Plastic waste (PW) has emerged as a global environmental concern due to its detrimental impact on ecosystems and human health. Traditional concrete heavily relies on natural aggregates like sand, gravel, and crushed stone, whose extraction leads to environmental degradation, including habitat destruction and resource depletion. Recently, the use of PW in concrete has gained attention as a sustainable alternative to these conventional aggregates. By incorporating PW as a partial replacement for natural aggregates, the construction industry can reduce its reliance on finite resources while also addressing the issue of PW. However, despite its potential environmental benefits, the incorporation of PW into concrete has primarily been explored through experimental studies, which are often time-consuming and resource-intensive. Therefore, this study aims to optimize the utilization of waste plastic in concrete through machine learning (ML) techniques, specifically Multi-Expression Programming (MEP) and Gene Expression Programming (GEP). A comprehensive literature review was conducted to compile a database for evaluating the compressive strength (CS) and tensile strength (TS) of PW concrete. The most influential parameters, such as plastic (P), gravel (G), water (W), cement (C), sand (S), and age (A), were considered as inputs in the models' development. The models developed were thoroughly evaluated using multiple statistical measures. Additionally, sensitivity analysis was conducted to discern and highlight influential factors that have a significant impact on the predicted outcomes. The findings indicate that both MEP (CS_R2 = 0.88, and TS_R2 = 0.89) and GEP (CS_R2 = 0.87, and TS_R2 = 0.88) models performed well, with MEP demonstrating slightly superior performance. Sensitivity analysis highlights the significant influence of cement (25.63 % and 24.53 %) and plastic (22.4 % and 23.44 %) on concrete strength properties. Furthermore, the equations provided by GEP and MEP models are simple to use from a practical perspective. Overall, this study contributes to sustainability efforts by promoting the incorporation of waste materials in concrete mixtures, thereby reducing reliance on cement.
Read full abstract