Evaluation of FHWA Pile Design Method against the FHWA Deep Foundation Load Test Database Version 2.0

Nikolaos Machairas, Gregory A. Highley, Magued G. Iskander

Research output: Contribution to journalArticlepeer-review


The efficacy of the FHWA pile design method was explored using data made possible by the 2017 release of FHWA Deep Foundation Load Test Database (DFLTD) v.2. Information contained within DFLTD v.2 was leveraged to evaluate the most common pile design methods against failure loads obtained via in situ static load testing procedures. In the process, the authors developed a custom relational database and software to batch process the available information. The scope was limited to impact-driven, un-tapered, steel and concrete piles, loaded in compression, using a static load test. DFLTD v.2 contains 3,116 unique-combination project/exploration/foundation/test cases of which 213 contained sufficient data to permit batch processing of the results to compute the axial load capacity and interpret the static failure load, according to the study scope. Scatter between measured (interpreted) and predicted capacities is significant; the computed capacity was off by a factor of 2 in many tests. The range in calculated to measured capacities (Qc/Qm) was from 0.12 to 8.88, and the mean Qc/Qm was 1.48. Preliminary evaluation suggests that the method performs better in clay than in sand, and overpredicts the capacities of long and larger diameter piles. The authors trust that this study will permit engineers and state agencies to better understand the efficacy of the most commonly employed design methods, resulting in a more resilient infrastructure.

Original languageEnglish (US)
Pages (from-to)268-277
Number of pages10
JournalTransportation Research Record
Issue number52
StatePublished - Dec 1 2018

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering


Dive into the research topics of 'Evaluation of FHWA Pile Design Method against the FHWA Deep Foundation Load Test Database Version 2.0'. Together they form a unique fingerprint.

Cite this