Classification and Regression Tree Model to Predict the Probability of a Product being Backordered in Supply Chain

Abstract

Supply chain uncertainties pose a massive and ever-present challenge for modern companies. These uncertainties can manifest in two contrasting scenarios: supply surplus, where companies have excess items, and supply shortages, where there is an insufficient quantity of goods. Each situation demands a different approach from businesses to adapt to the varying outcomes and maintain a competitive edge in the market. Product backordering is one of the important things that companies need to deal with in an uncertain supply chain. A backorder occurs when a customer-ordered product or service is not in stock or cannot be supplied immediately, and the customer has to wait. Companies striving for a balance in managing backorders. Machine learning models can help to determine the probability of a product being backordered. In this research, we develop Classification and Regression Tree (CART) model that uses previously known parameters to predict the likelihood of a product being backordered. We also use different model parameters to evaluate the accuracy of the model.