Driving retail logistics efficiencies with AI warehouse management systems

“The order-to-dispatch time is a critical metric in retail operations, and shippers are working to reduce the time between order placement and the goods leaving the warehouse,” said Jacopo Bruna, who recently joined the Amazon Shipping operations team.

Prior to joining Amazon Shipping, Bruna worked in Amazon fulfillment centres for eight years and has seen the evolution of facilities from first-generation buildings to high-tech operations with integrated robotics and artificial intelligence. This front-row view into Amazon’s automation and technology investments has given Bruna valuable insights into fulfilment centre best practices to increase productivity and keep products moving.

“We are obsessed within fulfillment centres with minimising processing times for picking, packing, and shipping,” he explained. “Amazon is always innovating to improve shipping cut-off times and is leveraging tools for flow management and managing volumes across various stages of the production line.”

Bruna shared his knowledge of what is working for Amazon and how AI warehouse solutions could work in retailers’ warehouses and fulfillment centres.

Algorithms that optimise picking - the process of selecting inventory for shipment - are a valuable tool to help shippers prioritise those items that have to leave the warehouse first. “To minimise the order-to-dispatch time, you only pick those orders and items that need to depart right away. Picking algorithms do a dynamic and constant assessment of items that need to depart and can skip the less urgent items,” Bruna explained.

However, that doesn’t mean orders with two- or three-day delivery times have to wait, especially if there are low backlogs of orders and capacity is still available within the truck. In those situations, the algorithm can search orders geographically and prioritise those headed to the same destination even if they have different delivery speeds, helping to maximise space within the truck, reduce waste, and cut costs.

Sorting and packing items in warehouses are some of the most labor-intensive tasks, but the picking times must be perfectly balanced with packing times. “If you can pick 10,000 items per hour, you need to pack 10,000 items per hour,” Bruna said.

For example, Amazon’s AI algorithms and robotics streamline packing processes to sort packages more quickly and accurately. Robots powered by AI algorithms now handle much of the physical movement of goods in fulfilment centres, determining the most efficient routes for sorting and packaging items.

These AI-powered machines can handle a range of tasks, from placing items into boxes to determining the optimal way to fit items to reduce space and packaging materials, reducing the time it takes to prepare packages for delivery and minimising waste.

Amazon's robotic system Sequoia, for example, uses AI, robotics, and computer vision systems - technology that allows computers to analyse and understand visual data - to transport inventory, which reduces the time it takes to process an order through a fulfillment centre by up to 25%.

Packing algorithms also apply to the packing of trucks and trailers. The technology assesses orders, their destination, and the needed delivery times to consolidate orders within the same truck to reduce costs, optimise space utilisation, and improve sustainability.

As technology advances, sophisticated machine learning models, autonomous robots, and real-time data analytics are elevating warehouse automation and efficiency to new heights, which can play a key role in moving products quickly and efficiently from retailer warehouses into the first mile of delivery.

From first mile, let Amazon Shipping take over - arranging collections from retailers’ warehouses and safe delivery to their customers’ doorsteps. Learn more about getting started with Amazon Shipping here.