Numerical Optimization of the Input Factors and Responses of an Experimental Design Reinforce PVC Composite

Abstract

In this study, the input factors (PVC, rattan, plantain peduncle, and temperature) and the responses (flexural strength and modulus of elasticity) of an experimental design to reinforce PVC composite were investigated. The assessment of suitability and ultimately the choice of the models were done based on the values of statistical parameters like coefficient of determination (R2 value), p value, F value, etc. The experimental data was carried out by fitting a second-order model in the Design Expert model library to the data to estimate the unknown model parameters. This process was accomplished by applying multiple regression analyses to the data. The optimum values of the responses were obtained by numerical optimization based on the criterion of desirability. The results obtained showed that the design was adequate, and the models developed from it were useful because the off-diagonal values were close to zero. Also, increasing the proportion of PVC resulted in a steep and significant increase in the flexural strength of the composite. More so, this trend was similar to the flexural strength for the response surface and the corresponding contour plots representing the relationship between PVC, rattan, and modulus of elasticity. Besides, at higher temperatures, the PVC material could act as a binder, which results in a strong interaction between the components of the composite material.