Utilising Machine Learning to Trend and Advise on Food Prices at Wholesale Vegetable Markets

There is a growing concern of food insecurity in Trinidad and Tobago, exacerbated by disruptions from the COVID-19 pandemic and the Russia-Ukraine war. With regional reports indicating that millions of people in the Caribbean are facing food insecurity, there is a pressing need for robust solutions to stabilise agricultural markets and food prices. Currently, Trinidad and Tobago lacks an advanced price forecasting system for agricultural commodities. The current National Agricultural Market Information System (NAMIS) used by the National Agricultural Marketing and Development Corporation (NAMDEVCO) monitors the prices but does not provide forecasting.

 

This research proposes the implementation of an AI-based price forecasting system to enhance NAMIS, focusing on three key commodities: limes, tomatoes, and lettuce. By leveraging machine learning techniques such as Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs), the system aims to account for both linear and non-linear factors affecting price fluctuations. Hybrid forecasting methods, which combine traditional statistical models with AI techniques, are recommended for their ability to improve accuracy in complex, volatile markets. The proposed system could significantly aid policymakers in crafting interventions to stabilise prices, support local farmers, and enhance food security in Trinidad and Tobago.