Statistical and neural network assessment of the compression index of clay-bearing soils


ÖZER M., Isik N. S. , ORHAN M.

BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT, vol.67, no.4, pp.537-545, 2008 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 67 Issue: 4
  • Publication Date: 2008
  • Doi Number: 10.1007/s10064-008-0168-8
  • Journal Name: BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.537-545
  • Keywords: Artificial neural networks, Clay soils, Compression index, Regression analysis, FUZZY MODEL, PARAMETERS, STRENGTH, PREDICTION, EQUATIONS, BEHAVIOR
  • Gazi University Affiliated: Yes

Abstract

The compression index is used to estimate the consolidation settlement of clay-bearing soils. As the determination of compression index from oedometer tests is relatively time-consuming, empirical equations based on index properties can be useful. In this study the performance of widely used single and multi-variable empirical equations was evaluated using a database consisting of 135 test data. New empirical equations were developed utilizing least square regression analysis. In addition, an artificial neural network (ANN) with eight input variables was also developed to estimate the compression index. It was concluded that ANN provides the best results.