A Comparative Study of Prediction Model Case Study Commercial Cane Sugar Prediction

Abstract

Commercial Cane Sugar (CCS) is major concern in the sugar industry, impacting both production quantity and the trading systems of sugarcane farmers. The purpose of this article is to examine the factors related to CCS using the Stepwise Regression method. This research comprises two case studies: Case 1 focused on 13 factors related to sugarcane planting, while Case 2 investigates 17 factors associated with both planting and cultivation care. The study then incorporates various factors found to impact CCS levels to construct three forecasting models: Artificial Neural Network, Decision Tree, and Random Forest, together with a K-fold Cross-Validation. The findings reveal that Random Forest is the most effective CCS forecasting model, followed by Decision Tree and Artificial Neural Network.