Amazon optimises packaging with AI

Instead of corrugated cardboard boxes, more than 50 per cent of European Amazon shipments are now sent in smaller, recyclable shipping packaging, such as paper bags or envelopes. The aim is to further reduce the amount of packaging with the help of artificial intelligence and machine learning.
Amazon customised packaging Amazon customised packaging
(Image: Amazon)

Instead of corrugated cardboard boxes, more than 50 per cent of European Amazon shipments are now sent in smaller, recyclable shipping packaging, such as paper bags or envelopes. Since 2019, the company has also sent 700 million consignments without any additional packaging. And with the help of artificial intelligence and machine learning, the amount of packaging is to be further reduced.

Items that do not require additional Amazon packaging are suitable for dispatch in original packaging. The products are dispatched in the manufacturer's original packaging and are only labelled with an address sticker. This not only reduces waste, but also also makes consignments easier, which in turn reduces the delivery emissions per parcel and the need for recycling. For items that require additional packaging, Amazon uses paper bags and cardboard shipping bags, which require less packaging material, create less empty space in each parcel shipped and minimise the need for recycling. are up to 90 per cent lighter than cardboard boxes made of corrugated cardboard of comparable size.

„We use machine learning and artificial intelligence to reduce packaging and be more sustainable for our customers. The Operations Innovation Lab in Italy illustrates these efforts.“

Pat Lindner, VP Mechatronics and more sustainable packaging at Amazon

Display

In the Operations Innovation Lab near Milan Amazon is testing the latest packaging technologies for Europe. The lab has special technologies such as a innovative labelling robot, which automates the application of address labels for products that can be shipped by Amazon without additional delivery packaging, as well as an automated machine that produces customised packaging. This machine is also used in Amazon's logistics centre in Mönchengladbach, where it has already packed thousands of items.

In recent years, Amazon has also significantly reduced the amount of packaging with the help of artificial intelligence and machine learning. These technologies help to identify products that can also be delivered safely in lighter, smaller packaging. Packaging Decision Engine is an AI model, which helps to determine the most efficient packaging options for shipping millions of items. The model was trained by data scientists, to understand a variety of product attributes, such as the shape and shelf life of an item, and to analyse customer feedback on the performance of different packaging options. The model is constantly learning and helps to reduce packaging consumption.

Amazon is also testing across Europe a new, recyclable, paper-padded envelope. The envelope is made entirely of paper and has been optimised in terms of weight, flexibility and robustness. It is much lighter than cardboard envelopes and boxes, is easier to fill without wasting space and requires less CO2 during production and transport. Because if the parcels are the right size, Amazon can load more parcels into each van.

Customised automated packaging

Amazon has also developed new machines that produce recyclable, customised packaging on demand, reducing the volume of packaging for thousands of everyday items. With the help of an integrated sensor, the machine, which also in the logistics centre in Mönchengladbach The system is used to pack items such as video games, kitchen appliances, sports equipment and office supplies that were previously shipped in cardboard boxes and cardboard packaging. It then cuts the paper bag from paper rolls in real time to pack the items securely. The paper bags are securely sealed using heat-sealing technology. This allows the machines to pack quickly and accurately while minimising empty space. No adhesive is required to seal the packaging, which further reduces the consumption of resources. By packing the items in recyclable lightweight paper packaging, which is customised without padding, the machines avoid an average of more than 26 grams of packaging per consignment.

Improvement of address labels

In addition to reducing packaging and improving recyclability, the company is also investing in innovative labelling technologies. For example, Amazon is testing a new Universal Robotic Labeller, an automatic high-speed labelling machine that can Optimised placement and adhesion of address labels on products shipped in original packaging. With the new machine, address labels can be automatically applied to smaller and unusual packaging. This saves material, reduces costs and increases speed and efficiency. Amazon is also testing a new machine in Europe and the USA. New Easy Peel Label technology, which enables customers to remove the shipping labels from their original packaging after delivery.

SourceAmazon