Multimodal Analysis in the 2010 Highway Capacity Manual

Multimodal Analysis in the 2010 Highway Capacity Manual It’s not just cars anymore!

Jamie Parks, AICP HCAT Conference

May 9-10, 2011

Multimodal LOS in the 2010 HCM

History and background

Overview of methods

– Pedestrian

– Bicycle

– Transit

Example applications



Overview of methods

– Pedestrian

– Bicycle

– Transit


Level of Service

Describes user perceptions of transportation facilities

– A-F scale

Defined in the Highway Capacity Manual (HCM)

Many jurisdictions require LOS analysis for transportation studies

– What’s measured matters!

Why Measure Level of Service?

Provides a consistent, systematic evaluation of existing conditions

Puts results in terms that transportation professionals and the public can understand

Provides an objective way to identify needs and prioritize improvements

Provides a way to evaluate different improvement types and cross sections

Multimodal Analysis in the HCM: 1950 - 1985 Manuals

1950 HCM

– Streetcars and buses impact vehicle capacity at traffic signals

– Pedestrian impacts on vehicle capacity addressed indirectly

1965 HCM

– LOS concept introduced

– Short (11-page) chapter on bus transit, with little quantitative info

1985 HCM

– Pedestrian and bicycle chapters introduced

History of Multimodal Analysis in the HCM: HCM2000

Expanded pedestrian chapter

– Service measures: space per pedestrian, average delay, average travel speed

Expanded bicycle chapter

– Service measures: average travel speed, average delay, hindrance

Revised transit chapter

– Four passenger-oriented service measures: frequency, hours of service, passenger load, reliability

NCHRP 3-92 – Production of the 2010 HCM

HCM’s 5th Major Revision (1950, 1965, 1985, 2000)

Project began in October 2007

2010 HCM shipped in March 2011

Significant changes:

– Integrated Multimodal Approach

– Multi-volume re-organization

– Incorporates New Research

– Increased Emphasis of Alternative Tools

HCM Focus Group Findings

Many jurisdictions don’t require multimodal analyses

– Therefore, they are not performed

Jurisdictions that do want to perform bike/ped analyses don’t find the current HCM capacity-based measures useful

– Maryland & Florida use measures of user comfort

Most bike & ped facilities don’t have capacity issues

– No need to analyze them using HCM procedures

10

Multimodal LOS Measure Issues

Current HCM method focuses on speed, delay, and space

NCHRP 3-70 research (and intuition) suggest these aren’t the key factors

Auto volumes highly important to bike & ped service quality

HCM 2010 considers a broader range of factors for ped and bike analysis

HCM2000: Ped LOS A HCM2000: Ped LOS D

11

2010 HCM Approach

Focus on the traveler perspective

– Quality of Service: perception of how well a facility operates from traveler perspective

Allow evaluation of intermodal interactions and trade-offs

Mode Affected

Impacting Mode

Auto Ped Bike Transit

Auto

Auto & HV volumes Turning patterns

Lane configurations

Minimum green time Turn conflicts

Mid-block crossings

Turn conflicts

Passing delay

Heavy vehicle

Blocking delay

Signal priority

Ped

Auto & HV volumes Cycle length

Driver yielding Turn conflicts

Traffic separation

Sidewalk crowding Crosswalk crowding

Cross-flows

Shared-path conflicts

Bicyclist yielding

Heavy vehicle

Transit stop queues

Stop cross-flows Vehicle yielding

Bike

Auto & HV volumes

Auto & HV speed

On-street parking

Turn conflicts

Traffic separation

Shared-path conflicts

Min. green time

Turn conflicts Mid-block xings

Bike volumes

Heavy vehicle

Blocking delay Tracks

Transit Auto volumes

Signal timing

Ped. env. quality Minimum green time

Turn conflicts

Mid-block crossings

Bike env. quality

Bike volumes Bus volumes



Overview of methods

– Pedestrian

– Bicycle

– Transit


Traveler-Perception Models

Recent research has quantified traveler perceptions of multimodal facilities to develop QOS indexes

– Indexes incorporate multiple factors (e.g., volumes, lane widths, etc.)

Models allow more service-quality factors to be considered than traditional HCM measures

Models set LOS thresholds based on survey responses to actual conditions

Anticipated that future research will develop similar indexes for other facilities (e.g. roundabouts)

Perception Models

LOS based on a weighted index

– Combination of multiple variables

– Example:

Ped Signal LOS = 0.00569 (RTOR+PermLefts) + 0.00013 (TrafVol x

TrafSpeed) + 0.0681 (# LanesCrossed 0.514) + 0.0401ln(PedDelay) –

RTCI (0.0027PerpTrafVol – 0.1946) + 1.7806

LOS Ped LOS Score

A ≤2.00

B >2.00–2.75

C >2.75–3.50

D >3.50–4.25

E >4.25–5.00

F >5.00

15

Service Measures in the 2010 HCM

System Element Service Measure Provided

Chapter Auto Ped Bike Transit

Freeway Facility 10

Basic Freeway Segment 11

Freeway Weaving Segment 12

Fwy. Merge/Diverge Seg. 13

Multilane Highway 14

Two-Lane Highway 15

Urban Street Facility 16

Urban Street Segment 17

Signalized Intersection 18

Two-Way Stop 19

All-Way Stop 20

Roundabout 21

Interchange Ramp Term. 22

Off-Street Ped-Bike Facility 23

- Based on traditional service measure

- Based on traveler perception index

Multimodal LOS Defined for Urban Streets

MMLOS measures the degree to which the urban street design and operations meets the needs of each mode’s users

Four level of service results for the street:

– Auto, Transit, Bicycle, Pedestrian

A combined LOS is not calculated

Conceptual MMLOS Results

Mode AM Peak PM Peak

Auto C E

Transit B C

Bicycle D C

Pedestrian C D

17

MMLOS Urban Street Applications

Segments

– All four modes

Signalized Intersections

– Auto, ped and bike mode

Facility

– All four modes

segment facility



Overview of methods

– Pedestrian

– Bicycle

– Transit


LOS A LOS B LOS C

LOS D LOS E LOS F

Pedestrian LOS in the 2000 HCM

Pedestrian LOS in the 2000 HCM

What does it not include?

– Motor vehicle traffic volume

– Traffic speed

– Intersection delay

– Separation from traffic

– Adjacent land uses

– Driver yielding behavior

Pedestrian HCM 2000 material retained, but supplemented

LOS at Unsignalized Crossings

Estimates pedestrian delay

Allows consideration of different crossing treatments

Based on 4 factors – Traffic volume - # of lanes crossed

– Crossing distance - Motorist yield rate


Example:

2-lane arterial with marked crosswalk, but nobody is yielding… Inputs: 1,000 peak-hour vehicles

2 lanes crossed 30 feet crossing distance 10% yield rate

Output: Average delay = 44 seconds

Ped LOS = E


Example (cont.):

Install rapid-flash beacons to improve driver compliance…

Inputs: 1,000 peak-hour vehicles

2 lanes crossed

30 feet crossing distance

80% yield rate

Output: Average delay = 6 seconds

Ped LOS = B

24

Pedestrian LOS: Urban Street Segments

Factors include:

– Outside travel lane width (+)

– Bicycle lane/shoulder width (+)

– Buffer presence (e.g., on-street parking, street trees) (+)

– Sidewalk presence and width (+)

– Volume and speed of motor vehicle traffic in outside lane (-)

Pedestrian density considered separately

Worse of density LOS/ segment LOS used to determine LOS

25

Pedestrian LOS: Signalized Intersections

Factors include:

– Permitted left turn and right-turn-on-red volumes (-)

– Cross-street motor vehicle volumes and speeds (-)

– Crossing length (-)

– Average pedestrian delay (-)

– Right-turn channelizing island presence (+)



Overview of methods

– Pedestrian

– Bicycle

– Transit


Common Factors Affecting Cyclists

Proximity of bicyclist to motor vehicles

Speed of traffic

Volume of motor vehicle traffic

Percent heavy vehicles

Pavement condition

28

Bicycle LOS: Urban Street Segments

Factors include:

– Volume and speed of traffic in outside travel lane (-)

– Heavy vehicle percentage (-)

– Pavement condition (+)

– Bicycle lane presence (+)

– Bicycle lane, shoulder, and outside lane widths (+)

– Number of driveways (-)

– On-street parking presence and utilization (+/-)

29

Bicycle LOS: Signalized Intersections

Factors included:

– Width of outside through lane and bicycle lane (+)

– Cross-street width (-)

– Motor vehicle traffic volume in the outside lane (-)

Bicycle LOS E

• No shoulder • 12.5 foot lane • 50 MPH • 2 lanes, undivided • 8,000 ADT • 7% trucks

Bicycle LOS B

• 9.5 foot shoulder • 12 foot lane • 45 MPH • 2 Lanes, undivided • 17,000 ADT • 8% Trucks

Bicycle LOS – Road Diet Example

ADT = 13,500 vpd

Lanes = 2

Pavement = 4 (good)

Wt = 12 ft

Wl = 0 ft

Speed = 30 mph

12' 12' 12' 12'

48'

BLOS Evaluation:

LOS score Category

3.58 D

48'

12'14'12'5' 5'

Bicycle LOS - After

ADT = 13,500 vpd

Lanes = 1

Pavement = 4 (good)

Wt = 17 ft

Wl = 5 ft

SPp = 30 mph

LOS score Category

2.07 B

BLOS Evaluation:

Bicycle LOS Model Notes

Heavily dependent on shoulder/bike lane width

Based on perceptions of typical bicyclist

– Cyclists are diverse

Represents typical conditions, not anomalies

Does not include slope

Does not capture emerging facility types

– Shared lane markings

– Colored pavement

– Bike boxes

– Cycle tracks

Shared-Use Path Bicycle LOS

Calibrated user perception index for bikes on shared-use paths

4 key variables:

– # of “meeting events” with other users

– # of delayed passing attempts

– Path width

– Presence of centerline

Shared-Use Path LOS

Uses volumes of user types to estimate meetings and passings

– Based on assumed speed distributions for each user type

– Procedure is complex, but Excel calculator is available



Overview of methods

– Pedestrian

– Bicycle

– Transit


2010 HCM Transit Objectives

New transit LOS measure desired

– Single measure to facilitate comparisons with other modes and for compatibility with HCM

– LOS thresholds tied to user satisfaction

- LOS grades mean the same thing across modes

– Opportunity for comparing impacts of other modes, where impacts exist

Approach

Develop a model that relates LOS to factors that:

– Have been shown to be important to customer satisfaction

– Can be readily quantified

– Can be related to ridership, or changes in ridership

The more satisfying the service, the more likely people are to use it

– Based on on-board survey results showing important factors

Rank

Summary of On-Board Survey Factors

Virginia 2B Virginia 38B Portland 14 Portland 44 Florida 18

1 Frequency Frequency Frequency Frequency Frequency

2 Wait time Reliability Close to home Reliability Wait time

3 Reliability Wait time Reliability Close to home Close to home

4 Close to home Close to dest. Wait time Close to dest. Reliability

5 Service span Close to home Close to dest. Wait time Service span

6 Close to dest. Service span Service span

7 Friendly drivers

Model Inputs

Only includes factors inside the right-of-way and which can be affected by agency actions

– Frequency

– Speed (travel time rate)

– Reliability & stop amenities (excess wait time)

– Crowding (perceived travel time rate adjustment)

– Pedestrian LOS

Model Output

Transit LOS score

– A function of:

- Transit wait/ride score (weighted 89%)

- Pedestrian LOS (weighted 11%)

– Addresses all three trip components

– Weightings based on on-board survey results that found that walk-to-the-stop satisfaction accounted for 11% of overall satisfaction

Details in NCHRP Report 616

Example Application: Portland



Overview of methods

– Pedestrian

– Bicycle

– Transit


Data Requirements

Typical Transportation Analysis Data Collection Efforts – Peak Hour Traffic Counts

- Passenger Vehicles

- Pedestrians

- Bicycles

- Heavy Vehicles

– Roadway Inventory - Sidewalks

- Bicycle Lanes

- Transit Stops/Amenities

- Transit Schedule

- Posted Roadway Speed

- Roadway Cross-Section

- Median Treatment

- Illumination

– Signalized Intersection Data - Signal Timing Sheets

- Signal Phasing

- Right-turn on Red

Data Requirements

Additional MMLOS Data Collection – Weekday PM Peak Hour Traffic Counts

- Number of Vehicles by Lane

– Roadway Inventory - Length of Roadway Segment - Roadway Cross-Section Dimensions

– Sidewalks – Landscape Strip – Bicycle Lanes – On-Street Parking – Travel Lane – Median

- Number of Trees/Bushes - Percent Occupancy of Parking - Pavement Condition Rating - Number of Driveways within Segment

– Transit Inventory - Bus Occupancy - Transit Reliability - Average Trip Length

Example – NoMa Transportation Plan

Rapidly developing neighborhood near downtown DC

Project for DDOT to proactively and strategically prepare for change

– Improve safety, comfort, and efficiency of all transportation modes

MMLOS in NoMa

Proposed modifications included:

– Lane reductions

– 2-way to 1-way couplets

– Bike lanes

– Sidepaths/cycle tracks

– Sidewalk widening

Analysis

Applied to 6 key corridors

– Analyze existing and no build conditions

– Assess the impacts of proposed cross-sections

– Demonstrate benefits to stakeholders

Used “research” quality spreadsheet

– Analysis will be easier with software

Example: K Street

No Build

– High traffic volumes

– No bike lane

– Peak-hour restricted parking

– Ped LOS = D

– Bike LOS = D

Build (One-way conversion)

– Dedicated bike facility

– No sidewalk widening

– Ped LOS = D

– Bike LOS = B

Example: K Street

Results

Do nothing and LOS gets worse

Proposed cross-sections have significant multi-modal benefits

Bicycle LOS Pedestrian LOS

Example 2: Russell Street, Missoula, Montana

1.5 Mile-long Study Area 1 of 5 Bridge Crossings 2/3 lane cross section with limited pedestrian/bicycle facilities 20,000 – 25,000 ADT Important bike commute route Commercial and residential uses

3rd St

14th St Mount Ave

5th St

Broadway St N

Ru

ss

ell

St

53

Alternative Analysis – Cross-section and Traffic Control

Bicycle, Pedestrian, and Transit LOS Summary

Overall LOS for Alternatives and Options

DEIS Alternatives

Alt 1 – No Build

Existing 3 Lane

Volumes 5 Lane

Volumes Alt 2 Alt 3 Alt 4 Alt 5-R Option 6

Bike LOS - Southbound F F F F F E E F

Bike LOS – Northbound F F F F F E E F

Ped LOS - Southbound D D E C C C C D

Ped LOS – Northbound D D D C C C C C

Transit LOS - Southbound D D D D D D D D

Transit LOS – Northbound D D D D D D D D

Legend

LOS = A, B, or C

LOS = D

LOS = E or F

Overall Analysis Summary

DEIS Alternatives

Alt 1 (No Build)

Performance Measure 3-Lanes 5-Lanes Alt 2 Alt 3 Alt 4 Alt 5-R Option 6

Intersection Operations (LOS) 6 7 3 3 1 2 5

Corridor Operations (Travel Time) 2 3 4 4 1 4 7

Safety (Predicted Average Crash Frequencies)

6 7 2* 1* 4 3 5

Pedestrian LOS 6 7 3 3 1 1 5

Bicycle LOS 5 7 3 3 1 2 6

Transit LOS 2 2 1 1 1 1 1

* Best ranking due in large part to lower traffic volume scenario (3 lane demand versus 5 lane demand).

MMLOS Sample: NE 3rd Street (Business 97)

Five-Lane Cross-Section

Posted Speed of 35 mph

Bicycle Lanes

Curb-tight Sidewalks

Multiple Driveways Throughout

Long Traffic Signal Cycle Lengths

ADT Approximately 20,500

– 5.4% Trucks (AM)

– 3.5% Trucks (PM)

85-Foot Cross-Section

BAT Route 1

– 40-Minute Headways

– No Shelters or Benches

Data Forms: Segment Data

Data Forms: Roadway Cross-Section

Data Forms: Traffic Data

Data Forms: Transit Data

Analysis Results: Existing 3rd Street Section

3rd Street Strategies and Goals

Freight Route?

Primary Regional Connection?

Primary Transit Route?

Pedestrian Corridor?

Bicycle Corridor?

Cross-Section Option 1: Increased ROW

Cross-Section Option 2: Removal Of Bicycle Lanes

Cross-Section Option 3: Removal of Travel Lane

Cross-Section Option 4: Transit Improvements

Cross-Section Option 5: Access Management

Summary of Options

Other – Effect of Lane Widths on Capacity

Saturation flow rate the same for10-foot and 12-foot lanes – No capacity-basis for denying 10-foot lanes

Lane Width

Sat. Flow Adj. Factor

9 0.90

10 0.93

11 0.97

12 1.00

13 1.03

14 1.07

Lane Width

Sat. Flow Adj. Factor

< 10 0.96

10 – 12.9 1.00

≥ 13 1.04

HCM 2000 HCM 2010

HCM 2010 Summary

Non-auto modes will be integrated into the 2010 HCM far better than before

Urban street LOS method will facilitate ”complete streets” evaluations – Relative service quality provided for each mode’s travelers

– Trade-offs of different improvement alternatives or future demand scenarios can be evaluated

HCM 2010 Summary (cont...)

Provides agencies a way to quantify the relative benefits and disadvantages of roadway cross-section standards and design modifications

Provides a methodology for multi-modal performance standards or alternative mobility standards

Some important policy considerations: – Vehicular/Pedestrian/Bicycle/Transit Hierarchy?

– Multi-modal LOS standards?

Future Considerations

Address additional facilities – Roundabouts

– Ramp terminals

– Others

Incorporate emerging treatments – Cycle tracks

– Sharrows

– Others

Connect to safety research

Learn from practical applications!

Questions?

Thank you!

Jamie Parks

[email protected]

(410) 347-9610

mailto:[email protected]

Documents

Multimodal Analysis in the 2010 Highway Capacity Manual