JJCE Submission Home

Strength Modelling Of Soil Geotechnical Properties From Index Properties

Last Update2018-09-12
TitleStrength Modelling Of Soil Geotechnical Properties From Index Properties
Author(s)Author #1
Author title:
Name: Olumide Aderemi Afolabi
Org: Graduate Student, Civil Engineering Department, University of Ibadan, Ibadan, Nigeria

Author #2
Author title:
Name: Olaniyi Diran Afolayan
Org: Lecturer, Civil Engineering Department, Covenant University, Ota, Nigeria.

Other Author(s)
Contact AuthorAuthor #2
Alt Email:
KeywordsStrength models, Regression models, Linear and non-linear relationships
AbstractThis research work presents strength models developed for the class of soils encountered. An empirical, analytical model is developed to predict the CBR and shear stress of soil from its index properties and grade size, with a view to reducing time, effort and cost usually incurred in determining these tests in the laboratory for future planning, design and construction projects. Soil samples were collected from various locations in Ife central local government. Various available index property tests, such as sieve analysis, Atterberg limit test and specific gravity test were carried out and classification of samples performed. Compaction test, California Bearing Ratio test and triaxial test were also carried out with the Optimum Moisture Content (OMC), Maximum Dry Density (MDD), unsoaked California Bearing Ratio (CBR), internal angle of friction φ, cohesion c and shear stress determined. Regression models related to these index properties together were developed and tested to ascertain their effectiveness. The study showed that about 44.4% of the soil mass of the Ife central local government is poorly graded soil with gravel, followed by 33.3% of well graded soil with gravel. Linear and non-linear relationships were generated between various soil indicies and engineering properties through correlation analysis with a reliable coefficient of determination (R). Poorly graded soil with gravel cannot be effectively correlated because of its weak coefficient of determination.
Paperview paper 4469.pdf (222KB)