Improving operations and product availability at stores globally

Decathlon StockBot robot by PAL Robotics at Singapore store (image: PAL Robotics)

Decathlon is a family-owned company, founded in 1976 in France, and inspired by the belief that sports and the outdoors should be accessible to everyone. They focus on ever-evolving and high-performance design, the best quality at the lowest possible price, and expert advice that customers can access in-store or online. 

The company manages the research, design, production, logistics, and distribution of its products including sports equipment and clothing in-house; as well as partnering with global suppliers. 

The company aims for an optimised supply chain and minimal retail footprint and usually favours large superstores averaging 4,000 m2 in size and reaching 9,000 m2 in larger stores. Optimising the supply chain also contributes to sustainability - by being able to reduce logistics efforts, lessening the need for unnecessary movement of stock between sales points, and stock optimisation by cutting down on overstocking of products.

In retail stores such as Decathlon, radio-frequency identification (RFID) is often used to provide accurate data on a store’s inventory or stock, as the technology transmits data from RFID tags on products to readers. Complete RFID solutions are made up of different components that work together: tags, readers, and antennae, as well as supporting software.

Automating the RFID processes at Decathlon stores, by introducing AMR robots by PAL Robotics as the RFID solution would be the challenge here. The AMRs would need to work effectively in a dynamic environment, with moving obstacles, including people, in stores during opening hours.

With the deployment of RFID robots, Decathlon wanted to take the opportunity to try to automate the counting of stock - since stock counting manually is a less repeatable process, prone to errors, and dependent on the performance of team members. They also wanted to help the team be more available to advise customers and not work on non-added value tasks such as counting stock. 

PAL Robotics’ StockBot solution is an Autonomous Mobile Robot (AMR) designed for inventory tracking and data collection in retail. StockBot is equipped with RFID antennas and readers so that the robot is able to move autonomously in stores to detect the tags that retailers use on products, then create the full inventory plus the location of the items in the store, which is not possible when stock taking manually with hand-held devices. StockBot is easily deployed within stores as it does not require any layout modifications and is integrated with existing software systems. The solution is typically set up in each store in one day.

At Decathlon stores, RFID robots perform SLAM – simultaneous localization of their own location and mapping of any area. SLAM includes Spatial AI, the spatial reasoning that allows robots to operate in general ways where they are, and helps them to understand their environment, such as the store or warehouse, better. This includes environment understanding by sensor fusion (laser + 3D cameras), analysis of obstacles and finding the optimal way to avoid them, as well as automatic generation of new maps of local regions where important changes are detected.

Inventory management assisted by StockBot allows the creation of demand forecasts based on data, and the AI used learns more and it constantly improves its forecasts to become more efficient. In this case, the robot integrates with the Inventory Management software used by Decathlon.

The AMR StockBot was customised in order to be introduced at Decathlon stores globally. The StockBot solution was initially deployed in Decathlon, Singapore. Following the success of this trial, the solution was chosen for roll-out in stores across multiple countries worldwide, including France, The Netherlands, Spain, Portugal, Poland, Ireland, Italy, Taiwan, and Australia.

StockBot is programmed to automatically read all of the inventory each day (daily inventories) so that the next day the teammates have the stock counted with data available on their smartphones, and are able to check for any discrepancies. In Decathlon stores with the solution, StockBot is considered a new teammate that assists workers and improves the capability to have stock available, as this is equal to huge benefits for the customer. 

Customer benefits 

Key benefits for Decathlon stores have been to conduct more frequent inventories in order to gain better visibility of stock, and of course, to optimise operations, resources, and costs.

The implementation of StockBot allows the company to increase the frequency of inventories as the robot is in operation continuously. The implementation at the store in Singapore enabled the team to finish inventories in a space of 5000 m2 within 5 hours, compared to having 80 teammates working a 12 hour shift. Janice Morin, Sports Leader for Trail Running in Decathlon Singapore explained, “before we had StockBot stock-taking work was very laborious, it was very manual and we allocated a lot of time to do our inventory. Now when we finish all the inventory we can allocate our time to do other things.”

As well as improving in-store operations, the retailer also aimed for better product availability rates. Mathieu Blanchard, Technical Director of Decathlon South East Asia said, “for any retailer, ensuring that customers are visiting your store and will find the products they are looking for is crucial.”

In addition, all the time and costs that employees spent on taking inventories can be spent on other more value-added tasks such as direct customer service, which, as a result, leads to a better and more complete customer experience.

Making decisions with the data provided by automation with a robotic tool means improving the efficient management of products, reducing costs, and reducing response time to product demand. The StockBot solution also helps Decathlon stores to determine easily and efficiently, a comparison of item sales depending on product placement in-store.

Additionally, the use of StockBot has helped Decathlon stores to maintain the success of facilities they offer such as ‘click and collect’ services, by identifying the products in the store to help prevent frustrastion for customers over stock discrepancies or outdated stock information, and reducing overstocking in stores where the demand is lower in comparison with other locations.

AI and digitalization have contributed to the success of the solution at Decathlon through a number of features:

• Robot navigation (wheel odometry) & environment perception (cameras RGB-D) 
• Traceability tags on products
• Virtual shielding (RFID parameters) in-store
• Tags positioning (triangulation) to track product location

The future

PAL Robotics and Decathlon are discussing potential additional collaborations - not only with inventory tracking and data collection through StockBot, but also RFID use cases in intralogistics.

PAL Robotics continues to respond to requests for the roll-out of StockBot in Decathlon stores in additional countries in Asia and Europe, as well as to increase the number of robots in stores where StockBot is already deployed. Thanks to the partnership with Decathlon, PAL Robotics also is able to work on developing further capabilities on StockBot.