The most popular construction material, concrete, is also a known pollutant that has a negative impact on sustainability due to resource depletion, energy consumption, and greenhouse gas emissions. As a result, in order to boost concrete's long-term viability, efforts should be focused on minimising its environmental effects. This study aimed to develop a prediction model for the compressive strength of various mixes in order to construct ecologically acceptable concrete mixtures. The concrete mixes that were employed in this study to construct our suggested prediction model are concrete mixtures that comprise both ground granulated blast-furnace slag (GGBFS) and recycled aggregate concrete (RAC). To forecast the compressive strength of eco-friendly concrete, the multivariate polynomial regression (MPR) white-box machine learning model was created. The model was contrasted with the other two machine learning models: support vector machine (SVM), a black-box machine learning model, and linear regression (LR), a white-box machine learning model. In terms of R2 (coefficient of determination) and RMSE (root mean absolute error) measures, the newly proposed model beats the previous two models and demonstrates robust estimate capabilities.