Safety Study - Update on Crash Surrogate Study

2007 Before Data - Speed and Lateral Placement

February 2008
All of the before data, including the vehicle tracking, has been completed. Preliminary analysis should start in the March and April timeframe. Scheduling for the April data collection and curve restriping started this month.

December 2007
The before data reduction was completed, but the tracking task was still not complete. The data can be viewed through the link above, which also contains restriping information.

During the preparation to set project limits for restriping the study curves, several things were determined. First, it was decided to only restripe the treatment site curves with 6-inch wide edgelines, provided that the spring retroreflectivities were not too low, and that restripe retroreflectivities were kept within a range of 200 to 300 mcd/m2/lux. This was decided to avoid one of the problems with restriping contracts, which is inadvertently creating a wider line from difficulties with restriping directly over the existing marking. This problem was evident in the fact that the average edgeline marking width from each curve ranged from 4 to 6 inches. The restriping retroreflectivity range was set to minimize confounding the impact of line width with retroreflectivity. If the spring retroreflectivity values are below 150, then, additional curves may be restriped. If control site curves are restriped, they will be restriped with 4-inch wide edgelines. Second, it was decided to add one additional curve to the data set to help round out the study design. The before data for this curve will be collected in early April prior to restriping this curve. With the concern over the spring retroreflectivities, the field team that collects the before data for the 19th curve will also collect spring retroreflectivity measurements for all of the curves.

September 2007
The research team completely reviewed all of the initial data, and was still in the process of reducing the data for analysis. It was projected that the data reduction will be complete in December of 2007. The reduction process included developing various macros to efficiently reduce the data from its raw format into individual vehicle speeds and lateral positions as tracked through each counter in each curve. Then, manual spot checks of the reduced data were conducted to verify the proper reduction by the macros.

Simultaneously while the data reduction was being conducted, the research team was putting together restriping project limits for each of the curves to be submitted to TDOT by the end of the month. The research team continued to be in direct contact with TDOT, and the team had been told this timeframe will be adequate to ensure the curves are restriped prior to the after data collection period during the summer of 2008.

August 2007
The research team completed the data collection of the before-data at the 18 curves in Tennessee. The research team then began the data reduction process.

July 2007
The research team continued collecting data at the 18 curves. To this point, the research team had been able to collect data at 4 curves each week.

June 2007
The research team traveled to Tennessee on July 9th to collect the before data at the 18 curves. The plan was to visit at least 4 curves a week, and barring weather or equipment issues, complete the data collection by August 10th. The research team also met with Jim Maxwell of TDOT to start coordinating the restriping of the study curves for next summer. At the time, it was planned to install pavement markings that are 2 inches wider than the current markings. If the current markings were 4 inches wide, they would be restriped with 6 inches; if they were 5 inches wide, they would be restriped with 7 inches, etc. The curves will be restriped from approximately 500 feet prior to the warning sign all the way through the curve in each direction. Hence, the centerline restripe will be the length of the curve plus approximately 1000 feet. The centerline will be restriped with 4-inch wide markings.

May 2007
A researcher traveled to Tennessee to visit the 18 horizontal curves to be used during the field testing of wider edgelines in the safety portion of the FHWA 1907 project under SAFETEA-LU. The researcher, accompanied by three TDOT staff members, verified what the appropriate advisory speed and/or curve warning sign should be if required in each curve based on the MUTCD. It should be noted that the 10, 12, and 14 degree ball-banks readings were used rather than the current 16 degree ball-bank reading as stated in the 2003 MUTCD. This was done because the current 16 degree ball-bank is being reviewed. After driving all 18 curves, the researcher found that 14 of the 18 curves required either adding an advisory speed or reducing the current advisory speed, and only 2 of the 18 curves neither required an advisory speed or warning sign. Only 4 of the curves initially had advisory speeds posted while 14 of the curves had warning signs. In addition, 2 of the curves that do have warning signs do not require them. They were left because it was felt there may be some crash history and/or local complaint that required it. Another point to note is that 2 curves are located within winding road segments, but this is not considered a problem because the speed limit was 35 mph, the advisory speed was 30 mph, and the tangent approaches allowed for at least 10 seconds of approach prior to the PC.

Update on Multi-State Retrospective Crash Study

Safety Study

Durability Study

PM Demo Main Page