Image-aided element shape generation method in discrete-element modeling for railroad ballast

Hai Huang, Erol Tutumluer

Research output: Contribution to journalArticlepeer-review

55 Scopus citations


The discrete-element method (DEM) provides more realistic results compared with those of continuum analysis for unbound construction materials such as railroad ballast. This paper introduces a digital-image aided particle-shape generation method for DEM to consider the shape effects of aggregate particles on assembly behavior. The BLOKS3D DEM program is used in this research with the capability of representing angular aggregates as discrete elements with user-defined particle morphological properties from imaging-based shape indices. A three-dimensional (3D) image analysis approach is available to construct discrete elements with shapes close to the morphological properties of actual aggregate shapes. A large-sized shear box is then used in direct-shear tests to validate this image-aided particle-shape generation DEM. The purpose of validation is to match the laboratory test results with the DEM simulation results by using one single set of model parameters. To that end, the sensitivity of DEM model parameters are investigated by conducting DEM shear-box simulations using different combinations of model parameters realistically chosen on the basis of previous research studies. The validation process is finally accomplished by statistically demonstrating that the DEM shear-box simulation results using one set of parameters can predict the laboratory shear-box test results reasonably well under various normal stress levels.

Original languageEnglish (US)
Pages (from-to)527-535
Number of pages9
JournalJournal of Materials in Civil Engineering
Issue number3
StatePublished - 2014

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Building and Construction
  • General Materials Science
  • Mechanics of Materials


Dive into the research topics of 'Image-aided element shape generation method in discrete-element modeling for railroad ballast'. Together they form a unique fingerprint.

Cite this