Estimation of Concrete Mixed Proportion and Concrete Compressive Strength Forecasting Employing the Simple and Multiple Linear Regression Models

Abstract

This study aims to 1) develop simple linear regression (SLR),develop multiple linear regression (MLR) equations, and 3) to validate the SLR and MLR models. The developed equations are used to predict both the concrete mix proportions and the compressive strength of concrete at 7 and 28 days of curing. A total of 29 concrete samples collected from various construction projects in Khammouane Province, and nearby areas were used to develop regression models. The dataset included key parameters such as compressive strength, maximum size of coarse aggregate, fineness modulus of sand, bulk specific gravity (SSD) of aggregates, water absorption of sand and aggregate, unit weight of coarse aggregate, entrapped air content, water content, cement content, water–cement ratio, slump value, and weights of the coarse and fine aggregates. These parameters were used to establish correlation equations for predicting concrete mix proportions and compressive strength. In addition, the properties of locally available materials were applied to verify the developed equations through laboratory retrial mixtures. The materials included mountain rock aggregate from Nakai Khia village and Mekong River sand supplied by Nonthakon Company in Thakhek District, Khammouane Province. Key aggregate properties such as specific gravity, water absorption, unit weight of coarse aggregate, and fineness modulus of sand were incorporated in the prediction process.

          The results indicate that the regression equations developed using SLR and MLR show high reliability and good predictive accuracy. These equations can effectively estimate concrete mixed proportions and compressive strengths at 7 and 28 days. Furthermore, laboratory retrial concrete mixtures showed good agreement with the predicted 28-day compressive strength, confirming the applicability of the developed models for practical concrete mix design and strength prediction.