Statistical analysis of porcelain tiles’ technical properties: full factorial design investigation on oxide ratios and temperature

Abstract

This study focuses on optimizing the composition and firing temperature of porcelain tiles using statistical analysis techniques. A full factorial design, including model adequacy checking, analysis of variance, Pareto charts, interaction plots, regression model, and response optimizer is employed. The key factors were the Seger ratios of SiO2/Al2O3, Na2O/K2O, MgO/CaO, and firing temperature. The response variables investigated were bulk density, water absorption, linear shrinkage, coefficient of thermal expansion (at 500 °C), and strength. The statistical analysis revealed highly significant results, which were further validated, confirming their reliability for practical use in the production of porcelain tiles. The study demonstrated the effectiveness of utilizing Seger formulas and properties of typical raw materials to accurately predict the final properties of ceramic tiles. By employing SiO2/Al2O3 = 5.2, Na2O/K2O = 1.50, MgO/CaO = 3.0, and firing temperature of 1180 °C, optimized properties, such as maximum strength, maximum bulk density, and minimum water absorption, was achieved with a composite desirability of 0.9821.