On-site concrete production is the most common kind of construction worldwide. Concrete, a composite material made by combining fine aggregate with cement coarse aggregate, and water, is a common building material. When it comes to tensile strength, ductility, and fracture resistance, plain concrete ranks quite low. Steel Fibre Reinforced Concrete (SFRC) has been the subject of intensive study because of its many advantages over standard concrete. The tensile and flexural strengths, as well as the ductility, of standard concrete are greatly improved by the inclusion of randomly scattered short fibres. However, the advantages can only be maximised with a well-thought-out mix design. Despite several academics' efforts to develop a standard approach for the design of steel fibre reinforced concrete (SFRC) mixtures, no such thing exists as of yet. Steel fibre reinforced concrete mixes may not be acceptable for the traditional ways of designing concrete mixes due to the increased water content required to achieve the same workability as without the inclusion of fibres. Fibre material, volume % of fibres, fibre aspect ratio, ratio of fine aggregate to coarse aggregate, aggregate - cement ratio, water-cement ratio, etc. all have significant impacts on the strength and workability attributes of fibre reinforced concrete. The purpose of this study was to create a new machine learning approach to predicting the workability and strength attributes of SFRC blends using a genetic algorithm (GA) and Artificial Neural Networks (ANN).

Steel Fibre Reinforced Concrete (SFRC), Genetic Algorithm (GA), Artificial Neural Networks (ANN), fibre material