Product defect prediction model in food manufacturing production line using multiple regression analysis (MLR)

This paper aims to develop an improved general mathematical model by focusing on human factors variables that related to the product defect in the manufacturing production line. This is because many studies found that almost 40% of total defects resulted from the operator error and the defects are usually not obvious and neglected. The objective to have defect prediction mathematical model to satisfy as early quality indicator of the manufacturing flow production line and assist the quality control team in manufacturing industries. Thus, the human factor variables will be investigate thoroughly and final model can be used to predict product defect on the line to improve product quality. Product defects quantity are identified and analyzed to determine the potential predictors for developing the mathematical model. A case study is offered that illustrates in a spice packaging semi-automated production line the effect that complexity variables have on assembly quality. By using Minitab, Multiple Regression analysis is conducted to model the relationship between the input variables towards response variables. From the analysis, the predicted data showed reasonable correlation with the observed data improved with adjusted R-Sq from 2.6% to 7.9%. Hence, the regression equation obtain is selected to be the prediction mathematical model for defects based on human factor input variables.