In today’s modern world, people are busy and food is mandatory for them to live. Food service providers help people by offering them good food. But they face a problem with inventory forecasting. Food inventory demand forecasting is a key component to food service providers. Food companies are more concerned with products having a short shelf-life and seasonal changes. The demand may depend on many hidden contexts and seasonal changes which remains to be non-trivial. In this paper we present an ensemble learning approach that employs dynamic integration of regressor for better handling of fluctuations in consumer demands. Demand forecasting is the process in which historical data is used to estimate the quantity of food item customer will purchase based on location, cuisine etc., It can be nearly impossible to have the required number of raw materials on hand for the preparation of food at any given time without proper demand forecasting processes. A food delivery service must deal with a lot of raw materials which are likely to decay, so it is important for such a company to accurately forecast daily and weekly demand. More inventory in the warehouse means more risk of wastage, and not enough inventory leads to out-of-stocks and push customers to seek solutions from the competitors. This decision tool is developed using Streamlit where the demands needed by the customers is forecasted using ensemble method named Random Forest in Machine Learning models, evaluated by Root Mean Squared Logarithmic Error (RMSLE) and visualized the forecasted data using Power BI.