VARIABLE SPEED LIMIT DEPLOYMENT EVALUATION I-270/I-255 Traffic and Safety Conference May 12, 2010 Missouri University of Science and Technology and HDR.

VARIABLE SPEED LIMIT DEPLOYMENT EVALUATIONI-270/I-255

Traffic and Safety Conference May 12, 2010

Missouri University of Science and Technology

and HDR Engineering

VARIABLE SPEED LIMITTask 1: Assessment of Law Enforcement and Public Opinions

Task 2: VSL System Evaluation over Time

Task 3: VSL System Evaluation over Space (Highway Segment)

VARIABLE SPEED LIMITTask 1a Continuous On-line Public Survey

Task 1b Law Enforcement Paper Survey (2 times)

VARIABLE SPEED LIMIT

Task 2.1 Average speed/lane by direction during peak and off-peaks

Task 2.2 Speed limit during peak period by direction

Task 2.3 Speed limit compliance by posted speed limit and direction

VARIABLE SPEED LIMITTask 2.4 Travel times by direction

Travel Time in person-minutes Travel Time Index = “Actual Travel

Rate/Free Flow Travel Rate”

Task 2.5 Average traffic volumes during peak periods by direction

Task 2.6 Capacity, Evaluation of Delay, Queue Clearance and Occupancy

VARIABLE SPEED LIMITTask 2. 7 Crash Rates

Crash Rates Severity Rates Affect of Speed Limit on Crash Rates

Task 2.8 Evaluation of Delay and Queue Clearance Times associated with crashes/incidents

VARIABLE SPEED LIMIT

Task 3 – The VSL System has 22 Controlled Segments over the 38 mile corridor

Four segments will be evaluated with similar performance measurements identified in Task 2

VARIABLE SPEED LIMIT

OBJECTIVES:Evaluation of performance

measures◦ Pre and Post VSL system installation

Quantify the difference in fundamental traffic flow variables

Quantity Potential Benefits using Delay only

Cost savings in dollars

VARIABLE SPEED LIMITAggregation of 30 second data Lane data used

◦Lane 2 adjacent to median and Lane 3 adjacent to shoulder

◦Lane 1 is the farthest left lane and lane 4/5 is the right/auxiliary lane

Purpose◦Useful for data analysis

Travel Time: 5 min data used

VARIABLE SPEED LIMIT

Segments

1 I-64 to Route 100 SB

2 Route 21 to I-44 NB

3 Route 367 to I-170 EB

4 Route 370 to I-170 WB

SEGMENT 1

SEGMENT 2

SEGMENT 3 SEGMENT 4

VARIABLE SPEED LIMIT

Time Periods Pre VSL Time Periods Post VSL Time Periods

1 Start to mid Sept 07 Start to mid Sept 08

2 Mid to Last October 07 Mid to Last October 08

3 Start to mid Nov 07 Start to mid Nov 08

4 Start to mid April 08 Start to mid April 09

5 Dec 07 to mid Feb 08 Dec 08 to mid Feb 09

VARIABLE SPEED LIMITDaily Assessment

◦Thursday◦Tuesday, Wednesday, and Thursday (monthly comparison)

Weather conditions◦Clear◦Rain ◦Snow

Interval: 15 min data used

VARIABLE SPEED LIMIT

Pre VSL System Installation Post VSL System Installation

6th September 2007 4th September 2008

13th September 2007 11th September 2008

18th October 2007 16th October 2008

25th October 2007 23rd October 2008

8th November 2007 6th November 2008

15th November 2007 13th November 2008

10th April 2008 9th April 2009

17th April 2008 16th April 2009

VARIABLE SPEED LIMIT

PRIMARY PERFORMANCE MEASURES 1) Speed2) Volume

- Capacity Estimation3) Occupancy Analysis4) Congestion Measures

Highway Capacity Manual (2000) definition for Roadway Capacity:

“the maximum sustainable flow rate at which persons or vehicles reasonably can be expected to traverse a point or a uniform section of a roadway during a given time period, under prevailing conditions.”

VARIABLE SPEED LIMIT

CAPACITY ESTIMATION Functions based on Greenshield’s

models Spline regression model (fitting

the two different conditions congested and un-congested with two different straight lines and the point of intersection is the roadway capacity)

VARIABLE SPEED LIMIT

VARIABLE SPEED LIMIT

OCCUPANCY ANALYSIS Occupancy is the percentage of time when a short section of the highway is occupied

Critical occupancy is the point where maximum flow occurs

CRITICAL OCCUPANCY

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0 7 14 21 28

Vol

um

e (V

ehs/

hr)

Occupancy (Percentage Time)

Pre VSL Uncongested Traffic Post VSL Uncongested TrafficPoly. (Pre VSL Uncongested Traffic) Poly. (Post VSL Uncongested Traffic)

Critical Occupancy= 12.8%

Critical Occupancy= 10.5%

Post: Steeper Slope

Pre: flatter slope

VARIABLE SPEED LIMIT

CONGESTION MEASURESMeasures of Effectiveness 

◦ Travel Time ◦ Total Delay ◦ Percent of Congested Travel

TRAVEL TIME Travel Time

= Actual Travel Rate * Segment Length * Vehicle Volume * Vehicle Occupancy

where:

Travel Time is in person-minutes,

Actual Travel Rate = 60/Average speed, in minutes per mile,

Length of the segment is expressed in miles,

Vehicle Volume is expressed in vehicles, and

Vehicle Occupancy is expressed in persons/vehicles, 1.29.

Travel time represents time of travel between detectors on the segment

MODIFIED EQUATION: TRAVEL TIME

L1

Station

D1

Station

D2

where:T1-2 = travel time from station D1 to station D2, person-minutes,L1 = distance between station D1 and station D2, miles,V D1 , V D2 = average speed measured at station D1 and station D2, VOL D1 , VOL D2 = volume measured at station D1 and station D2

TOTAL DELAY Total Delay = (Travel Time – PSL Travel Time)where:

Total Delay is expressed in persons - minute,

PSL stands for Posted Speed Limit

PSL would have been 60 mph in the pre condition and displayed varied speed limit in the post condition.

TOTAL DELAYModified Total Delay: Total Delay =

Post VSL Travel Time – Pre VSL Travel Time

Total Delay is expressed in persons - minute

PCT = Total Delay / Travel Time

PCT = percent of congested travel

VEHICLE MILES TRAVELEDVMT = Total Volume * Segment

Length

Average Volume Change Pre versus Post= between -0.2% and -1.0%

VARIABLE SPEED LIMIT

VARIABLE SPEED LIMIT

VSL ACTIVATION LOGIC

Occupancy: >= 7 %Volume: count > 10 vehicles/lane/30

seconds Average speed: < 60 mph

Activation SequencePosted variable speed limit signs start to

flash Posted speed limit adjusted by 5 mph (40 to 60 mph)

Posted speed limit meters mainline traffic flows

Analysis of 16 days of data◦Total Segments: 4◦Total Detectors: 17◦Types of Analysis (for individual

detectors): 5◦Types of Analysis (for a segment): 2

Output:◦17 * 5 * 8 = 680 + 2 * 4 = 688

analysis/dayData source approximately 1.4

billion lines of traffic data – average speed, volume, occupancy, per lane plus speed limit recommendation and current displayed

VARIABLE SPEED LIMIT

VARIABLE SPEED LIMIT

Final Report – End of 2010

Acceptance Plan Identification of VariablesAdaptive AlgorithmExplore European Concepts