Predicting Wheat Supply and Demand: A Systematic Literature Review of Machine Learning Approaches

Wheat domestication and use are strongly related to humans’ attempts to control their food supply and fend off hunger. Therefore, the main problem is a structural supply and demand imbalance, driven by rising demand and factors limiting wheat output. This paper will review the literature that has worked on wheat supply and demand as well as the papers that study wheat prices. The previous studies will be reviewed by the type of analysis: descriptive, predictive, and prescriptive; then we will go deeper into the predictive studies to review the machine learning algorithm used for prediction, which will be divided into five categories: tree-based approach, artificial neural network, statistical-based approach, grey model, and hybrid approaches. Based on that, the gaps that previous studies have not addressed will be examined.