The construction industry significantly impacts the environment, contributing around 30% of global greenhouse gas emissions. Concrete production, a major activity within this sector, is a major contributor to the CO2 emissions, since concrete is the most used material in the construction world. Despite cement comprising only 10–15% of the concrete mixture by mass, it accounts for up to 90% of the related greenhouse gas emissions. The concrete production process is inherently complex, with numerous unknowns. As a result, cement is often overused in concrete mixes to enhance mechanical properties and ensure safety margins. However, reducing the amount of cement in concrete mix designs can have a significant environmental impact. This means that improving the accuracy in the prediction of concrete mechanical properties is crucial for minimizing cement usage while maintaining structural safety. In this study, machine learning algorithms are used to predict the mechanical properties of concrete, with the goal of reducing cement usage and its environmental impact. A dataset of approximately 1100 mix designs is emplyed and integrated with meteorological data from the concrete batching plants. The research focuses on regression and classification models to estimate concrete properties, considering environmental and meteorological parameters at the production site.