Predicting the Behaviour of Stabilized Lateritic Soils Treated with Green Crude Oil (GCO) by Analysis of Variance Approaches

The behaviour of the stabilized lateritic soil obtained from Amaoba, Nigeria treated with green crude oil was statistically studied using analysis of variance with interaction, Kruskal-Wallis test, and expected mean square methods to validate the effect of the additive on the strength properties of the stabilized soil. First a preliminary test was conducted on the soil to classify the soil as an A-2-6 soil according to the AASHTO classification system. The soil was treated with varying percentages of crude oil; 0, 2, 4, and 6% by weight of the soil. The results obtained showed that the GCO improved the strength properties of the treated soil. Finally the prediction model was used to validate the reaction that brought about strength gain, flocculation, carbonation, cation exchange and densification of the stabilized soil matrix. The three analyses of variance approaches agreed on the hypotheses tests conducted, which rejected the null hypotheses and showed that, to achieve soil stabilization, there must be interaction between the additives and the treated lateritic soil and different percentage by weight of treatment affect the treated soil in different ways.

Keywords:

Analysis of Variance , Geotechnical Engineering , soil stabilization , Kruskal-Wallis Test , Soil Strength Behaviour