Empirical Model For Estimating Compression Index From Index Properties Of Fine Grained Soil

Abstract

Compressibility of soil is an important engineering property and is related to settlement of the foundation or structures. Compression index is one of the soil parameter that is required for analysis of foundation settlement, however, the determination of compression index (Cc) is expensive and time consuming. In order to mitigate these complexities, equations have been developed by several researchers to predict compression index using index properties of soil which are easier to conduct in the laboratory. This thesis illustrates prediction of compression index value from index properties of Mojo fine grained soils. After collecting 25 representative samples, laboratory index property tests consists of natural moisture content, Atterberg limit, free swell and one-dimensional consolidation tests were conducted. Accordingly the index properties results obtained were correlated with compression index using SPSS software. From the single and multiple regression analysis it was found that the multiple regression analysis made strong correlation (Cc = -0.253+0.004LL+0.004PL+0.003NMC) with coefficient of determination of R-square (R2) = 0.917 for a sample of 25. After validating the developed correlation with control test result, the correlation between the compression index and the remaining parameters on the other hand yield relatively weaker relationship compared with that of the LL, PL & NMC. Result of the validity of the newly developed correlation with control test results shows that, correlation of compression index value with soil index properties is valid only to estimate settlement for preliminary design purposes.