PREDICTING THE COMPRESSIVE STRENGTH OF PALM KERNEL SHELL CONCRETE USING FULL FACTORIAL EXPERIMENTAL DESIGN

Abstract

Research has shown that the strength characteristics of Palm Kernel Shell Concrete (PKSC) are highly dependent on the mix parameters. This paper, through series of trial mixes, aim to achieve a model by which mix proportions can be selected for a target compressive strength using palm kernel shell as aggregate in concrete. This was done by testing the compressive strength of concrete containing varying total aggregate-cement content from ratio 3 to 6 in steps of 0.7 and water-cement ratios of 0.50, 0.55 and 0.60. The palm kernel shell content too was varied as a percentage replacement of coarse aggregate from 0 to 100% in steps of 10. From the trial mixes, the maximum, medium and minimum values of concrete mix variables were determined. A three-level three-factor full factorial design of experiments was used to investigate their effects. A total of 32 tests were carried out, including five control experiments (center points). A linear regression model was employed to model the measured compressive strength, this model predicts the compressive strength from a combination of water-cement ratio, palm kernel shell content and cement content. The model yielded a good prediction of compressive strength and provided good correlations with the data used in this study. The coefficient of determination, R-squared was 0.985, implying the model is highly reliable. Moreover, the proposed model proved to be valid in predicting the compressive strength of concretes whose mix parameters varied from the data used to develop the model.