Comparing the efficacy of GEP and MEP algorithms in predicting concrete strength incorporating waste eggshell and waste glass powder

• The present study used the techniques of gene expression programming (GEP) and multi-expression programming (MEP) to assess the compressive strength (CS) and flexural strength (FS) and develop predictive models of sustainable mortar modified with waste eggshell powder (WEP) and waste glass powder (WGP) as a replacement of cement. In order to get more insights into the impact and relation of raw components on the CS and FS of a developed sustainable mortar, a comprehensive study using the SHapley Additive exPlanations (SHAP) methodology was performed. When comparing the efficiency of both employed models, it was seen that the MEP model exhibited superior performance with an R2 value of 0.871 and 0.894 for CS and FS, as compared to the GEP model, which had an R2 value of 0.842 and 0.845 for CS and FS respectively.

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Materialteknik (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Materials Engineering (hsv//eng)

TEKNIK OCH TEKNOLOGIER  -- Materialteknik -- Annan materialteknik (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Materials Engineering -- Other Materials Engineering (hsv//eng)

Nyckelord

Cement mortar

Eggshell powder

Glass powder

Machine learning

Prediction models

Operation and Maintenance Engineering

Drift och underhållsteknik

Publikations- och innehållstyp

ref (ämneskategori)

art (ämneskategori)