Investigating the strength performance of 3D printed fiber-reinforced concrete using applicable predictive models

Abstract The construction sector is quickly adopting 3D printing because of its many benefits, such as the capacity to build complex geometries, speed up timeframes, increase sustainability, and improve safety. Making changes to the mixture composition of 3D-printed fiber-reinforced concrete (3DP-FRC) involves a lot of trial and error due to the many interdependent variables. In order to estimate the compressive strength (CS) and flexural strength (FS) of 3DP-FRC, the present study used gene expression programming (GEP) and Multi expression programming (MEP) for machine learning (ML). We ran a sensitivity analysis to go further into how important the input parameters were. Among the models, MEP had better predictive performance for FS and CS than GEP did, with R 2 values of 0.958 and 0.978, respectively. In contrast, the GEP model found lower R 2 values of 0.945 for CS and 0.928 for FS. Sensitivity analysis exposed that for CS, water-binder ratio, silica fume, and water content were the most influential parameters, while load distribution, sand content, and fly ash had the highest impact for FS. The developed ML models provide a reliable means of estimating the strength characteristics of 3DP-FRC for sustainable building design based on various input parameter values, offering significant time and cost savings compared to traditional laboratory testing.