Development of specification limits for asphalt pavements based on quality control and quality assurance data

Jian-Shiuh Chen, Min Chih Liao, Ching Hsiung Wang

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)


Asphalt content and aggregate gradation measurements for hot-mix asphalt (HMA) concrete were collected during the 2008 construction season to develop statistics for quality control (QC) and quality assurance (QA) programs for the Taiwan Highway Bureau (THB). Data were analyzed to determine if the quality characteristics followed a normal distribution in order to compare contractor and THB measurements and to develop specification limits. The quality test data were shown to follow a normal distribution; therefore, appropriate statistics could be developed from the normally distributed data. Differences between QC and QA test results were shown to be statistically significant for some of the mix properties. QA and QC comparisons indicated that QC data were less variable and tended to have more favorable test results that would give more favorable acceptance outcomes to contractors. This is a significant finding since highway agency pay factors assume the validity of QC data. A statistical process was established to determine a typical standard deviation value by taking into account the variability of both QC and QA data. The revised specification limits were developed as construction control tools for the quality characteristics of HMA mixtures. The revised specification limits were found to be loose enough to account for material, sampling, and testing variations but still tight enough to identify manufacturing and construction variability.

Original languageEnglish
JournalJournal of Testing and Evaluation
Issue number6
Publication statusPublished - 2011 Nov

All Science Journal Classification (ASJC) codes

  • Materials Science(all)
  • Mechanics of Materials
  • Mechanical Engineering


Dive into the research topics of 'Development of specification limits for asphalt pavements based on quality control and quality assurance data'. Together they form a unique fingerprint.

Cite this