Investigating the Relationship of As-Constructed Asphalt Pavement Air Voids to Pavement Performance (2021)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

139 C H A P T E R 6 Implementation Plan The research team expects that most highway agencies will review the results of this study and determine if their agency’s construction standards should be modified to achieve improved performance of their asphalt pavements. That decision will be based on how the agency’s climate, traffic, pavement structure, and pavement performance compare to the data and models used in the study. This action will be a program change, not a project-by-project decision. To aid agencies in accomplishing implementation, this plan outlines the steps that should be taken to ensure that the study is disseminated to agencies. The research team recommends the following five methods to achieve broad dissemination: • Distribution of the report and User Guide – This report and the accompanying User Guide should be made available to the highway community using normal NCHRP publication practices and be added to the documents in InfoPave. • Webinar presentation – The content of this report should be consolidated into a 60 to 90-minute webinar. The key topics presented would be extent of the LTPP dataset used, the results of each analysis, how the results can be used, and examples of applying the models. The webinar should be recorded, edited, and made available on InfoPave. • TRB presentation and workshop – After a number of agencies have reviewed the report and examined their own data and program, a TRB workshop could be used to present the findings of the study and how the study was used by agencies. The agency presentations should include agencies that elected to change their program, agencies that elected to keep their program as-is, and those agencies that elected to repeat the analysis with in-state pavement data. • Regional conference presentations – There are numerous regional venues to present the research findings to encourage agencies to examine the research. A key advantage to presenting the research at regional meetings is that most of the participants will have similar climate, traffic, and pavement material conditions so the presentation can focus on those parameters. • Access to the regression and ANN models through InfoPave – The foundation of this study was the LTPP database, which is managed by FHWA and accessible through InfoPave. The models, particularly the ANN models, cannot be simply documented in the report. An appropriate host site is needed so that agencies can readily input their data and obtain model prediction results. The research team used available LTPP data to formulate a national understanding of the impact of as- constructed AV on asphalt pavement performance. The extent and nature of the data are limited to the sections selected from the LTPP program and must take into account that as-constructed AV was not a primary LTPP research objective. An agency’s pavement program may not be well represented in this study’s dataset and they could elect to do an in-house study following the process from this study and all in-state data. The following guidance is given for agencies that consider this implementation approach. More detailed discussion of the steps for an in-house study are provided in the User Guide. • Identify what types of pavements (new construction and/or rehabilitation) will be included. If rehabilitation is selected, consider narrowing the types of rehabilitation that will be included. • Identify what pavement performance measures will be included. Consideration should be given to what performance measures are critical and how much historic data is available. Consider a minimum

140 of 8 years of annual measured rutting and ride performance data and 10 years of annual measured fatigue and thermal cracking data. • Identify what types of input data are available as well as the quantity and quality of that data. Demographic (climate and traffic) and construction data (layers and material properties) are key to defining each section in the analysis. If rehabilitation sections are included, input regarding the pre- rehabilitation condition of the existing pavement is relevant to rehabilitation performance. • Identify a sufficient number of pavement sections to establish a sound analysis. Consider a minimum of 10 sections for each subgroup scatter plot trend, Analysis Method 1. Consider a minimum of 100 sections for developing regression models, Analysis Method 2. Consider a minimum of 250 sections for developing ANN models, Analysis Method 3. These recommended quantities are for new and rehabilitation sections separately. • For Analysis Method 1, select appropriate demographic ranges for each subgroup and a single appropriate value for each performance measure. • For Analysis Method 2, perform step-wise regression to narrow the field of input variables and use model equations that apply a proper shape to the performance curve. • For Analysis Method 3, this analysis method requires special expertise in ANN modeling and a reasonable understanding of how the pavement input data is related. • For Analysis Method 2 and 3, randomly reserve 10% of the pavement sections to test the model (validation).