Square's System for Predicting Orders and Guiding Couriers

This 2018 patent from Square describes a system that predicts which restaurants will get orders and tells delivery drivers where to wait to be ready.

What it covers

This patent details a system designed to make food delivery more efficient. It analyzes past order data for many restaurants to predict which ones will receive orders during specific times and days. The system also estimates how long it will take to prepare those orders. Based on these predictions, and knowing where delivery drivers are, it tells drivers to move to specific waiting spots. This way, drivers can be closer to a restaurant just as an order is ready for pickup, reducing wait times for everyone.

What it doesn't cover

• —Systems that do not predict future orders based on past order information.
• —Systems that do not estimate order preparation times.
• —Systems that do not direct couriers to specific waiting locations before an order is ready.
• —Systems that do not consider the time of day or day of the week for predictions.
• —Systems that do not manage a network of 'active' couriers who are available for assignments.

The clever bit

The innovation lies in using historical order data not just to see what happened, but to predict *when* and *where* future orders will likely come from, and then proactively positioning couriers *before* the order is even placed, considering preparation time.

Why it matters

This patent is significant because it addresses a core challenge in the on-demand delivery industry: matching supply (couriers) with demand (orders) efficiently. By proactively predicting order volume and guiding couriers, it aims to reduce idle time for drivers and speed up deliveries, which is crucial for customer satisfaction and operational costs in services like DoorDash or Uber Eats.

Real-world examples

• 1.Square's own delivery services
• 2.Modern food delivery apps like DoorDash
• 3.Modern food delivery apps like Uber Eats
• 4.Modern food delivery apps like Grubhub