Paths to BMD Implementation

Across the USA, many highway agencies are beginning the journey toward a new way of designing and accepting asphalt mixtures using the balanced mix design (BMD) system. Like pioneers during the American westward migration in the 19th century, several states have already crossed into unchartered territory, while others are now looking for information on the best path to take. To help with that journey, NCAT has been working on a guide for BMD implementation as a primary deliverable for NCHRP project 10 107. The draft implementation guide is currently being vetted through a stakeholder review process. A webinar and workshops are being planned for early 2022. The draft guide includes eight tasks (briefly described as follows), along with many more details, recommendations and suggested resources. 

NCAT's draft guide includes eight tasks to BMD implementation.

Task 1: Motivations and Benefits

Although each agency will likely have several reasons for implementing BMD tests and related specifications, motives are typically rooted in dissatisfaction with the status quo. For example, an agency may want to improve the average service life of their asphalt overlays by “weeding out” poor-performing mixes that pass their existing specifications but just don’t perform well, or they may want to better assess the impact of using recycled materials and/or innovative additives on pavement performance. 

The original vision of the Superpave mix design system was to include mixture performance tests in the mix design process for moderate and heavily-trafficked pavements, but the proposed tests from the SHRP program were only used for a few special projects, primarily because they were not practical for routine use for the thousands of mix designs used each year. Most state DOTs have revised Superpave mix design and binder requirements to make incremental improvements in the quality of their asphalt mixtures over the past 20 years. However, the dependence of the Superpave mix design system on volumetric requirements to assure good durability, particularly voids in mineral aggregate (VMA), means volumetric design criteria are also dependent on an accurate aggregate bulk specific gravity, which unfortunately has such poor precision that it is not a reliable property to ensure mixture durability. 

Task 2: Overall Planning

Implementation of BMD will take as much effort as the implementation of Superpave two decades ago. To accomplish this, it is essential to have highway agency and industry champions who will help overcome institutional challenges and resistance to change. Successful implementation will also require a well thought out plan and a joint agency-industry technical committee to shepherd the process and make key decisions along the way. 

One of the first steps is to set long-term goals and intermediate milestones. For example, some states may want to fully implement BMD tests and associated criteria for mix design and project acceptance of asphalt mixtures, but will first work toward an intermediate milestone of implementing BMD just for mix design approval. Other states may desire to only utilize BMD tests and criteria for projects on higher traffic, and/or high profile roadways. In communicating the plan to all stakeholders, it will be important to provide estimated timelines for the short-term and long-term goals, as some of the tasks in this plan are likely to take several years to complete.

Task 3: Selecting Tests

The first major decision along the path to BMD implementation is the selection of performance tests, which should begin with a basic understanding of the types of asphalt pavement distresses that are most prevalent in the state. Basically, it’s critical to know what aspects of pavement performance need to be improved – whether it's reflection cracking, thermal cracking, moisture damage problems, raveling, rutting, or some other distress. 

Once the BMD technical committee has that established, important factors to consider in selecting the appropriate tests include correlation to field performance, cost of the equipment, test variability and overall turn-around time from sampling to results. NAPA’s online BMD Resource Guide is an excellent resource for information on the BMD tests and many other aspects of BMD implementation. Click here to access the NAPA BMD Resource Guide.

In order to validate the selected BMD tests, states are strongly encouraged to consider field experiments similar to the Long-Term Pavement Performance SPS-9A experiment, Verification of SHRP Asphalt Specification and Mix Design (i.e. Superpave). A few pooled-fund projects, such as the NCAT Test Track and NRRA experiments, may also be helpful when correlating field performance to BMD test results, but local experimental projects using mix designs developed with that state’s existing criteria and a range of BMD test results will be crucial for setting appropriate specifications further down the path of implementation. 

Task 4: Acquiring Needed Resources

Preparing for the purchase of BMD and ancillary equipment (e.g. ovens, saws, etc.) often needs to be done well in advance of each organization’s budgeting cycle. Keep in mind that some existing labs may be constrained by space or electrical capacity, so additional lab space and or electrical circuits may also be needed (which will require even more lead-time). Consideration also needs to be given to staffing demands, which will largely depend on the new tests selected, the frequency of testing, and how many of the existing requirements will remain in-place once BMD tests are added for mix design approval and/or production acceptance.

Task 5: Establishing Baseline Data

This task requires significant efforts by each state and the industry to gather information necessary for setting appropriate criteria for mix design approval and acceptance during mix production, assuming the latter is a goal set by the agency. One source of information will come from benchmarking studies, which refers to the testing of existing mix designs across the state to assess the distribution of results. Each state should conduct its own benchmarking study, which will have two parts. One part is for lab-prepared mixes, such as used in mix design. The second part involves testing plant mix samples to represent tests for acceptance. Preferably, one lab would be responsible for conducting the benchmarking testing since using more labs would introduce between-lab variability in test results. 

Another critical source of information will come from shadow projects. Shadow projects are paving projects on which mixes are sampled for the normal acceptance tests, but additional samples are also pulled for BMD tests that are performed in the background. Results of the BMD tests from shadow projects are not used to accept or reject materials, adjust payment or even to motivate mix production changes. Rather, the three goals of the shadow projects are to: (1) better familiarize both agency and contractor personnel with the selected tests, (2) add to the database of test results from the benchmarking studies, and most importantly (3) gather information on typical production variability. Recommendations on the number of shadow projects, the number and frequency of BMD tests per project, and how the samples are prepared and tested are discussed in the draft implementation guide.

Task 6: Specification and Program Development

This task involves the development of the specifications and quality assurance program to be used for implementation. The information gathered from the previous tasks will be used for this effort, and decisions must be made about what aspects of the current specifications need to remain the same and what existing criteria should be relaxed or completely eliminated. Consideration should be given to how to adapt the acceptance of lots based on specification criteria with two-sided limits (e.g. lab compacted air voids) to test criteria that are one-sided (e.g. minimum cracking index and maximum rut depth). If the BMD test results are used for pay adjustments, then there must be confidence that the results are truly performance related, which goes back to the need for field validation experiments in Task 3. 

Once the new specifications are vetted by stakeholders, then it’s time to plan and conduct pilot projects. Pilot projects are necessary to evaluate the new QA program requirements under actual contractual conditions. Pilot projects will go through the typical bidding-contracting process with the new QA requirements and specifications applied, including BMD testing required as part of mixture design and acceptance. The number of pilot projects should be determined by the agency, typically starting with just a few projects in the first year, then increasing to involve additional districts/regions and contractors each year. Lessons learned from pilot projects should be used to assess and modify the specifications and QA program requirements on a yearly basis, with more frequent modifications for more acute issues.

Task 7: Adapting Training Programs and Lab Accreditations

As an agency begins rolling out pilot projects, it will need to conduct just-in-time training to cover the new testing requirements and specification changes in the QA program. Likewise, the agency also needs to establish new independent assurance requirements for labs conducting tests used for mix acceptance to ensure the technicians and equipment meet the requirements of the new procedures. The state’s ongoing technician training and qualification program will also need to be adapted to include the changes and a plan to recertify technicians who are already qualified but are not familiar with the new tests and requirements. Many states already have a proficiency testing program for their current quality assurance tests as a means to ensure technicians perform the tests correctly and to continuously evaluate test precision information. These programs should be expanded to include the new BMD tests.

Task 8: Initial Implementation

Prior to full implementation of the BMD requirements, it is essential that the agency communicate the changes and new requirements to both industry and DOT personnel. This technology transfer can be accomplished through webinars, face-to-face meetings and workshops, and can also be supported by implementation teams that help contractors and DOT personnel get ready for the changes, address issues and interpret specification requirements. The initial implementation should be viewed as a new beginning with an expectation that the QA program will continue to evolve toward a system that efficiently assesses mix characteristics that relate to long-term performance.