Multiple Regression Model For Demolding Force Of Rubber Products Using A Full Factorial Design

Abstract

Demolding force is still unfamiliar even though it plays a crucial role in the gripper design process for automation systems in rubber product manufacturing nowadays. This study aims to determine how demolding force can be evaluated by using contact area and rubber hardness. A full factorial design was experimentally executed to examine the relation between contact area and rubber hardness on the demolding force while the Multiple Regression Model was used to estimate the demolding force. Analysis of the experiment demonstrated that contact area has influenced demolding force more than hardness by the p-value of 6.16 x 10-14 and 0.036 respectively. Finally, the multiple regression model that researchers developed has a correlation coefficient (R2) of 74.9 %, which indicates how much variation in the contact area and rubber hardness is explained by the demolding force in the multiple regression model.