Prediction model of compressive strength for eco-friendly palm oil clinker light weight concrete: a review and data analysis

The increasing amount of waste materials from both industrial and agricultural sources poses an urgent danger to civilizations. Effectively managing these wastes through reuse or recycling is an important step towards reducing environmental harm. This article presents a comprehensive analysis of previous experimental endeavors that used palm oil clinker as a partial replacement for coarse aggregate in concrete, to assess its impact on the fresh and mechanical properties of the concrete. To understand the complex relationships between the data, data sets for training and testing are divided, providing the proposal of various models like linear, nonlinear, Gaussian progress, support vector machine, and ensemble boosting tree. The models are carefully evaluated using statistical characteristics such as the coefficient of determination, mean absolute error, root mean square error, and scatter index. After a comprehensive analysis, the best-performing and most accurate model is the Gaussian progress model, which has a scatter index of less than 0.1 and a coefficient of determination of 0.97 for testing data. The R2 value of Gaussian progress machine is higher by 17.3, 17, 3.4, and 0.7% compare to the R2 value of EBT, LR, NLR, and SVM model, the scatter index value of GPM also lower than SI value in EBT, LR, NLR, and SVM model by 158, 141, 51, and 13.7%This highlights how well it performs in comparison to other models and suggests that it could be a useful tool for predicting the effects of replacing coarse aggregate in concrete using palm oil clinker. For additional applications, statistical models can be presented that combine various types of lightweight aggregates to determine which will be more commonly used for individual or mixed applications.