Food Enterprises- Profit Growth Rate Prediction Based on LSTM from the Perspective of the Supply Chain

Abstract

Food is the fundamental guarantee of people's lives, and the food industry has always occupied an essential position in the national economy. Profit growth rate, as a measure of an enterprise's development ability, can intuitively reflect the change in operating profit for food enterprises. The accurate prediction of profit growth rate can provide a decision-making reference for enterprises in planning business objectives in the next stage. However, many factors affect the profit variation of a company, and it is hard to make accurate predictions using traditional statistical economics forecasting methods. Since the Long-Short Term Memory (LSTM) model can capture nonlinear relationships in time series analysis, we propose an LSTM-based model to predict the profit growth rate of enterprises by using the operational data of four seasons ahead. Moreover, due to the COVID-19 pandemic, the impact of supply chain integrity on enterprise operations is increasing. We introduce the information of the supply chain owned by the enterprise to predict the profit growth rate of the enterprise. The result of our model exhibits high prediction accuracy, which indicates that our model could provide practical guidance for companies' production and operation activities.