V3I1P83

Development of a Predictive Model for the Compressive Strength of Cement-Stabilized Lateritic Blocks Using Response Surface Methodology

Damini Righteous Gilbert^1*

Abstract

This study investigates the compressive strength development of lateritic blocks stabilized with cement contents ranging from 0% to 5% over curing periods of 1 to 100 days. A Central Composite Design (CCD) within the Response Surface Methodology (RSM) framework was employed to systematically explore the effects of cement content and curing age. Analysis of Variance (ANOVA) of the initial cubic model indicated overall significance (F = 6.27, p = 0.0184) with an R² of 0.9038, suggesting that 90.38% of the variability in compressive strength is captured by the model. Individual term analysis identified curing age (B), its quadratic (B²), and cubic (B³) terms as statistically significant (p < 0.05), while the effects of cement content were minor within the tested range. However, significant lack-of-fit (F = 609.48, p < 0.0001) and outlier patterns necessitated model simplification. A refined quadratic model retained curing age as the dominant factor, achieving an adjusted R² of 0.7596, standard deviation of 0.3326 MPa, and Adequate Precision of 10.002, confirming strong predictive capability. Experimental results demonstrated that blocks with 3–4% cement achieved peak compressive strengths of 3.82–3.88 MPa between 28 and 56 days, whereas higher cement contents showed diminishing returns. The final predictive model provides a reliable tool for optimizing block composition and curing schedules, supporting sustainable use of lateritic soils in load-bearing masonry while emphasizing the importance of rigorous statistical modeling for earthen materials.

Keywords:

Lateritic soil, compressive strength, response surface methodology, ANOVA, model diagnostics, sustainable construction