Modeling of Churning Machine Control by Experienced Operators in Industrial Butter Production Using an Artificial Neural Network

Abstract

The aim of this study is to develop an artificial neural network (ANN) model for the successful control of a butter churning machine in industrial butter manufacturing. To model the manipulation skill of experienced operators, a three-layered ANN model with variable selection by a forward selection method was employed. Using four inputs including cream flow rate, fat content, aging time, and cream temperature, the ANN model properly predicted the churning speed within the prediction error of 7.0%, compared with that of experienced operators.