This study presents a hybrid stacked deep learning model for forecasting potato consumption. The research aims to provide reliable predictions for potato consumption up to 2030, which is crucial for stakeholders in the food market, including farmers, sellers, and governments. The study employed various machine learning and deep learning models, including stacked long short-term memory (Stacked LSTM), convolutional neural network (CNN), random forest (RF), support vector regressor (SVR), K-nearest neighbour regressor (KNN), bagging regressor (BR), and dummy regressor (DR). The Stacked LSTM model demonstrated superior performance, achieving a mean squared error (MSE) of 0.0081, a mean absolute error (MAE) of 0.0801, a median absolute error (MedAE) of 0.0755, and a coefficient of determination (R²) value of 98.90%. These results indicate that the model can reliably forecast potato consumption. Potatoes are a vital food source, with a high nutritional value and widespread consumption. They are the third most eaten crop globally, with a significant role in many diets. The study highlights the importance of accurate forecasting for managing inventory, predicting food deficits, and ensuring a steady supply of food, especially in regions where potatoes are a staple. Environmental factors, cultural preferences, and dietary habits significantly influence potato consumption. Eastern Europe has the highest consumption rates, while some African countries have much lower rates. The study emphasizes the role of artificial intelligence in agriculture, including machine learning algorithms that help in decision-making and resource management. These technologies contribute to more efficient and sustainable agricultural practices.This study presents a hybrid stacked deep learning model for forecasting potato consumption. The research aims to provide reliable predictions for potato consumption up to 2030, which is crucial for stakeholders in the food market, including farmers, sellers, and governments. The study employed various machine learning and deep learning models, including stacked long short-term memory (Stacked LSTM), convolutional neural network (CNN), random forest (RF), support vector regressor (SVR), K-nearest neighbour regressor (KNN), bagging regressor (BR), and dummy regressor (DR). The Stacked LSTM model demonstrated superior performance, achieving a mean squared error (MSE) of 0.0081, a mean absolute error (MAE) of 0.0801, a median absolute error (MedAE) of 0.0755, and a coefficient of determination (R²) value of 98.90%. These results indicate that the model can reliably forecast potato consumption. Potatoes are a vital food source, with a high nutritional value and widespread consumption. They are the third most eaten crop globally, with a significant role in many diets. The study highlights the importance of accurate forecasting for managing inventory, predicting food deficits, and ensuring a steady supply of food, especially in regions where potatoes are a staple. Environmental factors, cultural preferences, and dietary habits significantly influence potato consumption. Eastern Europe has the highest consumption rates, while some African countries have much lower rates. The study emphasizes the role of artificial intelligence in agriculture, including machine learning algorithms that help in decision-making and resource management. These technologies contribute to more efficient and sustainable agricultural practices.