Estimation of strength, rheological parameters, and impact of raw constituents of alkali-activated mortar using machine learning and SHapely Additive exPlanations (SHAP)

Machine learning-aided catalyst screening and multi-objective optimization for the indirect CO2 hydrogenation to methanol and ethylene glycol process

“…31 For example, Kronberg et al 32 pointed out that SHAP provides robust and accurate feature attributes, making a significant contribution to understanding black box machine learning models and studying the basic structure–attribute relationships in data. Nazar et al 33 also explored the influence of various input parameters of alkali-activated materials on their prediction results through the SHAP method, and found that hydraulic lime has a strong interaction and positive impact on the yield stress and plastic viscosity and PV, while a negative effect on compressive strength. Besides, Liu et al used the SHAP interpretable model to analyze the features of the supercritical water gasification of coal, and believed that the influence of temperature and residence time is the highest on the gas production rate.…”

Comparative study of eco-friendly wire mesh configurations to enhance sustainability in reinforced concrete structures

“…Apart from the natural resources, plain concrete has several challenges such as low tensile strength, ductility, high porosity especially in severe service conditions, and resistance to crack propagation 10 . Recent and past studies took several approaches to improve concrete properties, resulting in quite different materials 11 – 15 . Serviceability criteria such as excessive crack width and deflection impair the appearance of the structure, weakening the member due to corrosion of steel and damaging non-structural members become more critical than the strength consideration 16 .…”