Abstract
APPLICATION OF ARTIFICIAL NEURAL NETWORK IN OPTIMIZATION OF SOAP PRODUCTION
Izabo Abigo*, Isaac E. Okwu and Barinyima Nkoi
ABSTRACT
This project was aimed at optimizing soap production profit using Neural Network programming. In this project a Neural Network was trained using a data set of solved linear programming problems. The objective function used in this training had two (2) variables and three (3) constraints equations. This trained Neural Network was used to optimize soap production profit for two different kinds of soap; bathing and laundry soaps. The Neural Network structure consisted of eleven (11) inputs and 3 outputs with a neural structure of 2 hidden layers and 50 neurons. The training algorithm used was feed-forward back propagation with a Bayesian Regularization error. The Neural Network results when compared with traditional simplex method of optimization, proved to be 98% accurate. The maximum projected profit was up to a 91% increase from ?30,625 to ?58,500 in a month. This research will increase the current rate of soap production in Patrich Global Enterprise and increase the profit of production while maintaining the same quantity of raw materials for monthly soap production.
[Full Text Article] [Download Certificate]