 (𝐹𝐹𝑆 × 𝐻𝑉%)
𝑃𝐹25𝑐𝑎𝑝 = 80.37105 + 14.44997(1.5) − 46.41831(√1.5) − 0.23367(62.43)

+ 0.84914(√62.43) − 0.56747(8) + 0.89427(√8 )
+ 0.00119(62.43 × 8)
𝑃𝐹25𝑐𝑎𝑝 = 35.90%

parameters for Passing Lane segments are used.
0 − ln [1 − 0 − ln [1 −
100 ] 100 ]
𝑚 = 𝑑1 ( 𝑐𝑎𝑝 ) + 𝑑2 ( 𝑐𝑎𝑝 )
0.25 [1,000] [1,000]
35.90 79.04
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.15808 ( ) − 0.83732 ( )
1,500 1,500
0.25 [1,000] [1,000]
𝑚 = −1.060

the parameters for Passing Lane segments are used.
0 − ln [1 − ] 0 − ln [1 − ]
100 100
𝑝 = 𝑒0 + 𝑒1 ( 𝑐𝑎𝑝 ) + 𝑒2 ( 𝑐𝑎𝑝 )
0.25 [ ] [ ]
1,000 1,000
0 − ln (1 − ) 0 − ln (1 − )
100 100
+ 𝑒3 √ 𝑐𝑎𝑝 + 𝑒4
√
𝑐𝑎𝑝
0.25 [ ] [ ]
1,000 1,000
35.90 79.04
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = −1.63246 + 1.6496 ( 100 ) − 4.45823 ( 100 )
1,500 1,500
0.25 [ ] [ ]
1,000 1,000
35.90 79.04
0 − ln (1 − ) 0 − ln (1 − )
− 4.89119√ 100 + 10.33057√ 100
1,500 1,500
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.897

𝑣𝑑 𝑝
(𝑚 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
868 0.897
(−1.060 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
𝑃𝐹 = 60.7%

Segment
Step 7a: Calculate the Flow Rate in Each Lane of the Passing Lane Segment
Equation 15-24 through Equation 15-27 are applied as follows.
𝐻𝑉% 8
𝑁𝑢𝑚𝐻𝑉 = 𝑣𝑑 × = 868 × = 69 veh
100 100
𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 0.92183 − 0.05022 × ln(𝑣𝑑 ) − 0.00030 × 𝑁𝑢𝑚𝐻𝑉

𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 0.92183 − 0.05022 × ln(868) − 0.00030 × 69
𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 0.561
𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 𝑣𝑑 × 𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 868 × 0.561 = 487 veh/h/ln
𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝑆𝐿 = 𝑣𝑑 × (1 − 𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 ) = 868 × (1 − 0.561) = 381 veh/h/ln
Step 7b: Calculate the Percentage of Heavy Vehicles in Each Lane of the
Passing Lane Segment
Equation 15-28 through Equation 15-30 are applied as follows.
𝐻𝑉%𝐹𝐿 = 𝐻𝑉% × 𝐻𝑉𝑃𝑟𝑜𝑝𝑀𝑢𝑙𝑡𝑖𝑝𝑙𝑖𝑒𝑟𝐹𝐿 = 8 × 0.4 = 3.2%
𝐻𝑉%𝐹𝐿 3.2
𝑁𝑢𝑚𝐻𝑉𝑆𝐿 = 𝑁𝑢𝑚𝐻𝑉 − (𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 × ) = 69 − (487 × ) = 54 veh
100 100
𝑁𝑢𝑚𝐻𝑉𝑆𝐿 54
𝐻𝑉%𝑆𝐿 = × 100 = × 100 = 14.2%
𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝑆𝐿 381
Step 7c: Calculate the Average Speed in Each Lane of the Passing Lane
Segment
Applying the equations and passing lane coefficient tables of Step 5
(Estimate the Average Speed), with the corresponding flow rate and heavy
vehicle percentage for each lane yields:
Sinit_FL = 60.7 mi/h; and
Sinit_SL = 60.6 mi/h.
The average speed lane differential adjustment is calculated with Equation
15-31.
𝐻𝑉%
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 = 2.750 + 0.00056 × 𝑣𝑑 + 3.8521 ×
100
8
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 = 2.750 + 0.00056 × 868 + 3.8521 ×
100
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 = 3.54 mi/h
Next, the average speed for each lane at the passing lane segment midpoint
is calculated with Equation 15-32 and Equation 15-33.
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 3.54
𝑆𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 = 𝑆𝑖𝑛𝑖𝑡_𝐹𝐿 + = 60.7 + = 62.5 mi/h
2 2
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 3.54
𝑆𝑃𝐿𝑚𝑖𝑑_𝑆𝐿 = 𝑆𝑖𝑛𝑖𝑡_𝑆𝐿 − = 60.6 − = 58.8 mi/h
2 2
Step 7d: Calculate the Percent Followers in Each Lane of the Passing Lane
Segment
Applying the equations and passing lane coefficient tables of Step 6
(Estimate the Percent Followers), with the corresponding flow rate and heavy
vehicle percentage for each lane yields
PFPLmid_FL = 44.5%, and
PFPLmid_SL = 35.6%.

The follower density, FD, for the midpoint of the passing lane segment is
estimated using Equation 15-34.
𝑃𝐹𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 𝑃𝐹 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝑆𝐿
( 100 × 𝑆 ) + ( 𝑃𝐿𝑚𝑖𝑑_𝑆𝐿
100 × 𝑆 )
𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 𝑃𝐿𝑚𝑖𝑑_𝑆𝐿
𝐹𝐷𝑃𝐿𝑚𝑖𝑑 =
2
44.5 487 35.6 381
( 100 × 62.5) + ( 100 × 58.8)
2
𝐹𝐷𝑃𝐿𝑚𝑖𝑑 = 2.9 followers/mi/ln
The follower density, FD, for the endpoint of the passing lane segment is
𝑃𝐹 𝑣𝑑 60.7 868
100 𝑆 100 57.9

There is no passing lane upstream of segment 2. Therefore, no adjustment is
needed to follower density for this segment. However, before proceeding to the
analysis of the downstream segments, the effective length of this passing lane
segment is determined. The first criterion for identifying the passing lane effective
length, that is, the point at which the percent improvement to percent followers
goes to zero, is applied by testing different values of DownstreamDistance in
Equation 15-36 until the value of %ImprovePF goes to zero. The resulting effective
length is 14.4 mi.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 𝐷𝑜𝑤𝑛𝑠𝑡𝑟𝑒𝑎𝑚𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒)]
+ 0.1 × max[0, 𝑃𝐹 − 30] + 3.5 × ln[max(0.3, 𝑃𝑎𝑠𝑠𝐿𝑎𝑛𝑒𝐿𝑒𝑛𝑔𝑡ℎ)]
− 0.01 × 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[14.4] + 0.1 × max[0, 69.7 − 30]
+ 3.5 × ln[max(0.3,1.5)] − 0.01 × 904) = 0
The second criterion for identifying the passing lane effective length, that is,
the point at which the downstream adjusted follower density reaches 95% of the
follower density immediately prior to the start of the passing lane, is applied by
testing different values of DownstreamDistance in Equation 15-36 through
Equation 15-38. The target value for FDadj is 95% of the follower density for
segment 1, or 0.95 × 10.7 = 10.17 followers/mi/ln. This target value is reached at a
downstream distance of 8.1 mi.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[8.1] + 0.1 × max[0, 69.7 − 30]
+ 3.5 × ln[max(0.3,1.5)] − 0.01 × 904) = 5.0
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 𝐷𝑜𝑤𝑛𝑠𝑡𝑟𝑒𝑎𝑚𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒

+ 0.1 × max[0, 𝑃𝐹 − 30] + 0.75 × 𝑃𝑎𝑠𝑠𝐿𝑎𝑛𝑒𝐿𝑒𝑛𝑔𝑡ℎ
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 8.1 + 0.1 × max[0, 69.7 − 30] + 0.75 × 1.5
− 0.005 × 904)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, −2.9) = 0
𝑃𝐹 %𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒
𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 %𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆
𝑆 × (1 + )
100
69.7 5.0 904
𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 0
58.8 × (1 + 100)
𝐹𝐷𝑎𝑑𝑗 = 10.18 followers/mi/ln
The passing lane effective length is taken as the shorter of the two values, in
this case 8.1 mi. The remaining downstream segments (3, 4, and 5) are all within
the passing lane’s effective length; therefore, an adjusted follower density will be
calculated for each of these segments.

From Exhibit 15-6, for a higher-speed highway, the LOS is B.

𝑉𝑑 820
𝑣𝑑 = = = 863 veh/h
𝑃𝐻𝐹 0.95
The demand flow rate is less than the capacity of a Passing Constrained
segment (1,700 veh/h); therefore, the calculation process proceeds to Step 3.


Segment 3 is a Passing Constrained segment similar to Segment 1. The only
aspect in which it differs from Segment 1 is its length, which is used in Equation
15-4 to determine the a coefficient. However, the length is multiplied by zero in
Equation 15-4 and therefore does not affect the final result. The calculated FFS is
the same as Segment 1, 62.43 mi/h.

The slope coefficient m is computed using Equation 15-8. The calculation is
the same as for Segment 1, except that the segment length is different.
1,500
𝑚 = max [0,0.0558 + 0.0542 × 62.43 + 0.3278 × √
1,000
+ max(0, 0.1029) × √1 + max(0, 0) × √8 ] = 3.944

The power coefficient p is computed using Equation 15-11. The calculation is
the same as for Segment 1, except that the segment length is different.
1,500
𝑝 = max [0,0.67576 + 0 × 62.43 + 0 × 1.00 + 0.12060 ×
1,000
1,500
− 0.35919 × √ + 0 × 8 + 0 × √8 + 0 × (1.00 × 8)]
1,000
𝑝 = 0.417

𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.417
863
𝑆 = 62.43 − 3.944 ( − 0.1)
1,000
𝑆 = 58.9 mi/h

Because segment 3 is straight, no adjustment to the speed estimate is

using Equation 15-18. The calculation is the same as for Segment 1, except that
the segment length is different.
𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(1.00) − 7.90866(√1.00) − 0.94321(62.43) +
1,500 1,500
13.64266(√62.43) − 0.00050(8) − 0.05500 (62.43 × ) + 7.1376 (√1,000)
1,000
𝑃𝐹𝑐𝑎𝑝 = 85.32%

is calculated using Equation 15-20. The calculation is the same as for Segment 1,
except that the segment length is different.
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(1.00) − 21.60000(√1.00) − 0.97853(62.43)
1,500
+ 12.05214(√62.43) − 0.00750(8) − 0.06700 (62.43 × )
1,000
1,500
+ 11.6041 (√ ) = 48.43%
1,000

Equation 15-22 is used to compute the slope coefficient m.
48.43 85.32
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑚 = −1.275

Equation 15-23 is used to compute the power coefficient p.
48.43 85.32
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
48.43 85.32
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.768

863 0.768
(−1.275 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ] = 68.0%

Segment
skipped.

68.0 863
𝐹𝐷 = × = 10.0 followers/mi/ln
100 58.9

As previously determined, this segment is within the effective length of the
upstream passing lane; therefore, an adjusted follower density value is
calculated. The downstream distance used in these calculations is 2.5 mi (passing
lane segment length + subject segment length).
Equation 15-36 is used to determine the percentage improvement to percent
followers.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 2.5)] + 0.1 × max[0, 69.7 − 30]
+ 3.5 × ln[max(0.3,1.5)] − 0.01 × 863)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = 15.7%
Equation 15-37 is used to determine the percentage improvement to average
speed.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 2.5 + 0.1 × max[0, 69.7 − 30] + 0.75 × 1.5
− 0.005 × 863)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = 1.8%
Equation 15-38 is used to determine the adjusted follower density.

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
𝑆 × (1 + 100 )
68.0 15.7 863

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 1.8
58.9 × (1 +
100)
This adjusted follower density result is used in Steps 10 and 11.

From Exhibit 15-6, for a higher-speed highway, the LOS is D (barely).
Segment 4: Passing Zone Segment

𝑉𝑑 800
𝑣𝑑 = = = 851 veh/h
𝑃𝐻𝐹 0.94
The demand flow rate is less than the capacity of a Passing Zone segment
(1,700 veh/h) and the calculation process can proceed to Step 3. However,
because Segment 4 is a Passing Zone segment, the demand flow rate in the
opposing direction is also required and is calculated as follows:
𝑉𝑜 500
𝑣𝑜 = = = 532 veh/h
𝑃𝐻𝐹 0.94

Passing Zone segment of vertical class 1, and therefore no adjustment is needed
to the segment length.

The determination of the FFS for Segment 4 is similar to that of the previous
segments, as they have similar geometric characteristics. It differs only in terms
of length (which does not play a role in determining FFS for level segments) and
percent heavy vehicles. Therefore, from Equation 15-3:
𝐹𝐹𝑆 = 62.7 − (0.0333)(7.5) − 0 − 0 = 62.45 mi/h

The slope coefficient m is computed using Equation 15-8. Passing Zone
segments use the same coefficients in this equation as do Passing Constrained
segments. However, the values for FFS, opposing flow rate, segment length, and
percent heavy vehicles are different for Segment 4.
532
𝑚 = max [0,0.0558 + 0.0542 × 62.45 + 0.3278 × √
1,000
+ max(0, 0.1029) × √0.50 + max(0, 0) × √7.5 ] = 3.752

The power coefficient p is computed using Equation 15-11.
532
𝑝 = max [0,0.67576 + 0 × 62.45 + 0 × 0.5 + 0.12060 ×
1,000
532
− 0.35919 × √ + 0 × 7.5 + 0 × √7.5 + 0 × (0.5 × 7.5)]
1,000
𝑝 = 0.478

𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.478
851
𝑆 = 62.45 − 3.752 ( − 0.1)
1,000
𝑆 = 59.2 mi/h

Because segment 4 is straight, no adjustment to the speed estimate is

Percent followers at capacity for a Passing Zone segment is calculated using
Equation 15-18. The coefficients used in the equation are the same ones used for
Passing Constrained segments.
𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(0.50) − 7.90866(√0.50) − 0.94321(62.45) +
532 532
13.64266(√62.45) − 0.00050(7.5) − 0.05500 (62.45 × ) + 7.1376 (√1,000)
1,000
𝑃𝐹𝑐𝑎𝑝 = 85.90%

is calculated using Equation 15-20.
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(0.50) − 21.60000(√0.50) − 0.97853(62.45)

+ 12.05214(√62.45) − 0.00750(7.5)
532 532
− 0.06700 (62.45 × ) + 11.6041 (√ ) = 48.06%
1,000 1,000

48.06 85.90
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( ) = −1.287
1,700 1,700
0.25 [ ] [ ]
1,000 1,000

48.06 85.90
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
48.06 85.90
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.791

851 0.791
(−1.287 ×{ } )
1,000
𝑃𝐹 = 100 × [1 − 𝑒 ] = 67.8%

Segment
skipped.

67.8 851
100 59.2

calculated. The downstream distance used in these calculations is the sum of the
lengths of Segments 2, 3, and 4, which is 3.0 mi.
followers.

%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 3.0)] + 0.1 × max[0, 69.7 − 30]
+ 3.5 × ln[max(0.3,1.5)] − 0.01 × 851)
speed.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 3.0 + 0.1 × max[0, 69.7 − 30] + 0.75 × 1.5
− 0.005 × 851)

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
𝑆 × (1 + 100 )
67.8 14.3 851

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 1.4
59.2 × (1 + 100)

From Exhibit 15-6, for a higher-speed highway, the LOS is D (barely).

Step 2: Determine Demand Flow Rates, Capacity and d/c Ratio
𝑉𝑑 795
𝑣𝑑 = = = 850 veh/h
𝑃𝐻𝐹 0.935
The demand flow rate is less than the capacity of a Passing Constrained
segment (1,700 veh/h); therefore, the calculation process proceeds to Step 3.


Segment 5 is a Passing Constrained segment similar to Segments 1 and 3. The
only aspect in which it differs from those segments is its length, which is used in
Equation 15-4 to determine the a coefficient, where it is multiplied by zero and
therefore does not affect the final result. The calculated FFS is therefore the same
as Segments 1 and 3, 62.43 mi/h.

The slope coefficient m is computed using Equation 15-8.
1,500
𝑚 = max [0,0.0558 + 0.0542 × 62.43 + 0.3278 × √
1,000
+ max(0, 0.1029) × √1.75 + max(0, 0) × √8 ] = 3.977

The power coefficient p is computed using Equation 15-11.
1,500
𝑝 = max [0,0.67576 + 0 × 62.43 + 0 × 1.75 + 0.12060 ×
1,000
1,500
− 0.35919 × √ + 0 × 8 + 0 × √8 + 0 × (1.75 × 8)]
1,000
𝑝 = 0.417

0.417
850
𝑆 = 62.43 − 3.977 ( − 0.1)
1,000
𝑆 = 58.9 mi/h


𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(1.75) − 7.90866(√1.75) − 0.94321(62.43) +
1,500 1,500
13.64266(√62.43) − 0.00050(8) − 0.05500 (62.43 × 1,000) + 7.1376 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 85.05%

is calculated using Equation 15-20.
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(1.75) − 21.60000(√1.75) − 0.97853(62.43)
1,500
+ 12.05214(√62.43) − 0.00750(8) − 0.06700 (62.43 × )
1,000
1,500
+ 11.6041 (√ ) = 48.96%
1,000

48.96 85.05
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( )
1,700 1,700
0.25 [1,000] [1,000]
𝑚 = −1.275

48.96 85.05
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
48.96 85.05
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.750

850 0.750
(−1.275 ×{ } )
1,000
𝑃𝐹 = 100 × [1 − 𝑒 ] = 67.7%

Segment
skipped.

67.7 850
100 58.9

calculated. The downstream distance used in these calculations is the sum of the
lengths of Segments 2–5, 4.75 mi.
followers.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 4.75)]
+ 0.1 × max[0, 69.7 − 30] + 3.5 × ln[max(0.3,1.5)]
− 0.01 × 850)
speed.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 4.75 + 0.1 × max[0, 69.7 − 30] + 0.75 × 1.5
− 0.005 × 850)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = 0%

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
𝑆 × (1 + )
100
67.7 10.2 850
𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 0
58.9 × (1 + 100)

From Exhibit 15-6, for a higher-speed highway, the LOS is D.
Step 11: Facility Analysis

The average follower density and the average LOS for the facility are computed
by averaging the length-weighted segment densities using Equation 15-39.
∑5𝑖=1 𝐹𝐷𝑖 × 𝐿𝑖
𝐹𝐷𝐹 =
∑5𝑖=1 𝐿𝑖
10.7 × 0.75 + 2.9 × 1.5 + 8.2 × 1.0 + 8.2 × 0.5 + 8.8 × 1.75
𝐹𝐷𝐹 =
0.75 + 1.5 + 1.0 + 0.5 + 1.75
40.1
𝐹𝐷𝐹 = = 7.3 followers/mi/ln
5.5
From Exhibit 15-6, for a higher-speed highway, the facility LOS is C. Exhibit
26-27 summarizes LOS results for each segment and the facility as a whole.
Length Follower Density Exhibit 26-27

Segment (mi) Type (followers/mi) LOS Example Problem 3:
1 0.75 Passing Constrained 10.7 D LOS Results
2 1.50 Passing Lane 2.9 B
3 1.00 Passing Constrained 8.2 D
4 0.50 Passing Zone 8.2 D
5 1.75 Passing Constrained 8.8 D
5.50 Facility 7.3 C
Discussion
The non–passing lane segments all operate at LOS D and the passing lane
segment operates at LOS B. The operating conditions within the passing lane
segment and its benefits several miles downstream result in an overall facility
LOS of C.
EXAMPLE PROBLEM 4: FACILITY ANALYSIS – MOUNTAIN ROAD

This example problem illustrates the computation of the LOS in the eastbound
direction of a 5.1-mi-long two-lane highway in mountainous terrain with two 0.5-
mi-long passing lanes. Grades reach 6% and there are a series of reverse curves
that constrain speeds.
This highway is a popular summer and winter recreational area access route.
It experiences heavy snowfall each winter. The road is extensively plowed
during winter and chains are required during winter storms. The Chapter 15
method is not appropriate for the analysis of winter operations.
This analysis focuses on summer, dry weather operations. Summer volumes
are higher than winter volumes.
The Facts
A diagram of the study facility, showing its six segments, is provided in
Exhibit 26-28. The study direction (eastbound) starts in the top left of the figure
and proceeds to the bottom center.
Exhibit 26-28
Example Problem 4:
Facility Diagram
Exhibit 26-29 provides volume and speed data for the facility, while Exhibit
26-30 provides grade and horizontal curve data. Tangent sections are either
straight or have horizontal curves with radii greater than 2,550 ft. Segment 2 has
a series of seven reverse curves of similar radius and central angle. To reduce
repetitive computations that would produce the same estimated speed, they have
been combined into a single long, curved subsegment. Segment 3 contains a short
passing lane 910 ft (0.17 mi) in length. Because this length is shorter than the
minimum passing lane lengths given in Exhibit 15-10 (i.e., is too short to be
effectively used as a passing lane), the passing lane is ignored. Segment 3 is treated
as a Passing Constrained segment instead, following the guidance on page 15-17.
Exhibit 26-29 Posted Directional Peak Heavy

Example Problem 4: Length Speed Limit Volume Hour Vehicles
Facility Volume and Segment Segment Type (mi) (mi/h) (veh/h) Factor (%)
Speed Data 1 Passing Constrained 1.3 55 1,100 0.90 8
2 Passing Constrained 1.0 55 1,100 0.90 8
5 Passing Lane 0.5 55 1,100 0.90 8
Exhibit 26-30 Horizontal Curvature

Example Problem 4: Segment Super-
Facility Grade and Horizontal Length Grade Sub- Horizontal Length elevation Curve
Curve Data Segment (mi) (%) segment Alignment (ft) (%) Radius (ft)
a Tangent 5,964 — —
1 1.3 4
b Curve 900 2 350
2 1.0 6
b 7 curves 4,280 2 500
3 0.5 6 Tangent 2,640 — —
4 1.3 4
b Curve 3,000 2 850
5 0.5 −3 Tangent 2,640 — —
6 0.5 −3 Tangent 2,640 — —
The facility has the following additional characteristics:

• Facility length = 26,928 ft (5.1 mi);
• Not within the effective length of upstream passing lanes;
• No turnouts;
• Lane width = 12 ft in all segments;
• Paved shoulder width = 6 ft in all segments; and
• Access points = 0 in all segments.
Objective
Estimate the LOS in the eastbound direction of the two-lane highway facility,
taking into account the effects of the passing lanes.

The facility was divided into homogeneous segments following the guidance
given in Step 1. The characteristics considered when segmenting the facility
included the ability to pass, lane geometry, grades, lane and shoulder widths,
posted speed limit, traffic demands, adjacent land uses, and driveways. Each
segment was designated as a Passing Constrained, Passing Zone, or Passing Lane
segment following the guidance on Segmentation given on page 15-4. The
resulting segment lengths and designations were shown in Exhibit 26-29.
Steps 2–10 of the two-lane highway analysis procedure are now followed for
each of the facility’s five segments, starting with the most upstream segment
(Segment 1) and proceeding in sequence to the downstream segments.

Equation 15-1 is used to convert the facility’s hourly demand volume to a
peak 15-min flow rate. Each segment has the same demand volume and thus the
same flow rate.
𝑉𝑑 1,100
𝑣𝑑 = = = 1,222 veh/h
𝑃𝐻𝐹 0.90
description of Step 2 on page 15-18, while the capacity of a Passing Lane segment
with vertical class 1 and 8% heavy vehicles is 1,500 veh/h, from Exhibit 15-5. The
demand flow rate is less than these capacities; therefore the calculation process
proceeds to Step 3.

Each segment is assigned a vertical alignment classification on the basis of
Exhibit 15-11. The results are shown in Exhibit 26-31.
Exhibit 26-31 Segment Length (mi) Grade (%) Vertical Class

Example Problem 4: 1 1.3 4 4
Vertical Alignment 2 1.0 6 5
Classifications by Segment 3 0.5 6 4
4 1.3 4 4
5 0.5 −3 1
6 0.5 −3 1
From Exhibit 15-10, all segment lengths lie between the minimum and
maximum lengths for their respective segment types. Therefore, no adjustment is
needed to any segment length.

The FFS is computed using Equation 15-2 through Equation 15-6. The
process will be demonstrated for Segment 1 and the results presented for the
remaining segments.
Because Segment 1 is a Passing Constrained segment, the opposing flow rate
vo is set at 1,500 veh/h in Equation 15-4 for the purposes of computing FFS. First,
the base free-flow speed BFFS is estimated using Equation 15-2. Next, Equation
15-5 and Equation 15-6 are used to determine factors relating to lane and shoulder
width and access-point density, which are used in the estimation of FFS.
𝐵𝐹𝐹𝑆 = 1.14 × 𝑆𝑝𝑙 = 1.14 × 55 = 62.7 mi/h
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)

𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 6) = 0
𝐴𝑃𝐷 0
𝑓𝐴 = min ( , 10) = min ( , 10) = 0
4 4
Unlike the previous example problems, Segment 1 is not level and therefore
the value of the coefficient a does not reduce to 0.0333. Equation 15-4 is used to
calculate a. This equation uses six coefficients a0 to a5, which are obtained from
Exhibit 15-12.
𝑎 = max [0.0333, 𝑎0 + 𝑎1 × 𝐵𝐹𝐹𝑆 + 𝑎2 × 𝐿

𝑣𝑜
+ max(0, 𝑎3 + 𝑎4 × 𝐵𝐹𝐹𝑆 + 𝑎5 × 𝐿) × ]
1,000
𝑎 = max [0.0333, −0.40902 + 0.00975 × 62.7 + 0.00767 × 1.3
1,500
+ max(0, −0.18363 + 0.00423 × 62.7 + 0 × 1.3) × ]
1,000
𝑎 = 0.335
Finally, the FFS is estimated by Equation 15-3.
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑎(𝐻𝑉%) − f LS − f A

𝐹𝐹𝑆 = 62.7 − (0.335)(8) − 0 − 0 = 60.0 mi/h
Exhibit 26-32 summarizes the FFS results for all segments.
Segment Exhibit 26-32

Variable 1 2 3 4 5 6 Example Problem 4:
BFFS (mi/h) 62.7 62.7 62.7 62.7 62.7 62.7 Free-Flow Speed Results
fLS (mi/h) 0 0 0 0 0 0
fA (mi/h) 0 0 0 0 0 0
a0 −0.40902 −0.38360 −0.40902 −0.40902 0.00000 0.00000
a1 0.00975 0.01074 0.00975 0.00975 0.00000 0.00000
a2 0.00767 0.01945 0.00767 0.00767 0.00000 0.00000
a3 −0.18363 −0.69848 −0.18363 −0.18363 0.00000 0.00000
a4 0.00423 0.01069 0.00423 0.00423 0.00000 0.00000
a5 0.00000 0.12700 0.00000 0.00000 0.00000 0.00000
a 0.335 0.457 0.329 0.335 0.0333 0.0333
FFS (mi/h) 60.0 59.0 60.1 60.0 62.4 62.4

veh/h, the equations given in Step 5 are used to estimate the average speed. The
process is demonstrated for Segment 1, with results summarized for all segments
afterwards.

The slope coefficient m is computed using Equation 15-8. This equation
requires six coefficients b0 to b5, which are obtained from Exhibit 15-13 for a
Passing Constrained segment. The exhibit references Equation 15-9 to calculate b3
and Equation 15-10 to calculate b4, and directly provides numerical values for the
other coefficients.
Equation 15-9 is used to determine the segment length coefficient b3. It uses
four coefficients c0 to c3 , which are obtained from Exhibit 15-15 for a Passing
Constrained segment.
𝑏3 = 𝑐0 + 𝑐1 × √𝐿 + 𝑐2 × 𝐹𝐹𝑆 + 𝑐3 × (𝐹𝐹𝑆 × √𝐿)
𝑏3 = −12.5113 + 0 × √1.3 + 0.2656 × 60.0 + 0 × (60 × √1.3)
𝑏3 = 3.4247
Equation 15-10 is used to determine the segment length coefficient b3. It uses
four coefficients d0 to d3, which are obtained from Exhibit 15-17 for a Passing
Constrained segment.
𝑏4 = 𝑑0 + 𝑑1 × √𝐻𝑉% + 𝑑2 × 𝐹𝐹𝑆 + 𝑑3 × (𝐹𝐹𝑆 × √𝐻𝑉%)
𝑏4 = −5.7775 + 0 × √8 + 0.1373 × 60.0 + 0 × (60.0 × √8)
𝑏4 = 2.4605
With all the coefficients now determined, the slope coefficient m can be
calculated:
𝑣𝑜
1,000
+ max(0, 𝑏4 ) × √𝐻𝑉% ]
1,500
𝑚 = max [3.2685,9.0115 − 0.1994 × 60.0 + 1.8252 × √
1,000
+ max(0,3.4247) × √1.3 + max(0,2.4605) × √8 ]
𝑚 = 10.147

The power coefficient p is computed using Equation 15-11. This equation
requires nine coefficients f0 to f8, which are obtained from Exhibit 15-19 for a
Passing Constrained segment.
𝑣𝑜 𝑣𝑜
1,000 1,000
+ 𝑓6 × √𝐻𝑉% + 𝑓7 × (𝐿 × 𝐻𝑉%)]
𝑝 = max [0.33950,0.67689 + 0.00534 × 60.0 − 0.13037 × 1.3
1,500 1,500
+ 0.25699 × − 0.68465 × √ − 0.00709 × 8
1,000 1,000
+ 0.07087 × √8 + 0 × (1.3 × 8)]
𝑝 = 0.519

The average speed is calculated using Equation 15-7.
𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.519
1,222
𝑆 = 60.0 − 10.147 ( − 0.1)
1,000
𝑆 = 49.2 mi/h
Exhibit 26-33 summarizes the average speed results for all segments.

b0 9.0115 23.9144 9.0115 9.0115 −1.1379 0.0558 Unadjusted Average Speed
b1 −0.1994 −0.6925 −0.1994 −0.1994 0.0941 0.0542 Results
b2 1.8252 1.9473 1.8252 1.8252 0.0000 0.3278
c0 −12.5113 −14.8961 −12.5113 −12.5113 0.0000 0.1029
c1 0.0000 0.0000 0.0000 0.0000 0.2667 0.0000
c2 0.2656 0.4370 0.2656 0.2656 0.0000 0.0000
c3 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
b3 3.4247 10.8869 3.4513 3.4247 0.1886 0.1029
d0 −5.7775 −18.2910 −5.7775 −5.7775 0.0000 0.0000
d1 0.0000 2.3875 0.0000 0.0000 0.1252 0.0000
d2 0.1373 0.4494 0.1373 0.1373 0.0000 0.0000
d3 0.0000 −0.0520 0.0000 0.0000 0.0000 0.0000
b4 2.4605 6.2989 2.4742 2.4605 0.3541 0.0000
b5 3.2685 3.5115 3.2685 3.2685 0.0000 0.0000
m 10.147 14.145 8.702 10.147 5.869 3.912
f0 0.67689 1.13262 0.67689 0.67689 0.91793 0.67576
f1 0.00534 0.00000 0.00534 0.00534 −0.00557 0.00000
f2 −0.13037 −0.26367 −0.13037 −0.13037 0.36862 0.00000
f3 0.25699 0.18811 0.25699 0.25699 0.00000 0.12060
f4 −0.68465 −0.64304 −0.68465 −0.68465 0.00000 −0.35919
f5 −0.00709 −0.00867 −0.00709 −0.00709 0.00611 0.00000
f6 0.07087 0.08675 0.07087 0.07087 0.00000 0.00000
f7 0.00000 0.00000 0.00000 0.00000 −0.00419 0.00000
f8 0.33950 0.30590 0.33950 0.33950 0.00000 0.00000
p 0.519 0.540 0.623 0.519 0.787 0.417
S (mi/h) 49.2 43.9 50.8 49.2 56.0 58.3

In this step, the average speed for each subsegment with a horizontal curve
is determined. There are three substeps: (a) identifying the horizontal alignment
classification for each subsegment with a horizontal curve, (b) calculating the
average speed for each subsegment with a horizontal curve, and (c) calculating
the adjusted average speed for the segment. The process will be demonstrated
for Segment 1, with results summarized for the other segments afterwards. Only
Segments 1, 2, and 4 contain horizontal curves.
Step 5d.1: Identify all Horizontal Curves Within the Segment

From the facts given previously, the horizontal curve in subsegment 1b has a
radius of 350 ft and a superelevation of 2%. From Exhibit 15-22, it is assigned a
horizontal alignment class of 4.
Step 5d.2: Calculate Average Speed for each Horizontal Curve Within the
Segment
The average speed for a subsegment with horizontal curvature is determined
using Equation 15-12 though Equation 15-15. First, Equation 15-14 is applied to
compute the base free-flow speed:
𝐵𝐹𝐹𝑆𝐻𝐶1𝑏 = min(𝐵𝐹𝐹𝑆𝑇 , 44.32 + 0.3728 × 𝐵𝐹𝐹𝑆𝑇 − 6.868 × 𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠1𝑏 )
𝐵𝐹𝐹𝑆𝐻𝐶1𝑏 = min(62.7,44.32 + 0.3728 × 62.7 − 6.868 × 4)
𝐵𝐹𝐹𝑆𝐻𝐶1𝑏 = min(62.7, 40.22) = 40.22 mi/h
Next, the FFS is computed using Equation 15-13:
𝐹𝐹𝑆𝐻𝐶1𝑏 = 𝐵𝐹𝐹𝑆𝐻𝐶1𝑏 − 0.0255 × 𝐻𝑉%
𝐹𝐹𝑆𝐻𝐶1𝑏 = 40.22 − 0.0255 × 8

𝐹𝐹𝑆𝐻𝐶1𝑏 = 40.02 mi/h
The slope coefficient m used in the determination of average speed is
computed using Equation 15-15 as follows:
𝑚 = max(0.277, −25.8993 − 0.7756 × 𝐹𝐹𝑆𝐻𝐶1𝑏 + 10.6294 × √𝐹𝐹𝑆𝐻𝐶1𝑏
+ 2.4766 × 𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠1𝑏 − 9.8238 × √𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠1𝑏 )
𝑚 = max(0.277, −25.8993 − 0.7756 × 40.02 + 10.6294 × √40.02 + 2.4766 × 4

− 9.8238 × √4)
𝑚 = max(0.277, 0.563) = 0.563
Finally, the average speed of subsegment 1b is computed by Equation 15-12.
𝑣𝑑
𝑆𝐻𝐶1𝑏 = min (𝑆, 𝐹𝐹𝑆𝐻𝐶1𝑏 − 𝑚 × √ − 0.1)
1,000
1,222
𝑆𝐻𝐶1𝑏 = min (49.2, 40.02 − 0.563 × √ − 0.1)
1,000
𝑆𝐻𝐶1𝑏 = min(49.2, 39.4) = 39.4 mi/h
Step 5d.3: Calculate Adjusted Average Speed for the Segment

Equation 15-16 is used to calculate Segment 1’s adjusted average speed by
taking a length-weighted average of the subsegment speeds.
∑2𝑖=1(𝑆𝑢𝑏𝑠𝑒𝑔𝑆𝑝𝑒𝑒𝑑𝑖 × 𝑆𝑢𝑏𝑠𝑒𝑔𝐿𝑒𝑛𝑔𝑡ℎ𝑖 )
𝑆=
𝐿
(49.2 × 5,964) + (39.4 × 900)
𝑆=
6,864
𝑆 = 47.9 mi/h
Exhibit 26-34 presents the calculation results for adjusted average speed for
each segment.
Exhibit 26-34 Segment

Example Problem 4: Variable 1 2 3 4 5 6
Adjusted Average Speed Horizontal class 4 3 — 2 — —
Results BFFSHC (mi/h) 40.22 47.09 — 53.96 — —
FFSHC (mi/h) 40.0 46.9 — 53.8 — —
m 0.563 0.933 — 1.401 — —
SHC (mi/h) 39.4 43.9 — 49.2 — —
Adjusted S (mi/h) 47.9 43.9 50.8 49.2 56.0 58.3
Note: — = not applicable, no horizontal curve in the segment.

The calculation of percent followers is demonstrated for Segment 1, with
results for all segments presented afterwards.

using Equation 15-18, applying eight parameters b0 to b7 obtained from Exhibit
15-24.
𝑜𝑣 𝑣0
𝑏6 (𝐹𝐹𝑆 × 1,000 ) + 𝑏7 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 58.29978 − 0.53611(1.3) + 7.35076(√1.3) − 0.27046(60.0) +

1,500 1,500
4.49850(√60.0) − 0.01100(8) − 0.02968 (60.0 × ) + 8.8968 (√1,000)
1,000
𝑃𝐹𝑐𝑎𝑝 = 92.74%

is calculated using Equation 15-20, applying eight parameters c0 to c7 obtained
from Exhibit 15-26.
𝑣𝑜 𝑣0
+ 𝑐6 (𝐹𝐹𝑆 × ) + 𝑐7 (√ )
1,000 1,000
𝑃𝐹25𝑐𝑎𝑝 = 103.13534 + 14.68459(1.3) − 23.72704(√1.3) + 0.664436(60.0)

1,500
− 11.95763(√60.0) − 0.10000(8) + 0.00172 (60.0 × )
1,000
1,500
+ 14.7007 (√ )
1,000
𝑃𝐹25𝑐𝑎𝑝 = 59.77%

0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = 𝑑1 ( 𝑐𝑎𝑝 ) + 𝑑2 ( 𝑐𝑎𝑝 )
0.25 [1,000] [1,000]
59.77 92.74
0 − ln [1 − ] 0 − ln [1 − ]
𝑚 = −0.29764 ( 100 ) − 0.71917 ( 100 )
1,700 1,700
0.25 [1,000] [1,000]
𝑚 = −1.747

0 − ln [1 − ] 0 − ln [1 − ]
100 100
𝑝 = 𝑒0 + 𝑒1 ( 𝑐𝑎𝑝 ) + 𝑒2 ( 𝑐𝑎𝑝 )
0.25 [ ] [ ]
1,000 1,000
0 − ln (1 − ) 0 − ln (1 − )
100 100
+ 𝑒3 √ 𝑐𝑎𝑝 + 𝑒4
√
𝑐𝑎𝑝
0.25 [ ] [ ]
1,000 1,000
59.77 92.74
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
59.77 92.74
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.762

𝑣𝑑 𝑝
(𝑚 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
1,222 0.762
(−1.747 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
𝑃𝐹 = 86.9%
Exhibit 26-35 presents the percent followers results for all segments.
Step 7: Calculate Additional Performance Measure Values for a Passing

Lane Segment
This step applies only to Segment 5, the passing lane segment. Equation 15-24
through Equation 15-33 are applied, as was demonstrated for the passing lane
segment in Example Problem 3. The results that will be used in the segment
midpoint follower density calculation are as follows:
FlowRateFL = 654 veh/h/ln;
FlowRateSL = 568 veh/h/ln;

SPLmid_FL = 61.1 mi/h;
SPLmid_SL = 56.8 mi/h;
PFPLmid_FL = 63.1%; and
PFPLmid_SL = 55.9%.

Follower density, FD, is estimated using Equation 15-35. For Segment 1, the
calculation is performed as follows:
𝑃𝐹 𝑣𝑑 86.9 1,222
100 𝑆 100 47.9
The follower density, FD, for the midpoint of the passing lane segment is
𝑃𝐹𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 𝑃𝐹 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝑆𝐿
( × ) + ( 𝑃𝐿𝑚𝑖𝑑_𝑆𝐿 × )
100 𝑆𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 100 𝑆𝑃𝐿𝑚𝑖𝑑_𝑆𝐿
2
63.1 654 55.9 568
( 100 × 61.1) + ( 100 × )
𝐹𝐷𝑃𝐿𝑚𝑖𝑑 = 56.8
2
𝐹𝐷𝑃𝐿𝑚𝑖𝑑 = 6.2 followers/mi/ln
Exhibit 26-35 presents the follower density results for all segments.

b0 58.29978 3.32968 58.29978 58.29978 61.73075 37.68080 Percent Follower and
b1 −0.53611 −0.84377 −0.53611 −0.53611 6.73922 3.05089 Unadjusted Follower Density
b2 7.35076 7.08952 7.35076 7.35076 −23.68853 −7.90866 Results
b3 −0.27046 −1.32089 −0.27046 −0.27046 −0.84126 −0.94321
b4 4.49850 19.98477 4.49850 4.49850 11.44533 13.64266
b5 −0.01100 −0.01250 −0.01100 −0.01100 −1.05124 −0.00050
b6 −0.02968 −0.02960 −0.02968 −0.02968 1.50390 −0.05500
b7 8.89680 9.99450 8.89680 8.89680 0.00491 7.13760
PFcap (%) 92.74 94.67 89.98 92.74 84.56 86.12
c0 103.13534 89.00000 103.13534 103.13534 80.37105 18.01780
c1 14.68459 19.02642 14.68459 14.68459 14.44997 10.00000
c2 −23.72704 −34.54240 −23.72704 −23.72704 −46.41831 −21.60000
c3 0.66444 0.29792 0.66444 0.66444 −0.23367 −0.97853
c4 −11.95763 −6.62528 −11.95763 −11.95763 0.84914 12.05214
c5 −0.10000 −0.16000 −0.10000 −0.10000 −0.56747 −0.00750
c6 0.00172 0.00480 0.00172 0.00172 0.89427 −0.06700
c7 14.70074 17.56610 14.70074 14.70074 0.00119 11.60410
PF25cap (%) 59.77 60.83 58.29 59.77 45.48 49.77
d1 −0.29764 −0.29764 −0.29764 −0.29764 −0.15808 −0.29764
d2 −0.71917 −0.71917 −0.71917 −0.71917 −0.83732 −0.71917
m −1.747 −1.897 −1.586 −1.747 −1.299 −1.317
e0 0.81165 0.81165 0.81165 0.81165 −1.63246 0.81165
e1 0.37920 0.37920 0.37920 0.37920 1.64960 0.37920
e2 −0.49524 −0.49524 −0.49524 −0.49524 −4.45823 −0.49524
e3 −2.11289 −2.11289 −2.11289 −2.11289 −4.89119 −2.11289
e4 2.41146 2.41146 2.41146 2.41146 10.33057 2.41146
p 0.762 0.823 0.696 0.762 0.791 0.760
PF (%) 86.9 89.3 83.9 86.9 78.2 78.5
FD (followers/mi) 22.2 24.9 20.2 21.6 17.1 16.5

Segment 5 is a passing lane. Therefore, an adjustment may be needed to
follower density in the downstream segment, Segment 6. Following the same
process outlined in Example Problem 3, the effective length is determined to be
4.4 mi (the adjusted follower density within 95% of the upstream follower
density controls the effective length in this case).
Because the end of the downstream, and last, segment is within the effective
length of the passing lane, Equation 15-36 through Equation 15-38 are used to
calculate the adjusted follower density for Segment 6. The downstream distance
used in these calculations is 1.0 mi (Segment 5 length + Segment 6 length).
followers in Segment 6 as a result of the upstream passing lane.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 1.0)] + 0.1 × max[0,86.9 − 30]
+ 3.5 × ln[max(0.3,0.5)] − 0.01 × 1,222)
speed in Segment 6 as a result of the upstream passing lane, and Equation 15-38
is used to determine the adjusted follower density.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 1.0 + 0.1 × max[0, 86.9 − 30] + 0.75 × 0.5
− 0.005 × 1,222)
𝐹𝑜𝑙𝑙𝑜𝑤𝑒𝑟𝐷𝑒𝑛𝑠𝑖𝑡𝑦𝑎𝑑𝑗 = × (1 − )×
𝑆 × (1 + 100 )
78.5 18.0 1,222
𝐹𝑜𝑙𝑙𝑜𝑤𝑒𝑟𝐷𝑒𝑛𝑠𝑖𝑡𝑦𝑎𝑑𝑗 = × (1 − )×
100 100 2.2
58.3 × (1 +
100)
𝐹𝑜𝑙𝑙𝑜𝑤𝑒𝑟𝐷𝑒𝑛𝑠𝑖𝑡𝑦𝑎𝑑𝑗 = 13.2 followers/mi

Each segment’s LOS is determined from Exhibit 15-6, using the column for a
higher-speed highway (posted speed limit equal to or greater than 50 mi/h). The
follower density in all segments exceeds 12 followers/mi; therefore, all segments
operate at LOS E.
Step 11: Facility Analysis

The average follower density and the average LOS for the facility are computed
by averaging the length-weighted segment densities using Equation 15-39.
∑6𝑖=1 𝐹𝐷𝑖 × 𝐿𝑖
𝐹𝐷𝐹 =
∑6𝑖=1 𝐿𝑖
22.2 × 1.3 + 24.9 × 1.0 + 20.2 × 0.5 + 21.6 × 1.3 + 6.2 × 0.5 + 13.2 × 0.5
𝐹𝐷𝐹 =
1.3 + 1.0 + 0.5 + 1.3 + 0.5 + 0.5
𝐹𝐷𝐹 = 20.0 followers/mi/ln

From Exhibit 15-6, for a higher-speed highway, the facility LOS is E. Exhibit
26-36 summarizes LOS results for each segment and the facility as a whole.
Length Follower Density Exhibit 26-36

Segment (mi) Type (followers/mi) LOS Example Problem 4:
1 1.3 Passing Constrained 22.2 E LOS Results
2 1.0 Passing Constrained 24.9 E
5 0.5 Passing Lane 6.2 C
5.1 Facility 20.0 E
Discussion
The main conclusions of this analysis are:
• The tight 350-ft radius horizontal curve at the end of Segment 1
significantly reduces speeds at the end of that segment. The other
segments are comparatively unaffected by their horizontal curvature.
• The 6% upgrade significantly affects speeds and percent followers. The
Passing Lane segment significantly reduces percent followers on the
downstream segment. However, the improvement is not large enough to
change the LOS from E to D for this mountainous highway.
• The percent followers and the follower density are high on this facility,
resulting in LOS E. The last segment of the facility (a downgrade
following a 0.5-mi passing lane) is slightly over the threshold for LOS E.
Average speeds for the non-passing lane segments range from 44 mi/h to
56 mi/h. Within the passing lane, the average speeds range from 56 to 61
mi/h across the two lanes. The demand/capacity ratio varies from 0.72 to
0.87 during the summer peak hour.
The long upgrade extending across several segments generates interactions
across the segments that are not well modeled by this macroscopic analysis
method for evaluating facilities. Consequently, microsimulation would be
recommended to verify and potentially refine the results.
As noted in The Facts section at the start of this example problem, Segment 3
contains a 910-ft passing lane that is too short to provide a substantial
operational benefit. One could analyze the effect of extending the passing lane
length to 0.5 mi, assuming it is actually feasible to extend the passing lane in this
mountainous terrain. In that case, Segment 3 would be analyzed as a Passing
Lane segment, and the follower density in Segment 4 would be adjusted to reflect
the effects of the passing lane. Another passing lane starts in Segment 5;
therefore, the analysis of the effects of the Segment 3 passing lane would not be
carried past Segment 4. With the improved passing lane, the follower density for
the facility would improve to 17.6 followers/mi/ln, although this still yields LOS E.
EXAMPLE PROBLEM 5: TWO-LANE HIGHWAY BICYCLE LOS

A segment of two-lane highway (without passing lanes) is being evaluated
for potential widening, realigning, and repaving. Analyze the impacts of the
proposed project on the bicycle LOS (BLOS) in the peak direction.
The Facts
The roadway currently has the following characteristics:
• Lane width = 12 ft,
• Shoulder width = 2 ft,
• Pavement rating = 3 (fair),
• Posted speed limit = 50 mi/h,
• Hourly directional volume = 500 veh/h (no growth is expected),
• Percentage of heavy vehicles = 5%,
• PHF = 0.90, and
• No on-highway parking.
The proposed roadway design has the following characteristics:
• Lane width = 12 ft,
• Shoulder width = 6 ft,
• Pavement rating = 5 (very good),
• Posted speed limit = 55 mi/h, and
• No on-highway parking.
Step 1: Gather Input Data

All data needed to perform the analysis are listed above.
Step 2: Calculate the Directional Flow Rate in the Outside Lane

Using the hourly directional volume and the PHF, calculate the directional
demand flow rate with Equation 15-40. Because this is a two-lane highway
segment without a passing lane, the number of directional lanes N is 1. Because
traffic volumes are not expected to grow over the period of the analysis, vOL is the
same for both current and future conditions.
𝑉 500
𝑣𝑂𝐿 = = = 556 veh/h
𝑃𝐻𝐹 × 𝑁 0.90 × 1
Step 3: Calculate the Effective Width

For current conditions, the hourly directional demand V is greater than 160
veh/h and the paved shoulder width is 2 ft; therefore, Equation 15-43 and
Equation 15-44 are used to determine the effective width of the outside lane.
Under future conditions, the paved shoulder width will increase to 6 ft;
therefore, Equation 15-42 and Equation 15-44 are used.
For current conditions,
𝑊𝑣 = 𝑊𝑂𝐿 + 𝑊𝑠 = 12 + 2 = 14 ft
𝑊𝑒 = 𝑊𝑣 − (%𝑂𝐻𝑃[2 ft + 𝑊𝑠 ]) = 14 − (0 × [2 + 2]) = 14 ft
Under the proposed design,
𝑊𝑣 = 𝑊𝑂𝐿 + 𝑊𝑠 = 12 + 6 = 18 ft
𝑊𝑒 = 𝑊𝑣 + 𝑊𝑠 − 2 × (%𝑂𝐻𝑃[2 ft + 𝑊𝑠 ]) = 18 + 6 − 2 × (0 × [2 + 6]) = 24 ft
Step 4: Calculate the Effective Speed Factor

Equation 15-46 is used to calculate the effective speed factor. Under current
conditions,
𝑆𝑡 = 1.1199 ln(𝑆𝑝𝑙 − 20) + 0.8103 = 1.1199 ln (50 − 20) + 0.8103 = 4.62
Under the proposed design,
𝑆𝑡 = 1.1199 ln (55 − 20) + 0.8103 = 4.79
Step 5: Determine the LOS

Equation 15-47 is used to calculate the BLOS score, which is then used in
Exhibit 15-7 to determine the LOS. Under existing conditions,
𝐵𝐿𝑂𝑆 = 0.507 ln(𝑣𝑂𝐿 ) + 0.1999𝑆𝑡 (1 + 10.38𝐻𝑉)2 + 7.066(1/𝑃)2 − 0.005(𝑊𝑒 )2
+ 0.760
𝐵𝐿𝑂𝑆 = 0.507 ln(556) + 0.1999(4.62)(1 + 10.38 × 0.05)2 + 7.066(1/3)2
− 0.005(14)2 + 0.760
𝐵𝐿𝑂𝑆 = 3.205 + 2.131 + 0.785 − 0.980 + 0.760 = 5.90
Therefore, the BLOS for existing conditions is LOS F. Use of the same process
for the proposed design results in the following:
𝐵𝐿𝑂𝑆 = 0.507 ln(556) + 0.1999(4.79)(1 + 10.38 × 0.05)2 + 7.066(1/5)2
− 0.005(24)2 + 0.760
𝐵𝐿𝑂𝑆 = 3.205 + 2.209 + 0.283 − 2.880 + 0.760 = 3.58
The corresponding LOS for the proposed design is LOS D, close to the
boundary of LOS C (BLOS = 3.50).
Discussion
Although the posted speed would increase as a result of the proposed
design, this negative impact on bicyclists would be more than offset by the
proposed shoulder widening, as indicated by the improvement from LOS F to
LOS D.
9. REFERENCES
Some of these references can 1. Zegeer, J. D., M. A. Vandehey, M. Blogg, K. Nguyen, and M. Ereti. NCHRP
be found in the Technical
Reference Library in Volume 4. Report 599: Default Values for Highway Capacity and Level of Service Analyses.
Transportation Research Board of the National Academies, Washington,
D.C., 2008.
2. Dowling, R., G. F. List, B. Yang, E. Witzke, and A. Flannery. NCFRP Report
41: Incorporating Truck Analysis into the Highway Capacity Manual.
D.C., 2014.
3. Washburn, S. S., and S. Ozkul. Heavy Vehicle Effects on Florida Freeways and
Multilane Highways. Report TRC-FDOT-93817-2013. Florida Department of
Transportation, Tallahassee, 2013.
4. Ozkul, S., and Washburn, S. S. Updated Commercial Truck Speed versus
Distance-Grade Curves for the Highway Capacity Manual. In Transportation
Research Record: Journal of the Transportation Research Board, No. 2483,
D.C., 2015, pp. 91–101.
5. Hu, J., B. Schroeder, and N. Rouphail. Rationale for Incorporating Queue
Discharge Flow into Highway Capacity Manual Procedure for Analysis of
Freeway Facilities. In Transportation Research Record: Journal of the
Transportation Research Board, No. 2286, Transportation Research Board of the
National Academies, Washington, D.C., 2012, pp. 76–83.
6. Elefteriadou, L., A. Kondyli, and B. St. George. Estimation of Capacities on
Florida Freeways. Final Report. Transportation Research Center, University of
Florida, Gainesville, Sept. 2014.
7. Brilon, W., J. Geistefeldt, and M. Regler. Reliability of Freeway Traffic Flow:
A Stochastic Concept of Capacity. In Proceedings of the 16th International
Symposium on Transportation and Traffic Theory, College Park, Md., July 2005,
pp. 125–144.
8. SAE International. Taxonomy and Definitions for Terms Related to Driving
Automation Systems for On-Road Motor Vehicles. Recommended Practice J3016.
Warrendale, Pa., June 2018.
9. Adebisi, A., Y. Liu, B. Schroeder, J. Ma, B. Cesme, A. Jia, and A. Morgan.
Developing Highway Capacity Manual Capacity Adjustment Factors for
Connected and Automated Traffic on Freeway Segments. In Transportation
Research Record: Journal of the Transportation Research Board, No. 2674,
D.C., 2020, pp. 401–415.
10. Jones, S. Cooperative Adaptive Cruise Control: Human Factors Analysis. Report
FHWA-HRT-13-045. Federal Highway Administration, Washington, D.C.,
Oct. 2013.
References Chapter 26/Freeway and Highway Segments: Supplemental

11. Krechmer, D., K. Blizzard, M.G. Cheung, R. Campbell, V. Alexiadis, J. Hyde,

J. Osborne, M. Jensen, S. Row, A. Tudela, E. Flanigan, and J. Bitner. Connected
Vehicle Impacts on Transportation Planning. Primer and Final Report. Report
FHWA-JPO-16-420. Federal Highway Administration, Washington, D.C.,
June 2016.
12. Nowakowski, C., J. O’Connell, S.E. Shladover, and D. Cody. Cooperative
Adaptive Cruise Control: Driver Acceptance of Following Gap Settings Less
than One Second. Proceedings, Human Factors and Ergonomics Society
Annual Meeting, San Francisco, Calif., 2010.
13. Goñi-Ros, B., W.J. Schakel, A.E. Papacharalampous, M. Wang, V.L. Knoop, I.
Sakata, B. van Arem, and S.P. Hoogendoorn. Using Advanced Adaptive
Cruise Control Systems to Reduce Congestion at Sags: An Evaluation Based
on Microscopic Traffic Simulation. Transportation Research Part C: Emerging
Technologies, Vol. 102, 2019, pp. 411–426.
14. Davis, S.C., and R.G. Boundy. Transportation Energy Data Book, Edition 37.
Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge,
Tenn., Aug. 2019.
15. Litman, T. Autonomous Vehicle Implementation Predictions: Implications for
Transport Planning. Victoria Transport Policy Institute, Victoria, B.C., Oct.
2019.
Chapter 26/Freeway and Highway Segments: Supplemental References

APPENDIX A: TRUCK PERFORMANCE CURVES
This appendix provides travel time versus distance curves for SUTs and TTs
for 50-, 55-, 60-, 65-, and 75-mi/h free-flow speeds (FFS). Curves for SUTs and TTs
for a 70-mi/h FFS are presented in Section 3 as Exhibit 26-5 and Exhibit 26-6,
respectively.
Exhibit 26-A1
FFS
Notes: Curves in this graph assume a weight-to-horsepower ratio of 100.

Triangles indicate where a truck reaches 55 mi/h, circles indicate 60 mi/h, diamonds indicate 65 mi/h, and
squares indicate 70 mi/h.
Exhibit 26-A2
FFS

Circles indicate where a truck reaches 60 mi/h, diamonds indicate 65 mi/h, and squares indicate 70 mi/h.
Appendix A: Truck Performance Curves Chapter 26/Freeway and Highway Segments: Supplemental
Exhibit 26-A3
FFS

Diamonds indicate where a truck reaches 65 mi/h and squares indicate 70 mi/h.
Exhibit 26-A4
FFS

Squares indicate where a truck reaches 70 mi/h.
Chapter 26/Freeway and Highway Segments: Supplemental Appendix A: Truck Performance Curves
Exhibit 26-A5
FFS
Exhibit 26-A6
FFS

Triangles indicate where a truck reaches 55 mi/h, circles indicate 60 mi/h, diamonds indicate 65 mi/h, and
squares indicate 70 mi/h.
Exhibit 26-A7
FFS

Circles indicate where a truck reaches 60 mi/h, diamonds indicate 65 mi/h, and squares indicate 70 mi/h.
Exhibit 26-A8
FFS

Diamonds indicate where a truck reaches 65 mi/h and squares indicate 70 mi/h.
Chapter 26/Freeway and Highway Segments: Supplemental Appendix A: Truck Performance Curves
Exhibit 26-A9
FFS

Squares indicate where a truck reaches 70 mi/h.
Exhibit 26-A10
FFS
APPENDIX B: WORK ZONES ON TWO-LANE HIGHWAYS
This appendix presents a method for estimating the capacity and operation
of work zones on two-lane highways when one of the two lanes is closed. This
method is based on research conducted by National Cooperative Highway
Research Program (NCHRP) Project 03-107 (B-1).
Work zones along two-lane highways can take three forms:
1. Shoulder closure. Work activity is limited to the shoulder of one direction
of travel and does not require lane reconfiguration. In this case, only the
direction of travel adjacent to the work zone is slightly affected.
2. Lane shift. Work activity extends beyond the shoulder, but both directions
of travel can be accommodated with a lane shift that utilizes the opposite
paved shoulder.
3. Lane closure. Work activity requires the closure of one of the two lanes.
Flaggers or temporary traffic signals are used to alternately serve one
direction of travel at a time. Both directions of travel can be significantly
affected.
The method presented in this appendix addresses the third scenario—lane This method addresses a one-
lane closure on a two-lane
closure—as it has the greatest impact on traffic operations. highway. Other types of work
zones, such as shoulder
closures or lane shifts, are not
CONCEPTS addressed.
A lane closure on a two-lane highway converts traffic flow from an
uninterrupted to an interrupted condition. With traffic control devices (flaggers
or signals) provided at each end, the operation of the lane closure can be
described in terms similar to those used for a signalized intersection:
• Capacity is the number of vehicles that can be processed through the work
zone per cycle or per hour. It can be determined based on the saturation
flow rate at the control points and the traffic control “cycle length.”
• Cycle length is determined by the flagging operations or signal timing at
each control point and the time required to travel through the work zone.
Travel time is dependent on the average travel speed of the platoons
traveling through the work zone. Factors that may influence travel speed
include posted speed limit, use of a pilot car, heavy-vehicle percentage,
grade, intensity of construction activity, lane width, lateral distance to the
work activity, and lighting conditions (day versus night).
Performance measures, including delay and queue length, can be calculated
by using capacity and cycle length.
WORK ZONE CAPACITY

The methodology for estimating the capacity of a work zone on a two-lane The work zone capacity
methodology is analogous to
highway with one lane closed is analogous to the capacity calculation for a two- the capacity calculation for a
phase signalized intersection. Average travel speed is estimated from a regression two-phase traffic signal.
model developed through observations of two directions of travel at three work

zones (B-1).
Chapter 26/Freeway and Highway Segments: Supplemental Appendix B: Work Zones on Two-Lane Highways
Step 1: Collect Data

For a typical capacity calculation, the analyst must specify traffic information
(including traffic demands, travel speed, and heavy-vehicle percentage), roadway
geometric configuration (e.g., lane width, lateral clearance, speed limit), and
work zone data (including work zone length, signal green time, and traffic
control plan).
A basic traffic flagger control process for a two-lane highway work zone
involving a lane closure is shown in Exhibit 26-B1. Direction 1 refers to the travel
direction whose lane is blocked by the work zone; Direction 2 refers to the travel
direction with the open lane.
Exhibit 26-B1
Traffic Control for a Two-Lane
Highway Work Zone Involving
a Lane Closure
Source: Schoen et al. (B-1).
Some data, such as average travel speed, saturation flow rate, and green
interval length, may be difficult to collect in the field. In Steps 2–4, the
mathematical models that can be used to estimate these data are presented.
Analysts must note that, for capacity calculations, field data are always more
desirable to use when available.
Measuring two-lane highway A procedure is given in Section 6 of Chapter 31, Signalized Intersections:
work zone saturation flow rates
requires a longer data Supplemental, for determining the saturation flow rate of a signalized
collection time than for a intersection. This procedure involves counting and timing the number of queue
signalized intersection because
of the longer cycle lengths discharge vehicles that pass through an intersection to determine the saturated
involved. vehicle headway. As two-lane highway work zone traffic control typically has a
much longer cycle length than a typical signalized intersection, the time period
for gathering saturation flow data is recommended to be 30–60 min. Of course, a
longer time period is generally more desirable when possible. The work zone
capacity can then be determined from the measured saturation flow rate and the
effective green–to–cycle length ratio.
Unlike the core two-lane highway procedure described in Chapter 15, the
work zone procedure requires that demand volumes be adjusted for the effects of
heavy vehicles and grades, using Equation 26-B1.
Appendix B: Work Zones on Two-Lane Highways Chapter 26/Freeway and Highway Segments: Supplemental
𝑉𝑖 Equation 26-B1
𝑣𝑖 =
𝑃𝐻𝐹 × 𝑓𝑔 × 𝑓𝐻𝑉
where
vi = demand flow rate (pc/h);
i = “d” (analysis direction) or “o” (opposing direction);
Vi = demand volume for direction i (veh/h);
PHF = peak hour factor (decimal);
fg = grade adjustment factor, from Exhibit 26-B2 or Exhibit 26-B3; and
fHV = heavy vehicle adjustment factor, from Equation 26-B2.
Grade Adjustment Factor Calculation

Exhibit 26-B2 shows grade adjustment factors for extended segments of level
and rolling terrain, as well as for specific downgrades. Exhibit 26-B2 is entered
with the one-direction demand flow rate vvph, in vehicles per hour. If demand is
expressed as an hourly volume, it must be divided by the PHF (vvph = V/PHF) to
obtain the appropriate factor. Other adjustment factor tables associated with
Equation 26-B1 are entered with this value as well.
One-Direction Adjustment Factor Exhibit 26-B2

Demand Flow Rate, vvph Level Terrain and Two-Lane Highway Work
(veh/h) Specific Downgrades Rolling Terrain Zone Grade Adjustment
≤100 1.00 0.67 Factor (fg) for Level Terrain,
200 1.00 0.75 Rolling Terrain, and Specific
300 1.00 0.83 Downgrades
400 1.00 0.90
500 1.00 0.95
600 1.00 0.97
700 1.00 0.98
800 1.00 0.99
≥900 1.00 1.00
Note: Interpolation to the nearest 0.01 is recommended.
Exhibit 26-B3 shows grade adjustment factors for specific upgrades. The
negative impact of upgrades on two-lane highway speeds increases as both the
severity of the upgrade and its length increase. The impact declines as demand
flow rate increases. At higher demand flow rates, lower speeds would already
result, and the additional impact of the upgrades is less severe.
Heavy Vehicle Adjustment Factor Calculation

Determining the heavy vehicle adjustment factor is a two-step process:
1. Passenger car equivalents are found for trucks (ET) and recreational
vehicles (RVs) (ER) under prevailing conditions.
2. A heavy vehicle adjustment factor is computed from the passenger car
equivalents with Equation 26-B2.
Exhibit 26-B3 Grade

Two-Lane Highway Work Directional Demand Flow Rate, vvph (veh/h)
Grade Length
Zone Grade Adjustment (%) (mi) 100 200 300 400 500 600 700 800 900
Factor (fg) for Specific 0.25 0.78 0.84 0.87 0.91 1.00 1.00 1.00 1.00 1.00
Upgrades 0.50 0.75 0.83 0.86 0.90 1.00 1.00 1.00 1.00 1.00
0.75 0.73 0.81 0.85 0.89 1.00 1.00 1.00 1.00 1.00
1.00 0.73 0.79 0.83 0.88 1.00 1.00 1.00 1.00 1.00
3 <3.5
1.50 0.73 0.79 0.83 0.87 0.99 0.99 1.00 1.00 1.00
2.00 0.73 0.79 0.82 0.86 0.98 0.98 0.99 1.00 1.00
3.00 0.73 0.78 0.82 0.85 0.95 0.96 0.96 0.97 0.98
4.00 0.73 0.78 0.81 0.85 0.94 0.94 0.95 0.95 0.96
0.25 0.75 0.83 0.86 0.90 1.00 1.00 1.00 1.00 1.00
0.50 0.72 0.80 0.84 0.88 1.00 1.00 1.00 1.00 1.00
0.75 0.67 0.77 0.81 0.86 1.00 1.00 1.00 1.00 1.00
1.00 0.65 0.73 0.77 0.81 0.94 0.95 0.97 1.00 1.00
3.5 <4.5
1.50 0.63 0.72 0.76 0.80 0.93 0.95 0.96 1.00 1.00
2.00 0.62 0.70 0.74 0.79 0.93 0.94 0.96 1.00 1.00
3.00 0.61 0.69 0.74 0.78 0.92 0.93 0.94 0.98 1.00
4.00 0.61 0.69 0.73 0.78 0.91 0.91 0.92 0.96 1.00
0.25 0.71 0.79 0.83 0.88 1.00 1.00 1.00 1.00 1.00
0.50 0.60 0.70 0.74 0.79 0.94 0.95 0.97 1.00 1.00
0.75 0.55 0.65 0.70 0.75 0.91 0.93 0.95 1.00 1.00
1.00 0.54 0.64 0.69 0.74 0.91 0.93 0.95 1.00 1.00
4.5 <5.5
1.50 0.52 0.62 0.67 0.72 0.88 0.90 0.93 1.00 1.00
2.00 0.51 0.61 0.66 0.71 0.87 0.89 0.92 0.99 1.00
3.00 0.51 0.61 0.65 0.70 0.86 0.88 0.91 0.98 0.99
4.00 0.51 0.60 0.65 0.69 0.84 0.86 0.88 0.95 0.97
0.25 0.57 0.68 0.72 0.77 0.93 0.94 0.96 1.00 1.00
0.50 0.52 0.62 0.66 0.71 0.87 0.90 0.92 1.00 1.00
0.75 0.49 0.57 0.62 0.68 0.85 0.88 0.90 1.00 1.00
1.00 0.46 0.56 0.60 0.65 0.82 0.85 0.88 1.00 1.00
5.5 <6.5
1.50 0.44 0.54 0.59 0.64 0.81 0.84 0.87 0.98 1.00
2.00 0.43 0.53 0.58 0.63 0.81 0.83 0.86 0.97 0.99
3.00 0.41 0.51 0.56 0.61 0.79 0.82 0.85 0.97 0.99
4.00 0.40 0.50 0.55 0.61 0.79 0.82 0.85 0.97 0.99
0.25 0.54 0.64 0.68 0.73 0.88 0.90 0.92 1.00 1.00
0.50 0.43 0.53 0.57 0.62 0.79 0.82 0.85 0.98 1.00
0.75 0.39 0.49 0.54 0.59 0.77 0.80 0.83 0.96 1.00
1.00 0.37 0.45 0.50 0.54 0.74 0.77 0.81 0.96 1.00
6.5
1.50 0.35 0.45 0.49 0.54 0.71 0.75 0.79 0.96 1.00
2.00 0.34 0.44 0.48 0.53 0.71 0.74 0.78 0.94 0.99
3.00 0.34 0.44 0.48 0.53 0.70 0.73 0.77 0.93 0.98
4.00 0.33 0.43 0.47 0.52 0.70 0.73 0.77 0.91 0.95
Note: Straight-line interpolation of fg,ATS for length of grade and demand flow permitted to the nearest 0.01.
1
Equation 26-B2 𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1) + 𝑃𝑅 (𝐸𝑅 − 1)
where
fHV = heavy vehicle adjustment factor,
PT = proportion of trucks in the traffic stream (decimal),
PR = proportion of RVs in the traffic stream (decimal),
ET = passenger car equivalent for trucks, and
ER = passenger car equivalent for RVs.
The passenger car equivalent is the number of passenger cars displaced from
the traffic stream by one truck or RV. Passenger car equivalents are defined for:
• Extended sections of general level or rolling terrain,
• Specific upgrades, and
• Specific downgrades.
Exhibit 26-B4 contains passenger car equivalents for trucks and RVs in
general terrain segments and for specific downgrades, which are treated as level
terrain in the lower-speed situations associated with work zones.
Exhibit 26-B4
Directional Demand Level Terrain and
Two-Lane Highway Work
Vehicle Type Flow Rate, vvph (veh/h) Specific Downgrades Rolling Terrain
Zone Passenger Car
≤100 1.9 2.7 Equivalents for Trucks (ET)
200 1.5 2.3
and RVs (ER) for Level
300 1.4 2.1 Terrain, Rolling Terrain, and
400 1.3 2.0
Specific Downgrades
Trucks, ET 500 1.2 1.8
600 1.1 1.7
700 1.1 1.6
800 1.1 1.4
≥900 1.0 1.3
RVs, ER All flows 1.0 1.1
Note: Interpolation to the nearest 0.1 is recommended.
Exhibit 26-B5 and Exhibit 26-B6 show passenger car equivalents for trucks
and RVs, respectively, on specific upgrades.
Step 2: Estimate Average Travel Speed

A simple estimation of average travel speed is obtained as follows. Speeds
for Directions 1 and 2 are calculated by Equation 26-B3 and Equation 26-B4,
respectively. Research on two-lane highway work zones (B-1) found that
Direction 2 (i.e., the direction whose lane is not closed) consistently had higher
average speeds than Direction 1.
𝑆1 = 0.615 × 𝑆𝑝𝑙 − 𝑓𝐿𝑆 − 𝑓𝐴 − 2.4 Equation 26-B3
𝑆2 = 0.692 × 𝑆𝑝𝑙 − 𝑓𝐿𝑆 − 𝑓𝐴 − 2.4 Equation 26-B4
where
Si = average travel speed in direction i (mi/h),
Spl = posted speed limit for the two-lane highway segment (mi/h),
fLS = adjustment for lane and shoulder width from Equation 15-5 (mi/h),
and
fA = adjustment for access-point density from Equation 15-6 (mi/h).
For two-lane highway work zones, these equations provide a constant speed
reduction of 2.4 mi/h in all conditions.
Exhibit 26-B5 Grade

Grade Length
Zone Passenger Car (%) (mi) 100 200 300 400 500 600 700 800 900
Equivalents for Trucks (ET) on 0.25 2.6 2.4 2.3 2.2 1.8 1.8 1.7 1.3 1.1
Specific Upgrades 0.50 3.7 3.4 3.3 3.2 2.7 2.6 2.6 1.9 1.6
0.75 4.6 4.4 4.3 4.2 3.7 3.6 3.4 2.4 1.9
3, 1.00 5.2 5.0 4.9 4.9 4.4 4.2 4.1 3.0 2.3
<3.5 1.50 6.2 6.0 5.9 5.8 5.3 5.0 4.8 3.6 2.9
2.00 7.3 6.9 6.7 6.5 5.7 5.5 5.3 4.1 3.5
3.00 8.4 8.0 7.7 7.5 6.5 6.2 6.0 4.6 3.9
4.00 9.4 8.8 8.6 8.3 7.2 6.9 6.6 4.8 3.7
0.25 3.8 3.4 3.2 3.0 2.3 2.2 2.2 1.7 1.5
0.50 5.5 5.3 5.1 5.0 4.4 4.2 4.0 2.8 2.2
0.75 6.5 6.4 6.5 6.5 6.3 5.9 5.6 3.6 2.6
3.5, 1.00 7.9 7.6 7.4 7.3 6.7 6.6 6.4 5.3 4.7
<4.5 1.50 9.6 9.2 9.0 8.9 8.1 7.9 7.7 6.5 5.9
2.00 10.3 10.1 10.0 9.9 9.4 9.1 8.9 7.4 6.7
3.00 11.4 11.3 11.2 11.2 10.7 10.3 10.0 8.0 7.0
4.00 12.4 12.2 12.2 12.1 11.5 11.2 10.8 8.6 7.5
0.25 4.4 4.0 3.7 3.5 2.7 2.7 2.7 2.6 2.5
0.50 6.0 6.0 6.0 6.0 5.9 5.7 5.6 4.6 4.2
0.75 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5
4.5, 1.00 9.2 9.2 9.1 9.1 9.0 9.0 9.0 8.9 8.8
<5.5 1.50 10.6 10.6 10.6 10.6 10.5 10.4 10.4 10.2 10.1
2.00 11.8 11.8 11.8 11.8 11.6 11.6 11.5 11.1 10.9
3.00 13.7 13.7 13.6 13.6 13.3 13.1 13.0 11.9 11.3
4.00 15.3 15.3 15.2 15.2 14.6 14.2 13.8 11.3 10.0
0.25 4.8 4.6 4.5 4.4 4.0 3.9 3.8 3.2 2.9
0.50 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2
0.75 9.1 9.1 9.1 9.1 9.1 9.1 9.1 9.1 9.1
5.5, 1.00 10.3 10.3 10.3 10.3 10.3 10.3 10.3 10.2 10.1
<6.5 1.50 11.9 11.9 11.9 11.9 11.8 11.8 11.8 11.7 11.6
2.00 12.8 12.8 12.8 12.8 12.7 12.7 12.7 12.6 12.5
3.00 14.4 14.4 14.4 14.4 14.3 14.3 14.3 14.2 14.1
4.00 15.4 15.4 15.3 15.3 15.2 15.1 15.1 14.9 14.8
0.25 5.1 5.1 5.0 5.0 4.8 4.7 4.7 4.5 4.4
0.50 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8
0.75 9.8 9.8 9.8 9.8 9.8 9.8 9.8 9.8 9.8
1.00 10.4 10.4 10.4 10.4 10.4 10.4 10.4 10.3 10.2
6.5
1.50 12.0 12.0 12.0 12.0 11.9 11.9 11.9 11.8 11.7
2.00 12.9 12.9 12.9 12.9 12.8 12.8 12.8 12.7 12.6
3.00 14.5 14.5 14.5 14.5 14.4 14.4 14.4 14.3 14.2
4.00 15.4 15.4 15.4 15.4 15.3 15.3 15.3 15.2 15.1
Note: Interpolation for length of grade and demand flow rate to the nearest 0.1 is recommended.
Exhibit 26-B6
Grade Grade
Zone Passenger Car (%) Length (mi) 100 200 300 400 500 600 700 800 900
Equivalents for RVs (ER) on 0.25 1.1 1.1 1.1 1.0 1.0 1.0 1.0 1.0 1.0
Specific Upgrades >0.25, 0.75 1.2 1.2 1.1 1.1 1.0 1.0 1.0 1.0 1.0
3,
>0.75, 1.25 1.3 1.2 1.2 1.1 1.0 1.0 1.0 1.0 1.0
<3.5
>1.25, 2.25 1.4 1.3 1.2 1.1 1.0 1.0 1.0 1.0 1.0
>2.25 1.5 1.4 1.3 1.2 1.0 1.0 1.0 1.0 1.0
0.75 1.3 1.2 1.2 1.1 1.0 1.0 1.0 1.0 1.0
3.5,
>0.75, 3.50 1.4 1.3 1.2 1.1 1.0 1.0 1.0 1.0 1.0
<4.5
>3.50 1.5 1.4 1.3 1.2 1.0 1.0 1.0 1.0 1.0
4.5, 2.50 1.5 1.4 1.3 1.2 1.0 1.0 1.0 1.0 1.0
<5.5 >2.50 1.6 1.5 1.4 1.2 1.0 1.0 1.0 1.0 1.0
0.75 1.5 1.4 1.3 1.1 1.0 1.0 1.0 1.0 1.0
5.5, >0.75, 2.50 1.6 1.5 1.4 1.2 1.0 1.0 1.0 1.0 1.0
<6.5 >2.50, 3.50 1.6 1.5 1.4 1.3 1.2 1.1 1.0 1.0 1.0
>3.50 1.6 1.6 1.6 1.5 1.5 1.4 1.3 1.2 1.1
2.50 1.6 1.5 1.4 1.2 1.0 1.0 1.0 1.0 1.0
6.5 >2.50, 3.50 1.6 1.5 1.4 1.2 1.3 1.3 1.3 1.3 1.3
>3.50 1.6 1.6 1.6 1.5 1.5 1.5 1.4 1.4 1.4
Note: Interpolation in this exhibit is not recommended.
Step 3: Estimate Saturation Flow Rate

If the saturation flow rate is not measured in the field, a directional
saturation flow rate can be estimated by using Equation 26-B5 with Equation
26-B6 and Equation 26-B7.
3,600 Equation 26-B5
𝑠𝑖 =
ℎ̂𝑖
with
ℎ̂𝑖 = ℎ0 × 𝑓speed,𝑖 Equation 26-B6
𝑓speed,𝑖 = 1 − 0.005(min[𝑆𝑖 , 45] − 45) Equation 26-B7
where
si = saturation flow rate for direction i (pc/h);
ĥi = adjusted time headway for direction i (s);
h0 = base saturation headway (s/pc) = 3,600/1,900 = 1.89 s/pc;
fspeed,i = speed adjustment for direction i (decimal); and
Si = average travel speed in direction i (mi/h).
Step 4: Estimate Green Time

The length of the green interval can be applied directly if a fixed-time signal
is applied at the work zone site. However, most work zones apply flagger
control, for which the green time in each cycle is not fixed. For flagger control
under relatively balanced directional demand conditions, a simple estimation of
optimal directional effective green time can be found by using Equation 26-B8.
20 0.0375𝑙 < 20
𝐺𝑜𝑝𝑡 = {0.0375𝑙 20 ≤ 0.0375𝑙 ≤ 60 Equation 26-B8
60 0.0375𝑙 > 60
where
Gopt = optimal effective green time for one direction (s), and
l = work zone length (ft).

To ensure traffic can be fully discharged in two directions, directional
effective green-time lengths must satisfy Equation 26-B9 with Equation 26-B10.
𝑣𝑖
𝐺𝑖 ≥ 𝐺𝑖,𝑚𝑖𝑛 = (𝐶 − 𝐺𝑖 ) Equation 26-B9
𝑠𝑖 − 𝑣𝑖
with
𝑙 𝑙
𝐶= + + 𝐺1 + 𝐺2 + 2𝐿𝑆 Equation 26-B10
𝑆1,𝑓𝑝𝑠 𝑆2,𝑓𝑝𝑠
where
Gi = effective green time for direction i (s),
Gi,min = minimum effective green time for direction i (s),
si = saturation flow rate for direction i (pc/h),
vi = demand flow rate for direction i (pc/h),

C = cycle length (s),
Si,fps = average travel speed in direction i (ft/s) = (Si × 5,280 ft/mi)/(3,600 s/h),
Si = average travel speed in direction i (mi/h), and
LS = start-up lost time (s).
Step 5: Calculate Capacity

Directional capacity is calculated by Equation 26-B11.
𝑠𝑖 𝐺𝑖
Equation 26-B11 𝑐𝑖 =
𝐶
where
ci = capacity for direction i (pc/h),
si = saturation flow rate for direction i (pc/h),
Gi = effective green time for direction i (s), and
C = cycle length (s).
The start-up lost time, the elapsed time between the last vehicle in the
opposing direction exiting the work zone and the entry of the first queued
vehicle traveling in the subject direction, is assumed to be independent of traffic
direction, as the two directions follow the same traffic control plan. A default
value of 2 s for each direction is recommended.
The total capacity ctotal (in passenger cars per hour) can be calculated by
summing the two directional capacities, as shown in Equation 26-B12.
𝑠1 𝐺1 + 𝑠2 𝐺2
Equation 26-B12 𝑐total = 𝑐1 + 𝑐2 =
𝐶
QUEUING AND DELAY ANALYSIS

The previous steps provide a simple procedure to check two-lane highway
work zone capacity. In practice, it might also be useful to have performance data
such as delay and queuing. Users can apply the model to determine the optimal
control plan while minimizing the vehicle delay and queuing data.
A simple way to estimate vehicle delay and queue length is by assuming
deterministic traffic flow for both directions. Exhibit 26-B7 shows the deterministic
queuing diagram for a two-lane highway work zone. Although more accurate
estimates can be calculated from microscopic simulations that incorporate
random processes, these estimates might be difficult to accomplish in practice
because of the extra time and resources required. Therefore, by a similar
procedure to that used in Chapter 19 for signalized intersection control delay
estimation, the incremental delay caused by random arrivals is added to the
deterministic queuing delay associated with the work zone. The interval gi
shown in the exhibit is the portion of the green time with saturated departures.
The maximum queue length for each direction Qi,max (in passenger cars) is the
height of the triangles in the queue length area of the exhibit. These lengths can
be calculated by Equation 26-B13 and Equation 26-B14 for Directions 1 and 2,

respectively.
𝑣1 𝑙 𝑙 Equation 26-B13
𝑄1,𝑚𝑎𝑥 = ( + + 𝐺2 + 2𝐿𝑆 )
3,600 𝑆1,𝑓𝑝𝑠 𝑆2,𝑓𝑝𝑠
𝑣2 𝑙 𝑙
𝑄2,𝑚𝑎𝑥 = ( + + 𝐺1 + 2𝐿𝑆 ) Equation 26-B14
Exhibit 26-B7
Directional Queueing Diagram
for a Two-Lane Highway
Lane-Closure Work Zone
Source: Schoen et al. (B-1).
For undersaturated conditions, directional vehicle delay caused by a two-lane

highway work zone with one lane closed can be represented by Equation 26-B15
𝑑 = 𝑑1 + 𝑑2 Equation 26-B15
where
d = control delay per passenger car (s/pc),
d1 = uniform control delay assuming uniform traffic arrivals (s/pc), and
d2 = incremental delay resulting from random arrivals and oversaturation
queues (s/pc).
For each direction i, the total directional uniform control delay per cycle Di,1 (in
seconds) is the triangle area in the queue length diagram (Exhibit 26-B7). It is
calculated as one-half the queue length multiplied by the queueing duration. Di,1
is given by Equation 26-B16.
𝑠𝑖 𝑣𝑖
𝐷1,𝑖 = (𝐶 − 𝐺𝑖 )2 Equation 26-B16
2(𝑠𝑖 − 𝑣𝑖 )
The average uniform delay in direction i is given by Equation 26-B17.
𝐷1,𝑖 𝑠𝑖 (𝐶 − 𝐺𝑖 )2
𝑑1,𝑖 = = Equation 26-B17
𝑣𝑖 𝐶 2(𝑠𝑖 − 𝑣𝑖 )𝐶
Finally, by following Equation 19-26 in Chapter 19, the average incremental

delay in direction i is given by Equation 26-B18.
8𝑘𝐼𝑋𝑖
Equation 26-B18 𝑑2,𝑖 = 900 𝑇 [(𝑋𝑖 − 1) + √(𝑋𝑖 − 1)2 + ]
𝑐𝑖 𝑇
where
T = analysis period duration (h),
k = incremental delay factor (decimal),
I = upstream filtering adjustment factor (decimal),
ci = directional capacity (pc/h) from Equation 26-B11, and
Xi = directional volume-to-capacity ratio or degree of saturation (unitless).
Values for k can be calculated with Equation 19-22 in Chapter 19. For fixed-
time control, k = 0.5. Because the purpose of calculating delay in a work zone
context is to identify the optimal effective green time, which is assumed to repeat
every cycle, a value for k of 0.5 is recommended for use in Equation 26-B18. It
incorporates the effects of metered arrivals from upstream signals or work zones.
If the work zone is isolated, then I = 1.0.
The average delay per passenger car is the sum of the directional total
delays, divided by the total number of passenger cars, as shown in Equation 26-
B19. Note that the traffic flow rates used in the equation are in units of passenger
cars per hour; therefore, vehicle delay is calculated in terms of seconds per
passenger car.
(𝑑1,1 + 𝑑2,1 )𝑣1 + (𝑑1,2 + 𝑑2,2 )𝑣2
Equation 26-B19 𝑑=
𝑣1 + 𝑣2
In equations calculating queue length and vehicle delay, all variables are
given by roadway or traffic parameters, except that directional effective green
time Gi should be determined by users. Thus users can change the traffic control
plan to optimize the result. Users must note, however, that they should not
arbitrarily choose an effective green-time value.
EXAMPLE CALCULATION
This subsection presents an example application of the methodology. An
isolated 1,000-ft-long work zone will be located on a rural two-lane highway.
Known peak hour roadway and traffic parameters are summarized in Exhibit 26-
B8 and Exhibit 26-B9.
Lane Width Shoulder Width No. of Access General Terrain Exhibit 26-B8
Direction (ft) (ft) Points per Mile Type Example Calculation: Work
1 12 3 0 Rolling Zone Roadway Parameters
2 12 3 0 Rolling
Speed Limit Traffic Demand Truck RV Exhibit 26-B9

Direction (mi/h) (veh/h) PHF Percentage Percentage Example Calculation: Work
1 45 300 0.88 10.0 10.0 Zone Traffic Parameters
2 45 300 0.88 10.0 10.0
Step 1: Collect Data

Most of the necessary data are provided in the problem statement. However,
the traffic demand Vi (in vehicles per hour) must be converted into a traffic flow
rate vi (in passenger cars per hour).
𝑉𝑖
𝑣𝑖 =
𝑃𝐻𝐹 × 𝑓𝑔 × 𝑓𝐻𝑉
This equation requires determining both an adjustment factor for grade (in
this case, general terrain) and an adjustment factor for heavy vehicles (which also
includes terrain effects). In addition, a peak hour factor (PHF) is applied; this was
given in the problem statement as 0.88 for each direction.
From Exhibit 26-B2, the grade adjustment factor fg for rolling terrain is 0.83,
while from Exhibit 26-B4, the truck PCE is 2.1 and the RV PCE is 1.1. The heavy
vehicle adjustment factor fHV can then be calculated from Equation 26-B2.
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1) + 𝑃𝑅 (𝐸𝑅 − 1)
1
𝑓𝐻𝑉 =
1 + (0.10)(2.1 − 1) + (0.10)(1.1 − 1)
𝑓𝐻𝑉 = 0.89
then
300
𝑣1 = 𝑣2 = = 461 pc/h
0.88 × 0.83 × 0.89
Step 2: Estimate Average Travel Speed

Average travel speed through the work zone is calculated with Equation 26-
B3 and Equation 26-B4 for Directions 1 and 2, respectively.
𝑆1 = 0.615 × 𝑆𝑝𝑙 − 𝑓𝐿𝑆 − 𝑓𝐴 − 2.4
𝑆2 = 0.692 × 𝑆𝑝𝑙 − 𝑓𝐿𝑆 − 𝑓𝐴 − 2.4
The speed limit Spl is given. From Equation 15-5, the adjustment for lane and
shoulder width fLS for 12-ft lane widths and 3-ft shoulder widths is as follows:
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)
𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 3)
𝑓𝐿𝑆 = 2.1 mi/h
Finally, from Equation 15-6, the adjustment for access point density is 0.0
mi/h when no access points are present. Then
𝑆1 = 0.615 × 45 − 2.1 − 0 − 2.4 = 23.2 mi/h

𝑆2 = 0.692 × 45 − 2.1 − 0 − 2.4 = 26.6 mi/h
Step 3: Estimate Saturation Flow Rate

Equation 26-B5 through Equation 26-B7 are used to estimate the saturation
flow rate through the work zone.
First, the speed adjustment factor is calculated for each direction as follows:
𝑓speed,𝑖 = 1 − 0.005(min[𝑆𝑖 , 45] − 45)
𝑓speed,1 = 1 − 0.005(min[23.2,45] − 45) = 1.11
𝑓speed,2 = 1 − 0.005(min[26.6,45] − 45) = 1.09
Next, an adjusted time headway is calculated for each direction as follows:
ℎ̂𝑖 = ℎ0 × 𝑓speed,𝑖
̂1 = 1.89 × 1.11 = 2.10 s
ℎ
̂2 = 1.89 × 1.09 = 2.06 s
ℎ
where the base saturation headway of 1.89 s/pc is as given in the text following
Equation 26-B6.
Finally, the saturation flow rate for each direction is calculated as
3,600
𝑠𝑖 =
ℎ̂𝑖
3,600
𝑠1 = = 1,714 pc/h/ln
2.10
3,600
𝑠2 = = 1,748 pc/h/ln
2.06
Step 4: Estimate Green Time

In Step 4, the effective green time length is determined. It may be difficult to
choose a green time value without knowing the traffic performance parameters,
but an estimate of the optimal value can be obtained with Equation 26-B8.
20 0.0375𝑙 < 20
𝐺𝑜𝑝𝑡 = {0.0375𝑙 20 ≤ 0.0375𝑙 ≤ 60
60 0.0375𝑙 > 60
As the work zone will be 1,000 ft long, the value 0.0375l computes to 37.5 s.
As 37.5 is between 20 and 60, it can be used directly; however, this value should
be checked to make sure it is long enough to discharge the vehicle queues.
Equation 26-B9 provides this check.
𝑣𝑖
𝐺𝑖 ≥ 𝐺𝑖,𝑚𝑖𝑛 = (𝐶 − 𝐺𝑖 )
𝑠𝑖 − 𝑣𝑖
The cycle length C is computed from Equation 26-B10, incorporating a
default value of 2.0 s for the start-up lost time.
𝑙 𝑙
𝐶= + + 𝐺1 + 𝐺2 + 2𝐿𝑆
𝑆1,𝑓𝑝𝑠 𝑆2,𝑓𝑝𝑠
1,000 1,000
𝐶= + + 37.5 + 37.5 + 2(2.0)
23.2 × 5,280/3,600 26.6 × 5,280/3,600
𝐶 = 134.0 s
then
461
𝐺1,𝑚𝑖𝑛 = (134.0 − 37.5) = 35.5 s
1,714 − 461
461
𝐺2,𝑚𝑖𝑛 = (134.0 − 37.5) = 34.6 s
1,748 − 461
As the optimal effective green time of 37.5 s is greater than the minimum
required time for each direction, it is accepted, and the process continues to
Step 5.
Step 5: Calculate Capacity

Directional capacity is calculated with Equation 26-B11.
𝑠𝑖 𝐺𝑖
𝑐𝑖 =
𝐶
(1,714)(37.5)
𝑐1 = = 480 pc/h
134.0
(1,748)(37.5)
𝑐2 = = 489 pc/h
134.0
As v1 < c1 and v2 < c2, this 1,000-ft work zone can serve the traffic demand
without accumulating vehicle queues when the effective green time is 37.5 s for
both directions.
Queuing and Delay

If desired, the maximum queue length and average vehicle delay can be
calculated for both directions. The maximum queue length is calculated from
Equation 26-B13 and Equation 26-B14 for Directions 1 and 2, respectively.
𝑣1 𝑙 𝑙
𝑄1,𝑚𝑎𝑥 = ( + + 𝐺2 + 2𝐿𝑆 )
461
𝑄1,𝑚𝑎𝑥 = (29.4 + 25.6 + 37.5 + 4.0) = 13 veh
3,600
𝑣2 𝑙 𝑙
𝑄2,𝑚𝑎𝑥 = ( + + 𝐺1 + 2𝐿𝑆 )
461
𝑄2,𝑚𝑎𝑥 = (29.4 + 25.6 + 37.5 + 4.0) = 13 veh
3,600
The average uniform delay by direction is calculated with Equation 26-B17.
𝑠𝑖 (𝐶 − 𝐺𝑖 )2
𝑑1,𝑖 =
2(𝑠𝑖 − 𝑣𝑖 )𝐶
(1,714)(134.0 − 37.5)2
𝑑1,1 = = 47.5 s/veh
(2)(1,714 − 461)(134.0)
(1,748)(134.0 − 37.5)2
𝑑1,2 = = 47.2 s/veh
(2)(1,748 − 461)(134.0)
The average incremental delay by direction is calculated from Equation 26-

B18 The recommended value of 0.5 is used for the incremental delay factor k, and
as the work zone is isolated, a value of 1.0 is used for the upstream filtering
adjustment factor I.
8𝑘𝐼𝑋𝑖
𝑑2,𝑖 = 900 𝑇 [(𝑋𝑖 − 1) + √(𝑋𝑖 − 1)2 + ]
𝑐𝑖 𝑇
461
2 (8)(0.5)(1.0) (
𝑑2,1 = (900)(1) [(
461 √
− 1) + (
461
− 1) + 480)] = 52.4 s
480 480 (480)(1)
461
2 (8)(0.5)(1.0) (
𝑑2,2 = (900)(1) [(
461 √
− 1) + (
461
− 1) + 489)] = 42.8 s
489 489 (489)(1)
Finally, the average delay per passenger car is given by Equation 26-B19.
(47.5 + 52.4)(461) + (47.2 + 42.8)(461)
𝑑= = 95.0 s
461 + 461
REFERENCE
B-1. Schoen, J. M., J. A. Bonneson, C. Safi, B. Schroeder, A. Hajbabaie, C. H.
Yeom, N. Rouphail, Y. Wang, W. Zhu, and Y. Zou. Work Zone Capacity
Methods for the Highway Capacity Manual. National Cooperative Highway
Research Program Project 3-107 final report, preliminary draft. Kittelson &
Associates, Inc., Tucson, Ariz., April 2015.

Chapter 26 - Freeway and Highway Segments Supplemental - 700

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 26 - Freeway and Highway Segments Supplemental - 700

Uploaded by

Copyright:

Available Formats

HIGHWAY CAPACITY

MANUAL 7th Edition

Copyright 2022 by the National Academy of Sciences. All rights reserved.

Printed in the United States of America

Hardcover International Standard Book Number-13: 978-0-309-08766-7

The three Academies work together as the National Academies of Sciences,

Learn more about the Transportation Research Board at www.TRB.org.

2. STATE-SPECIFIC HEAVY-VEHICLE DEFAULT VALUES .......................... 26-2

3. TRUCK ANALYSIS USING THE MIXED-FLOW MODEL .......................... 26-4

4. ADJUSTMENTS FOR DRIVER POPULATION EFFECTS ......................... 26-14

5. GUIDANCE FOR FREEWAY CAPACITY ESTIMATION .......................... 26-15

6. CONNECTED AND AUTOMATED VEHICLES .......................................... 26-22

7. FREEWAY AND MULTILANE HIGHWAY EXAMPLE PROBLEMS ....... 26-30

Chapter 26/Freeway and Highway Segments: Supplemental Contents

8. TWO-LANE HIGHWAY EXAMPLE PROBLEMS .........................................26-54

APPENDIX A: TRUCK PERFORMANCE CURVES .......................................26-102

APPENDIX B: WORK ZONES ON TWO-LANE HIGHWAYS ....................26-107

Contents Chapter 26/Freeway and Highway Segments: Supplemental

Exhibit 26-1 State-Specific Default Values for Percentage of Heavy

Chapter 26/Freeway and Highway Segments: Supplemental Contents

Exhibit 26-27 Example Problem 3: LOS Results ....................................................26-85

Contents Chapter 26/Freeway and Highway Segments: Supplemental

Chapter 26/Freeway and Highway Segments: Supplemental Introduction

2. STATE-SPECIFIC HEAVY-VEHICLE DEFAULT VALUES

Research into the percentage of heavy vehicles on uninterrupted-flow

Exhibit 26-1 Small Medium Large Small Medium Large

Two-Lane Multilane Two-Lane Multilane Exhibit 26-2

3. TRUCK ANALYSIS USING THE MIXED-FLOW MODEL

OVERVIEW OF THE METHODOLOGY

Notes: BP = breakpoint; FFS = free-flow speed; c = capacity.

Step 1: Input Data

Step 2: Compute Mixed-Flow Capacity Adjustment Factor and Capacity

CAF for the Auto-Only Case

CAF for Truck Percentage

CAF for Grade Effect

Once CAFmix is computed, the mixed-flow capacity can be computed with

Step 3: Compute Mixed-Flow FFS and FFS Adjustment Factor

𝜏𝑇𝑇 = 𝜏𝑇𝑇,𝑘𝑖𝑛 + 𝛥𝜏𝑇𝐼 Equation 26-7

3,600 Equation 26-8

Traffic Interaction Term

Auto-Only Speed for the Given Flow Rate

Kinematic Travel Rates for SUTs and TTs

Note: Curves in this graph assume a weight-to-horsepower ratio of 100.

Note: Curves in this graph assume a weight-to-horsepower ratio of 150.

Exhibit 26-7 Free-Flow Speed (mi/h)

Exhibit 26-8 Free-Flow Speed (mi/h)

FFS Adjustment Factor

Step 4: Compute the Speed–Flow Relationship Breakpoint for the

Step 5: Compute Mixed-Flow Speeds at Capacity and 90% of Capacity

Step 6: Compute the Exponent for the Mixed-Flow Model Speed–Flow

Scalib,cap = mixed-flow speed at capacity (mi/h),

Step 7: Compute the Mixed-Flow Speed Under Mixed-Flow Conditions

Step 8: Compute the Mixed-Flow Density Under Mixed-Flow Conditions

4. ADJUSTMENTS FOR DRIVER POPULATION EFFECTS

Exhibit 26-9 Level of Driver Familiarity CAFpop SAFpop

5. GUIDANCE FOR FREEWAY CAPACITY ESTIMATION

FREEWAY CAPACITY DEFINITIONS

Prebreakdown Flow Rate

Postbreakdown Flow Rate or Queue Discharge Flow Rate

CAPACITY MEASUREMENT LOCATIONS

Source: Elefteriadou, Kondyli, and St. George (6).

Upstream Bottleneck Downstream Exhibit 26-11

Upstream Bottleneck Downstream Exhibit 26-12

CAPACITY ESTIMATION FROM FIELD DATA