FoodZilla is the most advanced food waste technology on the market which is enabled with ML to maximize operational efficiency and data accuracy, reducing food waste. First, restaurants will save money by using Machine Learning to predict demand and waste from the past. We use data logged and seasonality prediction from previous months to predict both the demand and the wastage in the upcoming months. Using computer vision and object recognition techniques, we keep track of how much food is unused at the end of inventory periods. We then advise the restaurant owner to order an adjusted value. We also track the chopped vegetables/fruits by the restaurants which are wasted by them. For this, we take a Camera which is above the trash can which recognizes which item gets wasted. Second, the platform gives people who need food the most and restaurants a way to connect when their inventory reflects the wastage. When restaurants have excess food, it will post on the website where anyone can reach out to the restaurant to pick up the food at an agreed-upon location. Our solution reduces food waste two-fold, by using Machine Learning techniques to intelligently predict future orders, and in the case of wastage, provide a much-needed platform for restaurant owners to distribute their food.
Since our product is one such that AI has entered professional kitchen at scale. Hence, it was a challenge to gather the dataset of the restaurant's past inventory in order to predict the future stock of the restaurant. So we simulated the same for time series analysis by generating random values for stockings with respect to a particular time frame.