The use of generalized models for determining the behavior of geological materials is difficult because of three complications. Firstly, the stress-strain behavior is highly complex. Not only is it nonlinear, but it is also very dependant on stress history. Secondly, adequate measurements of many soil property values are difficult because of disturbance. Therefore, values measured on samples may not be representative of values in situ. Thirdly, any natural deposit is likely to be extremly variable. These complications made it difficult to decide whether discrepancies between predicted and observed behavior are a consequence of inadequacies in the model or as a result of parametric values being determined from non representative samples. To be of any real value, the models for the behavior of soil and rock need to be formulated so that numerical values are required for only a few parameters; these values can be determined with single test procedures; and, the variability of the material at site can be assessed at resonable cost.

  • Availability:
  • Supplemental Notes:
    • Refer to ASCE Proceeding Paper 12243 by IS Sandler, FL Dimaggio and GY Baladi, July 1976.
  • Corporate Authors:

    American Society of Civil Engineers

    345 East 47th Street
    New York, NY  United States  10017-2398
  • Discussers:
    • Pender, M J
  • Publication Date: 1977-7

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00165894
  • Record Type: Publication
  • Report/Paper Numbers: ASCE 13054
  • Files: TRIS
  • Created Date: Mar 29 1978 12:00AM