Data-driven predictive modeling of green concrete properties with fly ash: advancing sustainable construction through machine learning

The construction industry is a major contributor to environmental pollution due to its significant energy consumption and demand for raw materials. This results in the extensive use of natural resources. Addressing these issues requires advancements in construction processes and technology, focusing on innovation and digital transformation for sustainable futures. In recent years, there has been a growing adoption of cleaner processes in concrete production. This research aims to improve construction practices by employing soft computing techniques and machine learning (ML) technologies, including Gene Expression Programming (GEP), Artificial Neural Networks (ANNs), and Gradient Boosting (GB). These innovative digital tools are used to develop predictive models for estimating the properties of structural materials in green concrete, utilizing materials such as fly ash. The methodology involves a comprehensive literature review and the collection of a dataset comprising 239 data points. The data includes variables such as cement content, water, fly ash, fine aggregates, coarse aggregates, and curing age and is used to predict compressive strength. A sensitivity analysis assessed the impact of different input variables on the model’s output. The developed model provides a valuable tool for engineering practitioners to meet specific structural requirements for green concrete while ensuring structural integrity and sustainability. This research highlights the potential of innovation and digital transformation to enhance the construction industry’s capabilities and practices.