Multi-objective optimisation for mortar containing activated waste glass powder

Waste glass is inert and non-degradable which leads to enormous environmental sustainability troubles, but it can be reused in concrete due the potential of pozzolanic activity. This study proposes methods on activity excitation waste powder (WGP) including mechanical, chemical, mechanical-chemical activation. The results showed that mortar containing 30% 75 ?m WGP activated by method was optimal increase mechanical property reduce detrimental expansion. In addition, microstructural analysis conducted explore activation effect WGP-cement system. An artificial intelligence (AI) based multi-objective optimisation (MOO) model proposed seek mix proportions for unconfined compression strength (UCS), alkali-silica reaction (ASR), cost. A comprehensive dataset investigated 549 specimens UCS test 366 expansion test. Random Forest (RF) utilized prediction ASR values with hyperparameters tuned a firefly algorithm (FA). high correlation coefficients (0.93 0.91 ASR) verified feasibility FA-RF. Subsequently, FA-RF extended as objective function mi-objective (MOFA-RF) obtain consequent Pareto fronts. paper combined experiments, machine learning prediction, design mortar, provided basis practical application.