Mix design optimization of waste-based aggregate concrete for natural resource utilization and global warming potential

Previous research has shown there is often a need to incorporate additional binders to maintain mechanical properties when incorporating waste-based aggregates in concrete mixes. This represents a challenge for those seeking to develop ‘green’ or environmentally sustainable concretes as incorporation of additional binders may increase global warming potential (GWP), however, using waste-based aggregates provides a way to avoid further consumption of natural resources. In this study, a mix design optimization framework is developed to find optimal mixes of concrete comprising different types of waste-based aggregates including concrete, slag, glass, plastic and rubber, considering trade-offs with global warming potential and natural resource minimization. Integrating a large database containing 5321 concrete mixes incorporating waste-based aggregates, artificial neural networks (ANNs) are used to predict concrete slump, compressive strength, splitting tensile strength, flexural strength, elastic modulus and water absorption. Genetic algorithms are then used to identify mix designs that minimize GWP and natural resource utilization. Having presented the framework, a series of scenarios are evaluated to investigate the impact of prioritizing minimization of natural resource usage or carbon dioxide equivalent emissions, and then the impact of transportation distance. The results show that the developed ANN model performs well in predicting different properties of waste-based aggregate concretes and the developed framework is capable of identifying mix designs that reduce carbon dioxide emissions by 50% as compared to similar mixes with an equivalent compressive strength in the database.