Home » Uncategorized

13 Applied Artificial Intelligence & Internet of Things Industrial Use Cases in Retail Industry

“We used computer vision, deep learning algorithms and sensor fusion, much like you’d find in self-driving cars. We call it -Just Walk Out’ technology”


Much to the surprise of everyone in the retail industry, Amazon launched a new format of stores called “Amazon Go.” The most interesting feature of these stores is that shoppers will not have to check out at a counter with a clerk or even with an automated kiosk. Shoppers simply have to swipe into the store with an app, then walk out with their desired items – no checkout necessary. Amazon’s “Just Walk Out” capability is based on the twin technological fusion of IOT and AI technologies.

While “Just Walk Out” technology represents the future of what AI and IOT can do for retail, historically, retail has been a gateway for enterprise adoptions of IOT technologies. One prominent example is RFID’s, a key IOT variant which was introduced for real-time tracking and identification of items using SKUs on retail shelves or inventory items in trucks, rail freight or ships. As an early adopter of IOT technologies, retail can be seen as a leading indicator for what’s to come in the AI and IOT domains.

AI scientists are pioneering advanced strategies for cross-selling and up-selling that look to transform retail by analyzing market baskets and customer segments in such a way that will enhance the shopping experience for customers and maximize value for retailers. It can be definitively stated that the retail industry has a faster rate of adoption for AI and IOT technologies than other industry verticals.

Some unique possibilities offered by combining AI and IOT technologies follow:

  1. Clerk-less Stores: Extending the concept of “Just Walk out Technology,” we will see the emergence of completely automated stores, offering the convenience of checkout-free shopping made possible by the combination of RFID personalized trackers and AI, digitally–enabled vision powered by deep learning and sensor analytics (sensor fusion).
  2. Supply Chain Optimization: Retail is heavily dependent on efficient supply chains. AI combined with IOT has tremendous potential in this space. Here are two use cases:

Use Case B.1:  Real-time tracking of transport vehicles improves fleet or route utilization which in turn improves the replenishment schedule

Use case B.2:  Shipping and delivery lead time, especially for e-tailers, can not only be accurately predicted, it can also be optimized by AI algorithms which ultimately increases customers’ confidence in e-tailing outfits.

  1. Inventory Management Optimization: AI can be used to lower inventory costs by analyzing consumption data and making predictive inventory control decisions based on data run through AI algorithms
  2. Customer Experience Optimization: Data gleaned from RFID-based object trackers can be analyzed to know when an item is sold out and how fast it needs to be replenished so customers will be able to find what they are looking for or know exactly when it will become available again
  1. Customer Analysis and Segmentation: Given the availability of AI tools and techniques to micro-segment the customers into fine-grained segments offers retails to offer high levels of personalized items and services to different types of customers thereby increasing overall stickiness and increasing overall customer lifetime value.
  2. Marketing Campaign Management: AI can be used to personalize both traditional and digital marketing campaigns by running analyses on customer preferences and personas which will increase the likelihood of marketing campaign success.
  3. Real-time Shipping Trackers: IOT sensors make it possible to track different products at various stages of their shipping journey. Real-time information from the sensors can be passed on to customers and retailers. This has been a boon for e-tailers because digitally savvy customers have come to expect insight into the progress of their order and this technology has let retailers kindly oblige.
  4. Humanoid Robots: Humanoid robots can be used to improve the in-store customer experience by speeding up interactions and personalizing the shopping experience for the customer e.g. Pepper, a humanoid robot that can interact with customers and “perceive human emotions.” Pepper is already popular in Japan, where it’s used as a customer service greeter and representative in 140 SoftBank mobile stores.
  5. Delivery and Transportation: AI-enabled drone and robot delivery systems have already been deployed by various retailers, but optimizing last-mile delivery seems to be the largest hurdle. Such robots and drones will use extensive AI and sensor networks to operate effectively.
  6. Computer Vision Quality Control: Computer vision can detect defects in materials and SKUs before stocking them in retail stores or warehouse shelves. Whether it is physical deformation or spoilage, AI-powered computer vision will enhance retailer image and make sure customers are only purchasing goods that meet acceptable standards.
  7. Efficient Floor Planning: Thorough analysis of floor plans and space design can help retailers and warehousers organize aisles and cross correlate with SKU level sales data to efficiently organize shelves to increase sales throughput and ease of access. AI-based planning and optimization routines can continuously monitor this process.
  8. Warehouse Management Optimization: A recent story emerged of how Alibaba managed to handle $21 billion USD in sales in a single day by using robot-enabled warehouses. AI and IOT robots save space and maximize scale and don’t get tired.

While these are just some of the AI and IOT use cases in retail, we expect more innovations currently being developed will soon be available on the market to disrupt current trends.