The use of artificial intelligence (IA) in a delivery operation leverages not only logistics, but the entire ecosystem – restaurant, delivery person and consumer. At iFood, AI permeates different areas, from the way data is measured internally to delivery planning. “Artificial intelligence has the potential to enter all areas and all possible processes, being a catalyst to add more value to our operations”, explains Benjamin Mariotti Feldmann, head of Logistics Development at iFood.
In FoodTech logistics, statistical and operational research models, at different levels of complexity, are used, for example, to predict how many orders should arrive during a shift, in a city, and how many delivery partners need to be available during the period.
According to Feldmann, mathematical models, which aim to optimize, also make it possible to predict the routing of orders and the allocation of delivery people. “This is just to name a few examples. With each passing cycle, we map more and more opportunities for artificial intelligence, with new models and new improvements constantly implemented.”
In relation to delivery, the expert explains that balancing the three pillars of iFood (restaurant, delivery partner and end consumer) is the main challenge. “Everyone has different demands – which can sometimes seem like a paradox. How can I deliver efficiently and quickly and, at the same time, reduce the wait time for the delivery person at the restaurant? Artificial intelligence comes in precisely to break down these apparent paradoxes, helping and optimizing each person’s levers. For example: having an accurate model of dish preparation time allows the delivery person to arrive at the exact moment the order is ready, without affecting the customer experience at the end, who still receives their food in a short time.”
