|FHWA > HfL > Projects > Michigan Demonstration Project > Data Acquisition and Analysis|
Performance Contracting for Construction on M-115 in Clare County, MI
Data Acquisition and Analysis
Data collection on the MDOT HfL project consisted of acquiring and comparing data on safety, construction congestion, quality, and user satisfaction before, during, and after construction. The primary objective of acquiring these types of data was to provide HfL with sufficient performance information to support the feasibility of the proposed innovations and to demonstrate that PCfC can be used to do the following:
This section discusses how well MDOT project met the specific HfL performance goals in these areas.
The HfL performance goals for safety include meeting both worker and motorist safety goals during construction. No workers were injured during the construction of the M-115 project, so the contractor exceeded the HfL goal for worker safety (an incident rate of less than 4.0 based on the OSHA 300 rate).
MDOT set a goal of less than 1.0 crash per month (excluding animal crashes) during construction, based on three other projects constructed between 2004 and 2006 on M-115 and US-10 in Clare County and M-115 in Osceola County. The crash rates (excluding animal crashes) for these three construction projects adjusted for project length were 1.24, 0.33, and 0.99 per month, respectively. Two motorist incidents involving crashes with deer were reported over the 3.5-month construction period, resulting in a crash rate (excluding animal crashes) of 0.0 crashes per month.
From the Crash Analysis and Safety Review, dated March 22, 2006, this M-115 roadway segment experienced a total of 58 crashes, including 11 injuries and no fatalities, from 2000 to 2002. The majority of the crashes consisted of 38 (66 percent) animal crashes, seven (12 percent) fixed-object crashes, six (10 percent) miscellaneous single-vehicle crashes, and three (5 percent) overturn-type collisions. The remainder included the following crash types: one head-on, one rear-end, one side-swipe, and one head-on left-turn crash. No section of this roadway appeared on MDOT's 2000–2002 Bay Region Surveillance Report. A review of the fixed-object crashes indicated that the objects struck were four trees, two ditches, and one mailbox. Of the seven fixed-object crashes, five (71 percent) occurred during wet conditions: two icy/snowy conditions and three roadway conditions.
As part of this HfL M-115 construction project, rumble strips were constructed on the shoulder to alert animals to approaching vehicles, minimizing animal crashes and improving safety. An improvement in the pavement surface characteristics is expected to reduce wet condition crashes. These measures taken to improve long-term safety will be tracked for several years.
The performance goal on motorist delay was that no vehicle should be delayed by contractor operations more than 10 minutes beyond its normal travel time. The normal travel time at 55 mi/h (88.5 km/h) for 11 mi (17.7 km) was estimated at 12 minutes. The method of evaluation was to perform onsite total travel time measurements four times per week, twice during the weekdays (Monday through Thursday) and twice on the weekend (Friday through Sunday). Each measurement would include both directions of travel and the measurement for the direction with the highest delay would be recorded as the delay time. Incentives and disincentives were awarded based on this travel time.
To attain the maximum incentives, Central Asphalt Inc. chose several innovations that were not part of MDOT's original plans, including precast bridge construction, self-adjusting temporary signals to control single-lane traffic during precast bridge construction, 24-hour roadside patrol within the construction zone to minimize delays caused by breakdowns, and 11-ft-wide (3.3-m-wide) temporary traffic lanes during major construction stages to provide two-way traffic. As a result of these innovations, the average delay based on 54 measurements was 2 minutes and 16 seconds. The distribution of these measurements is shown in figure 28.
Sound Intensity Testing
Sound intensity (SI) measurements were taken on November 15, 2007, before reconstruction, using the latest industry standard onboard sound intensity (OBSI) equipment. The measuring device was the OR25 OROS (www.oros.com) analyzer with four GRAS (www.gras.com) 0.5-in (12.7-mm) microphones. The OROS NVGATE software processed the recorded data. The recorded data were analyzed with the third octave band approach and averaged logarithmically over the three runs and between leading and trailing edges.
The OBSI measurements were executed using two pairs of phase-matched sound intensity microphones attached to a bracket and adjacent respectively to the trailing and leading edges of the test vehicle rear wheel (figure 29). The microphones were set 4 in (101 mm) from the edge of the tire wall and 3 in (76 mm) off the ground, and the distance between the two pairs of microphones was 8 in (203 mm). The measurements consisted of three runs in each direction at a constant speed of 45 mi/h (72 km/h) using the standard reference test tire (SRTT), inflated at a pressure of 35 pounds per square inch (psi) (241 kilopascals (kPa)). Figure 30 shows the tread of the SRTT.
The system was calibrated before the OBSI measurements. After the SRTT was mounted on the vehicle, it was warmed up as the vehicle was driven for about 30 miles (48 km). The tire pressure was checked to verify the pressure of 35 psi ± 0.1 psi (241 kPa ± 0.7 kPa). The microphones were also calibrated using a Larson Davis signal generator and mounted on the bracket. After the OBSI measurements, another recording with the Larson Davis signal generator and data analysis confirmed that the microphone calibration was within tolerance.
The dual sound intensity probes simultaneously collect noise data from the leading and trailing tire-pavement contact areas, and the software uses Fourier transform to analyze the raw data signals over the full length of each test run to produce SI values. The values are normalized for environmental effects such as ambient air temperature and barometric pressure at the time of testing. The resulting A-weighted mean SI levels are filtered to produce the noise-frequency spectra in one-third octave bands, as shown in figures 31 and 32, for road and bridge sections.
The global noise levels for the northbound and southbound lanes are computed using a logarithmic addition of the intensity level corresponding to each frequency of the spectrum. Figure 33 shows the resulting spectra among the road and bridge sections. Table 6 includes the preconstruction global noise level measured at each bridge and road section and related statistics over three measurement runs for the northbound and southbound lanes.
The onboard preconstruction SI levels on M-115 in each direction of travel were as follows:
The average preconstruction SI level determined as described above was 99.4 dB(A).
On October 30, 2008, the postconstruction SI levels were acquired at 45 mi/h (72 km/h). The resulting A-weighted mean SI levels are filtered to produce the noise-frequency spectra in one-third octave bands, as shown in figures 34 and 35, for road and bridge sections. Figure 36 shows the resulting spectra among the road and bridge sections. Table 7 includes the postconstruction global noise level measured at each bridge and road section and related statistics over three measurement runs for the northbound and southbound lanes.
The onboard postconstruction SI levels on M-115 in each direction of travel were as follows:
The average preconstruction SI level determined as described above was 99.4 dB(A). These data suggest that the difference between pre- and postconstruction SI levels was significant and dropped from 99.4 dB(A) to 95.2 dB(A).
Smoothness measurements on the sections were collected by the Auburn University Automatic Road Analyzer (ARAN) van (figure 37) on the same days as the preconstruction and postconstruction OBSI measurements. The ARAN is a high-speed inertial profiler able to perform smoothness measurements of the pavement surface in both wheel paths. Smoothness is reported in in/mi (mm/km) as measured by the International Roughness Index (IRI). The latter consists of a mathematical assessment of the section profile aimed to quantify quality of the ride on a passenger car—the higher the IRI, the rougher the pavement, and the lower the IRI, the smoother the pavement. The ARAN van system provides data summarized every 25 ft (7.6 m) along the measured section.
The ARAN van performed three runs in each direction at a speed of 45 mi/h (72 km/h) and collected IRI data of the left wheel path (L-IRI), and right wheel path (R-IRI). The average of the two (A-IRI) was then calculated. Tables 8 and 9 show the preconstruction and postconstruction mean IRI of 115.5 and 37.8 in/mi, respectively. An analysis of the roughness data on the road and bridge sections indicated no significant differences. Table 8 shows that the southbound lane is rougher than the northbound lane before construction. Table 9 shows that following construction, there was no significant difference between the southbound and northbound lanes. Table 9 shows a dramatic improvement in smoothness and reduction in IRI following construction. Based on the field data collected after construction, the M-115 project exceeds both the HfL goals of IRI less than 48 in/mi and tire-pavement noise less than 96.0 dB(A) using the OBSI test method.
User satisfaction surveys were conducted before and after construction. This survey was difficult to sample because the users were seasonal tourists and MDOT had to substitute the major stakeholders to include businesses and homeowners. The following questions were included in the preconstruction survey:
A total of 46 responses were collected during the preconstruction survey. The results of the preconstruction survey, shown in figures 38 through 41, indicate a high level of dissatisfaction with the pavement condition and ride quality. A majority of those surveyed also indicated a high level of satisfaction with the proposed construction schedule and the daytime construction plan.
The following questions were included in the postconstruction survey:
A total of 43 responses were collected during the postconstruction survey. The results of the postconstruction survey, shown in figures 42 through 45, indicate that a majority of the respondents were very satisfied with the pavement condition and ride quality. The postconstruction survey also showed that more than half the respondents were somewhat dissatisfied or totally dissatisfied with the delays experienced in the work zone. This was a surprising find to MDOT because the average measured delay was 2 minutes and 16 seconds beyond the normal travel time and only one delay measured was beyond 10 minute maximum delay goal that was established for this project.. MDOT should evaluate the factors causing this apparent anomaly and adjust future goals and actions based on their findings.
This page last modified on 04/04/11