U.S. Department of Transportation
400 Seventh St., S.W.
Washington, D.C. 20590
Federal Highway Administration
July 17, 2003
Refer to: ROTO-l
Mr. Mark W. Bott
Traffic Operations Engineer
Michigan Department of Transportation
P.O. Box 30050
Lansing, MI 48909
Dear Mr. Bott:
Thank you for your June 19 letter requesting approval to experiment with a double solid line marking pattern for edgelines on some portions of approximately 83 miles of freeways in the areas to the north and east of Grand Rapids, in conjunction with a project to install shoulder rumble strips that are offset only 4 inches from the travel lane and to provide edgeline striping' within the rumble strips for enhanced wet night visibility ("rumble striping"). Your project will include some freeway sections where, in addition to the normal 4-inch wide right and left edgelines placed just within the travel lanes, you will add a second 4-inch wide solid edgeline on the shoulder, within the rumble strip, separated from the normal edgeline by 8 inches, on both the left and right shoulders. The left edgeline(s) will be yellow and the right edgeline(s) will be white.
From the information provided, we understand that Michigan department of transportation's preferred marking pattern for edgelines with the 4-inch offset rumble strips would be, to remove the existing edgeline stripe (on both left and right) from the travel lane and instead place a new 6-inch wide solid edge line 2 inches outside the travel lane, on the shoulder. Such an edgeline marking will have 2 inches of markings on the shoulder (within 2 inches of the travel lane) and 4 inches of markings within the rumble strip. A single 6-inch wide edgeline placed in this manner would not be in non-compliance with the Manual on Uniform Traffic Control Devices (MUTCD) and thus does not require experimentation approval.
You also indicate that, in the event your preferred edgeline-marking pattern produces unacceptable negative results (in terms of longitudinal joint maintenance or other considerations), a "fallback" edgeline pattern would be implemented. This pattern, labeled as Type B in your "Attachment C," is the one that features two 4-inch wide lines, one on the travel lane within 3 inches of the lane/shoulder joint, and one on the shoulder within the rumble strip. These two solid lines would be separated by 8 inches. This "fallback" marking pattern does not comply with the MUTCD, which calls for edgelines to be single lines and which defines a double solid line as a line which is prohibited to cross. Therefore, your Type B pattern does require experimentation approval.
We have reviewed your proposed experimentation and your evaluation plan and we approve of it, with some reservations as described below. Your experimentation request is approved for a 4-year time period. For future reference purposes, we have assigned the following official experimentation number and title to your request: "3-159(Ex)-Double Line Edgelines MIDOT. Please refer to this number in future correspondence.
Our reservations about your experimentation are centered on the issue stated in the first paragraph of your letter, particularly as it applies to the double white edgeline for the right shoulder. That is, the established meaning of a double white line is to prohibit crossing the line, yet that is not the intended meaning of the double white line that drivers will see as the right edgeline in your experimentation. Although double white lines may not have been used very much in Michigan, they are widely used in other States in applications including but not limited to:
We are willing to approve your experimentation as a "fallback" to your preferred marking pattern, because it is for only a limited portion of the Michigan freeway network and only for a limited time. However, we are very concerned about future implications, should Michigan and/or other States decide to adopt the 4-inch rumble strip offset for universal use and then use the double line edgeline "rumble striping" pattern with it. Universal striping of all of Michigan's freeways with double edgelines would condition drivers to a less restrictive meaning of double lines, and in our opinion this would likely lead to reduced respect for and effectiveness of the double line marking where it is really needed to convey a prohibition on lane changing for safety reasons. Therefore, the Federal Highway Administration is unlikely to approve a future revision to the MUTCD that would allow a double line edgeline marking where crossing the line is not prohibited. As "rumble striping" is gaining wider use for enhanced wet night visibility, we are encouraging States using this technique to consider use of either the wide edgeline (such as your Type A, or even wider) or else two 4-inch lines that are separated by a considerably larger dimension such that it appears different to the driver than the traditional double line marking.
We hope your overall project and this specific experimentation are both successful and we look forward to receiving your reports as the experimentation progresses. However, we would urge you to look ahead to the larger implications of possible future wide-scale marking of "rumble stripes" throughout Michigan and other States, and to develop a reasonable future marking pattern for this technique that does not look like a traditional double line.
If we can be of further assistance, please contact Mr. Scott Wainwright of our staff on -366-0857.I
Sincerely yours,
Vincent P. Pearce
Acting Director, Office of Transportation Operations
cc: Mr. Jim Baron, ATSSA
STATE OF' MICHlGAN DEPARTMENT OF TRANSPORTATION LANSING
JENNIFER M. GRANHOLM
GOVERNOR
GLORIA J. JEFF
DIRECTOR
June 19,2003
File No. 9-31
35-21-4B
35-40PM
Ms. Shelley J. Row, P.E.
Director, Office of Transportation Operations
HOTO, Room 3401
4007th Street, S.W.
Washington, D.C. 20590
Re: Request for Permission to Experiment with Double Line Pavement Markings
Dear Ms. Row:
Experimental Device: The Michigan Department of Transportation (MDOT) requests permission to experiment with a non-traditional use of double-line pavement marking. The double-line pavement marking will be applied as an edgeline, left and right, on some sections of freeway in the area of the state to the north and east of Grand Rapids. Whereas, the traditional meaning of a double-line marking is that the motorist should not cross the marking, the double-line marking in this case is simply a by-product of a broader safety initiative, and is not meant to prohibit crossing of the edgeline.
Background: The safety initiative that we are pursuing is a combination of shoulder rumble strip installation set close to the driving lane (4" offset) and edgeline striping on that rumble strip. We have determined from previous research that markings applied to standard milled rumble strips return a large quantity of light at night, and in the rain they outperform markings that are placed on non-rumbled pavement. In other words, simply by co-locating the shoulder rumble strip and the edge line, we are supplying the motorist with an all-weather marking at no additional cost to our program.
In addition to providing improved markings, this project is expected to increase the effectiveness of the rumble strip as a countermeasure to drowsy and distracted driving. Current MDOT standards for placement of the rumble strip are 24" offset from the right edge joint line, and 12" offset from the left edge joint line. These offset distances represent a compromise location, as agreed in Michigan by personnel representing "the disciplines of safety, maintenance and pavement design. To achieve the co-location with edgeline, the rumble strips on the experimental sections of this project have been moved in to a 4" offset from the edge joint line. The earlier contact with a drift off driver is expected to result in more successful intervention by the rumble strips. We note in Michigan that drift-off crashes are by far the most severe category of crashes that we experience on our freeway system.
MURRAY D. VAN WAGONER BUILDING.
P.O. BOX 30050. LANSING. MICHIGAN 48909
www.michigan.gov. (517) 373-2090
Ms. Shelley J. Row, P.E.June 19,2003
Illustrations:
Attachment A contains photographs that demonstrate wet night retro-reflectivity of a rumble strip proto-type marking placed on 1-75 for research purposes last year. (The rumble strip in this photograph is located at a 12" offset, so the striping pattern shown here is not directly comparable to the striping proposed on the upcoming project).
Attachment B demonstrates the experimental location of rumble strips on this project compared to current MDOT rumble strip and pavement marking edgeline standards.
Determination of marking schemes for this project:
Attachment C shows the two distinct marking schemes that will be applied to locations on this project where the rumble strip has been cut at 4" offset:
Patent disclaimer: The experimental markings we propose are not protected by patent or copyright.
Time period: We intend to collect safety data for three calendar years past the date of installation that is through the end of CY 2006. However, other measures of effectiveness such as vehicle positioning, noise, and public opinion will be complete by July 2004, and we expect to choose between the two striping patterns by November 2004.
Project location: US-31 north of Muskegon (23 miles); US-131 north of Grand Rapids (30 miles); 1-96 east of Grand Rapids (30 miles). Attachment D provides a map overview of the freeway sections, which will receive close-cut (4" offset) rumble strips on this project. The experimental double line will be installed on the southbound / westbound roadway on these routes.
Research plan: Several research efforts are planned to evaluate potential positive and negative benefits of this project. Vehicle positioning and noise research combined with public opinion surveys will address several key concerns about the relocation of both the rumbles and the edgelines from traditional Michigan placement. Crash data will be reviewed to investigate the effect of the experimental rumble/markings on drift-off and sideswipe crashes. The research Scope that will guide all these "non-marking" research efforts is attached as Attachment E.
Ms. Shelley J. Row, P.E.
June 19, 2003
Attachment F is the detailed work plan that has been approved to investigate the vehicle positioning, noise, and incidental contact concerns.
Attachment G is the detailed work plan that has been approved to investigate the durability and retro reflectivity of the marking schemes.
Reports: We agree to provide your office with semi-annual progress reports on our experimentation, and to provide you with a copy of our final report within 3 months following completion of all our studies.
Restoration to non-experimental condition: We agree to terminate the experimental markings if we detect that the markings are contributing to significant safety problems.
Sincerely,
Mark W. Bott
Traffic Operations Engineer
Enclosures
MWB:JGM:gb
cc: J. Friend
J. Culp
A. Uzcategui
J. Finch
D. Morena (FHWA)
Contract to Evaluate Grand Rapids markings/rumble project
Background:
In a projected May 2003 letting, MDOT will let a contract for installation of shoulder rumble strip on 165 miles of freeway in the MDOT Grand Region. The installation of the rumble strips and subsequent striping of pavement markings may begin as early as mid-June and should be completed by September 2003.
On approximately 80 miles of freeway, the rumble strips will be installed according to current MDOT policy at 12" offset to the left edge of driving lane and 24" offset to the right edge of driving lane. On approximately 85 miles of this freeway, the rumble strips will be offset 4" from existing driving lane, both left and right. On this 85 miles, several pavement marking strategies will be tried:
The performance of the pavement markings at different locations on this project will be
evaluated under separate MDOT/MSU Contract # 2000-0232, Control Section 84900, Job Number 51324.
Scope:
The consultant should devise and conduct research that will shed light on the following questions / issues:
NOTE: MDOT reserves the right to any camera equipment purchased for the
NOTE: MDOT reserves the right to any camera /video equipment purchased through this contract.
RESEARCH PROPOSAL
Attachment F
SUBMITIED TO: Ms. Jill Morena
Traffic and Safety Division ,
Michigan Department of Transportation
PO. Box 30050
Lansing, MI 48909
SUBMITTED BY: Dr. William C. Taylor, Professor
Civil & Environmental Engineering
3566 Engineering Building
Michigan State University
East Lansing, MI 48824
OFFEROR: Michigan State University, Contract & Grant Administration
301 Administration Building, East Lansing, MI 48824
Point of Contact: Fred Salas, Senior Contract & Grant
Administrator
Phone: 517/355-8729
TYPE: Institute of Higher Education
PROJECT TITLE: Assessment of the Operational Impacts of New Rumble Strip Pavement Paint Combinations
PROJECT DATES: May 15,2003 through June 30,2004
AMOUNT: $199,698.00
APPROVALS:Ronald Harichandran, Chairperson
Civil & Environmental Engineering
Margaret Wade,
Director Division of Engineering Research
Fred Salas
Sr. Contract & Grant Administration Contract and Grant Administration
Brian T. Day
Sr. Contract & Grant Administration Contract and Grant Administration
MSU DUNS No. 19-3247145 MSU Federal ID No. 38-6005984
Work Plan
a) Vehicle positioning and joint integrity
We propose to use 2 road segments where the rumble strips are placed 4" from the joint line for this part of the study; 1-96 and US-131. Each of the road segments will start with two sites selected for study. One near the beginning of the segment and one near the end of the segment. In addition, one control site on 1-96 and one control site on US-131 where the rumble strips are placed 2' from the joint line will be selected.
Before data will be collected at each of these six sites to be used in our comparisons. Approximately one month after the rumble strips are installed, a second set of data will be collected at these sites. Data will be collected a third time after a minimum of 4 months after the, rumble strips are installed. (This data may be as long as 8 months after installation, depending on the weather). This will allow us to study the effect of familiarity overtime and over distance (scope item 7).
An additional site located on a curve will be identified in each of the segments. The before observation and the 1 month" and 4 month observations will be made at these sites as well. We will compare the vehicle location at these sites with the vehicle location on straight segments. This adds 2 sites (one in each segment) where data will be collected three times.
The site on a curve, and one site on 'a straight section on each segment will be used to test the effect of day versus night. Data will be collected at these 4 sites at night, and a comparison will be made to the vehicle position data from the day observations from above.
The next set of data will be collected to study the vehicle positions when the road surface is wet. Two of the locations on each study segment will be observed under both wet-day, and wet-night conditions, one straight road location and the curve location.
The final test will be to determine the effect of snow. The same four sites observed in the wet condition will also be observed both in the day and at night, when there is snow covering the shoulder.
This experimental design will allow us to test each of the conditions identified in the project scope. This requires 44 separate observations. We propose to film each condition for 2 hours on each of 2 days using 2 cameras, one over the left edge line and one over the right edge line. This will yield approximately 350 hours of film to be reduced.
The data reduction will involve two operations. First, lateral position detectors will be spaced at approximately 6" intervals starting two feet inside the edge line and extending to 1 foot beyond the edge line. The counts from the various detectors will be used to determine:
The second operation will be a measurement of the lateral position of trucks, measured to the nearest 0.1 foot. Depending on the truck volume, either all trucks will be measured, or a sample of trucks large enough to establish the average lateral position will be measured. We will also determine the, lateral position of a sample of the automobiles to determine if the lateral placement differs from that of the trucks. The data will be segregated into four categories, with the lateral position determined for each category.
b) Speed
Speed measurements will be made only at one site in each segment. If there is a permanent monitoring station in any of the segments, this data will be used to, test all conditions included in the lateral placement studies (day/night; wet/dry; snow; and combination of wet-night, etc.).
If a segment does not contain a permanent monitoring, station, the Transportation Planning Division of MDOT will be requested to place speed-monitoring devices on the segments to collect a sample of data under each condition of interest.
It is also possible to obtain speed data from the films, but our experience with.' the variable speed limit in work zones study suggests that the camera view needed for the lateral placement study may result in inaccurate speed determinations. Thus a separate filming session under each of the eight conditions will be required at each segment. This filming session will be conducted immediately after the filming of traffic to be used in the lateral placement study.
c) Noise
The number of times a vehicle encounters the rumble strips under each condition at each site will be obtained from the films as noted above.
A noise meter will be used to determine the noise level (dbA) at the outer edge of the shoulder and at the RIW line for the before condition. Following the installation of the rumble strips, noise measurements will be made under controlled conditions. The vehicle type and vehicle speed will be varied to produce a. noise level distribution.
In addition to these observations, an audio recorder will be used to record the noise level at six sites to be selected by MDOT. A 2-hour recording will be made twice at each site (day and night). Recordings will also be made during the noise meter observations, and the playback volume calibrated to reproduce the volume level observed in the field.
d) Public and Motorist Acceptance (no included in this proposal)
e) Crashes (not included in this proposal, because data will not be available until summer 2005)
f) Useable Shoulder Width
We propose to drive each of the 4 segments (including the segments used as control sites) used in this study each time we go to the field to film. The distance from the- edge joint to all vehicles parked on the shoulder will be measured. This will provide data from approximately 70 days on each segment. If we find that the sample is not sufficient for statistical analysis, observations will be made on additional days to obtain the appropriate sample size.
g) Motorist Learning Curve
This objective is covered in tasks a, b and c.
Attachment G
Research Objective
The future objective of this project is to study the visual and reflective performance of two painted rumble strip configurations on selected interstate roadways in Grand Region. The main effort will be to determine the benefits, if any, of applying waterborne paint to a rumble strip. This will involve assessing day and night reflectivity under various weather conditions, using mobile and hand-held reflectometers along with still and video cameras. Based on previous information from the 2002 Gaylord project, there should be a significant improvement in wet- night reflectivity due to the presence of glass beads on a profiled surface. Effects of winter maintenance, snow cover, water pooling, and other environmental factors will be studied to determine whether painted rumble strips provide better year round delineation than conventional edge lines. This will include a comparison between the single six-inch line and a double four inch line configurations.
Research Approach
Many factors are involved in determining if painted rumble strips, located off the edge of metal, are advantageous. Reflectivity will be measured using a Laserlux van in mobile and static modes to try to separate the contribution of beads on flat and profiled surfaces along the rumble strip. A hand-held reflectometer and digital still camera will be used to augment the Laserlux in measuring how light interacts with glass beads on a profiled surface. Instrumental reflectivity measurements are not useful in documenting how the eye actually perceives light. A video camera will record before and after scenes of project sites for a subjective review of both line configurations under rainy, snowy, day, and night conditions.
Oct 2002 - April 2003 Project Tasks
May - Sept 30. 2003 Grand Region Tasks
1. Prepare day and night video of2003 rumble strip test sites before construction to document existing pavement marking delineation.