Utilizing machine learning approaches within concrete technology offers an intelligent perspective towards sustainability in the construction industry: a comprehensive review

This article provides an overview of the current applications and trends of machine learning (ML) techniques in concrete technology, focusing specifically on sustainability. The use of concrete and its several constituent components is one of the most important sustainability challenges facing the construction industry. In particular, there are serious risks to environmental damage, waste management, biodiversity, and public health associated with the industrial production of cement. Since biodegradable concrete is made from building and demolition waste rather than natural resources and has promising low-carbon potential, it is increasingly being identified as a suitable solution for these major issues. Physical design becomes more intricate when new solid waste resources are included, such as recycled aggregates, additional cementitious ingredients, and geopolymers. However, traditional models based on assumptions and linear regression are not adequate to assess this multi-level material systems’ performance. The research results show that ML models can optimise concrete mixtures in multiple ways while recognising multicollinearity, and they can also precisely evaluate the performance of sustainable concrete. Finding concrete competencies by ML modelling proved to be a time- and money-saving method that could still produce results with reasonable accuracy.