Amy L Simpson

Effectiveness of Maintenance Treatments of Flexible Pavements

To achieve effective pavement maintenance, the life expectancy and timing of treatment applications need to be determined. The Long Term Pavement Performance (LTPP) program includes the Specific Pavement Study-3 (SPS-3), which focuses on this subject. The treatments applied are chip seals, crack seals, slurry seals, and thin overlays. In studying the life expectancy it is not feasible to wait for all the sections in the experiment to fail. Thus, there is a need to determine the life expectancy while making efficient use of the available data collection funds. Survival data analysis is a statistical technique that meets this need by accounting for the portion of the sections in which the exact time the treatment lasted is not known. The application of this technique to flexible-pavement maintenance is presented. In addition, some results of the LTPP SPS-3 experiment are presented to the highway community. The focus is on the LTPP Southern Region. The results showed that the probability of failure was two to four times higher for the sections that were in poor condition at the time the treatment was applied than those sections that were in better condition. The median survival times for thin overlays, slurry seals, and crack seals were 7, 5.5, and 5 years, respectively. The chip-seal sections had not yet reached the 50 percent failure probability after 8 years of the SPS-3 experiment. Accordingly, chip seals appear to have outperformed the other treatments investigated in this study in delaying the reappearance of distress.

Collecting and Interpreting Long-Term Pavement Performance Photographic Distress Data: Quality Control-Quality Assurance Processes

The Long-Term Pavement Performance (LTPP) program is making use of photographic technology to provide, first, detailed, distress-specific condition data for use in the development and validation of pavement performance models and, second, a permanent, objective, high-resolution record of pavement condition over the full length and width of the sections in the program. Because high-quality distress data are critical to the success of the LTPP program, numerous quality control-quality assurance (QC-QA) processes have been implemented over the life of the program. Those processes address data quality before the start of data collection, during data collection, during data interpretation, after data interpretation, and after the uploading of data to the LTPP database. Some of the processes are a direct result of advance planning, based on past experience, while others are the result of lessons learned in the course of the program. Some were implemented early in the program, while others were introduced well into the program. The Distress Identification Manual, distress rater accreditation workshops, time series review, database checks, data studies and analyses, and feedback reports are just a few of the elements that make up the full suite of QC-QA processes. A detailed summary of the QC-QA processes associated with the LTPP photographic distress data is presented.

A study of the suitability of the KENTORT II eastern shale oil for asphalt paving applications

Abstract The uncertain future of petroleum reserves has fuelled the search for alternative resources. A feasibility study was conducted to determine potential paving applications of the oil extracted from eastern shale by the KENTORT II process. The eastern shale oil (ESO) in this study was separated into two drastically different viscosity portions, designated as ‘hard’ and ‘soft’ ESO. It was hypothesized that the ‘hard’ portion might enhance the asphalt performance by increasing the stiffness. It was discovered that the ‘hard’ ESO modified asphalt properties deteriorate significantly with time. On the other hand, the ‘soft’ ESO was found to exhibit desirable properties in an asphalt recycling application. Further studies are recommended to fully characterize the binder and mixture properties of ESO modified/rejuvenated asphalts.

OFF-THE-WALL PAVEMENT DISTRESS VARIABILITY STUDY

Variability of pavement surface distress data collection has always been an area of significant concern. When conducting evaluations of distress data manually (with raters observing pavements in question, interpreting what they see, and recording on paper) the process is subject to human errors. To minimize the impact of such human errors on these important pavement performance data, sophisticated equipment has been developed to eliminate as much of the human intervention as possible. Such technology is not without its own limitations of precision and bias. With both methodologies being used for the collection of surface distress data for the long-term pavement performance (LTPP) program, questions regarding precision and bias have been identified. In attempting to define the variability of the data for incorporation in stochastic analyses, it has become apparent how diverse and complex these distress data truly are. To adequately quantify the precision and bias, a detailed experiment was designed to evaluate the errors inherent in the different distress data collection methodologies. The facet of the experiment reported targets the variability of human distress surveyors and the biases associated with conducting surveys from film, using a slightly different projection system. Specifically, a collection of surveyors was assembled to establish the variability associated with experienced raters versus novice raters, engineers versus engineering technicians, and teams versus individuals.

Trends in Deflection with Application of Repeated Loads: Impact on Deflection Data Averaging

Falling weight deflectometer (FWD) testing is now considered routine for the evaluation of pavement structures. Averaging of FWD load and deflection data collected at the same location and similar drop loads is common before further analysis. The stated reason for this averaging is to decrease the effect of the random error inherent in FWD deflection sensors. Implicit in this methodology is the assumption that there is no significant trend in deflection results with successive load applications. FWD data from the Long-Term Pavement Performance program were analyzed to determine whether a significant trend exists between deflection and drop number. In a majority of data sets, a statistically significant trend does exist for at least one deflection sensor. The trends are most common for the highest load level and the center deflection sensor. These trends are less often practically significant but are common enough that averaging deflection data can be expected to increase total error in a large minority of cases.

Sampling to Evaluate Performance on the Interstate Highway System

The Moving Ahead for Progress in the 21st Century Act highlighted the need to monitor the performance of pavement sections over time. An important aspect of performance monitoring is collecting a statistically significant sample of the network so conclusions can be made about changes in performance and types of maintenance and rehabilitation needed for maintaining these assets. This paper reviews the sampling requirements on the Interstate highway system to draw accurate conclusions about the network. Specifically, the paper answers the following questions: Is two-way data collection necessary? Do data have to be collected in more than one lane in a direction? What is the optimum reporting length? For what percentage of the network should data be collected? Should data be collected annually?

Pavement Smoothness at Weigh-in-Motion Sites: Developing Specifications for the Long-Term Pavement Performance Program

Accurate traffic data are vital to the success of the Long-Term Pavement Performance (LTPP) program. To minimize dynamic motions and therefore improve the accuracy of traffic loading data at weigh-in-motion (WIM) scales, pavement smoothness specifications were developed to screen sites for excessive truck dynamic loading that exacerbates scale error beyond ASTM recommended tolerances. WIM scale error was related to pavement profile characteristics by using a large simulation study of the response of virtual trucks over measured profiles. A distribution of error was compiled over the truck population for steer axle, tandem axle, and total vehicle weight at each site. The error distributions were summarized by their 95th percentile absolute error levels. The error levels assigned to each site were then used as a correlation standard for the proposed roughness indices and for the selection of corresponding threshold values. Instead of a single index, two versions of the Butterworth filter were selected for use in the specifications. One addresses short-range roughness and the other long-range roughness. The shortand long-range WIM error indices were then statistically related to WIM scale error to set the threshold values. An overview is presented of the development of the pavement smoothness specifications, including indices and threshold values.