Over the past 20 years or so, the majority of state DOTs have transferred primary responsibility for quality control (QC) of asphalt materials over to the producers doing the work. States still run independent quality assurance (QA) tests to verify contractors' test results, but far more tests are run by the contractors.

Still at issue in many states, however, is the matter of which tests are used for acceptance and payment to contractors for completed jobs. “There are a number of states trying to get contractor tests used for acceptance, and there are a lot of compelling reasons to do that,” said Dave Newcomb, vice president of research and technology at the National Asphalt Pavement Association, Lanham, Md.

As state transportation budgets are cut back and the number of state technicians declines, contractors can pick up the slack, said Newcomb. And if states can quickly verify contractors' QC tests, “it would provide better feedback on issues like contractor payment,” he said.

Only a small number of states have exemplary programs for the use of contractor tests for acceptance, said John D'Angelo, asphalt team leader, Office of Pavement Technology with the Federal Highway Administration (FHWA) in Washington, D.C. “A lot of states need to improve their systems, and a lot of the problems have to do with how they analyze their data. A lot of states aren't running enough tests, but there are ways around that.”

FHWA lets states use contractors' test results for payment, provided the states verify that the contractors' results are representative of the actual material being produced. For example, if a state runs one test, the contractor runs four, and all tests compare within statistical limits, payment still should not be based on the contractor's four tests, D'Angelo said.

“We have codified that that is not an acceptable way to do it,” said D'Angelo. “You should have five to 10 tests by the state to compare to the contractor's tests. The state doesn't have to run as many tests as the contractor, but the state should combine several days' worth of tests to compare to the contractor's results.” That may take longer for contractors to get paid, but the matter can be resolved, he said.


D'Angelo cited Kansas and Georgia as two states that have what he called “reasonable” QC/QA programs. At the Kansas DOT (KDOT), the state's asphalt specifications use statistical processes to establish whether a contractor's test results will be used for pay—and also for pay adjustment.

Rick Kreider, assistant bureau chief for KDOT's Bureau of Materials and Research, explained the process. First, the state must determine if the contractor's and KDOT's test results come from the same population. That is accomplished through a statistical process known as the F&t test, which compares the variance and means between the two groups of tests—state's and contractor's. Twenty tests from the contractor are used for this comparison.

Next is the establishment of pay adjustments using percent within limits (PWL), a quality measure that evaluates how much of a given product meets the specification. PWL uses statistical values from test results to represent the total material placed within a lot. KDOT can then use the PWL to set incentive and disincentive pay adjustments for a given lot. To receive 100% of pay requires a contractor to achieve 90% PWL. If the PWL number exceeds 90%, then the contractor will receive an incentive up to a total of 7% for the combined pay adjustments.

KDOT specifies two pay adjustment factors in hot mix asphalt: air voids and density. For air voids, there are four contractor tests and one state test per lot. A lot is typically 3000 tons of asphalt mix, which is divided into four sublots (one contractor test per sublot). Density requires 20 contractor tests and 10 state tests per lot. A lot is typically one day's production, which is divided into 10 sublots. For each sublot, two contractor tests and one state test are run. All sampling follows a random selection process.