Computer Vision for Retail Shelf Monitoring: Optimizing On-Shelf Availability
When you visit a grocery store and cannot find the product you are searching for, it can be quite frustrating. Making sure that products are readily available on store shelves is vital to increase profitability and customer satisfaction. When a desired product is not found where it is expected to be, known as “out-of-stock” (OOS), it can notably affect consumer behaviour negatively. Based on research done by Corsten and Gruen, when the product you are looking for is not present on the shelf where it is supposed to be then:
- 31% of people will go to another store to find it
- 26% will choose for a different brand
- 19% might choose a different size of the same brand
- 15% might postpone their purchase for later
- 9% may end up not buying anything at all
In 2021, the retail sector in the United States experienced major disruptions characterized by widespread empty shelves, marking it as a challenging year for several. A recent analysis conducted by NielsenIQ sheds light on the impact of on-shelf availability issues, revealing staggering figures.
“Empty shelves resulted in a staggering $82 billion in missed sales for U.S. retailers throughout the year. Weekly, “out of shelf” items cost retailers an average of $1.4 billion. The decline reached a record high of $1.75 billion in the third quarter of the year, showing the severity of the situation.”
Various studies have explored automating On-Shelf Availability (OSA) monitoring from diverse angles. Radiofrequency identification (RFID) tagging is used to track product quantities on shelves, but its implementation cost and integration into existing systems pose challenges. Recently, there has been growing interest in Computer Vision approaches to monitor OSA.
In this blog, we delve into the concept of how we can optimize Retail On-Shelf Availability monitoring using Computer Vision algorithms and their significance in retail operations.
What is Smart Shelf-Monitoring?
“Shelf Monitoring”, an advanced technology, offers real-time monitoring and analysis for retail stores. By employing AI cameras and image recognition software, this device can detect low stock, identify misplaced items, and even provide valuable insights into consumer behaviour and preferences.
Why Retail Shelf Monitoring?
Shelf monitoring is especially important today when customer expectations keep rising. Businesses need to keep an eye on their customer’s wishes and needs to stay relevant and successful.
- Improving Customer Experience: Customers expect to find what they are looking for when they visit a retail store. If the desired products are unavailable or hard to come by, it can result in a negative customer experience. Real-time Shelf Monitoring ensures that the products are adequately stocked and easily accessible.
- Optimizing Inventory Management: By monitoring the shelves, retailers can gain valuable insights into their inventory levels and make informed decisions about stocking and restocking products. This can help them in optimizing their inventory management processes, minimizing waste, and ultimately cutting costs.
- Identifying Trends: Retail Shelf-Monitoring can also provide valuable data and insights into consumer behaviour and preferences. By analysing the sales performance of different products, retailers can adjust their marketing strategies based on which products are selling well and which ones are not.
- Increasing Sales: When products are out of stock or difficult to find, customers may opt to purchase from a competitor instead. By ensuring that products are well-stocked and easy to find, retailers can increase their sales and revenue.
How Does Retail Shelf Monitoring Using Computer Vision Work?
This On-Shelf-Availability Monitoring with Computer Vision solution involves real-time monitoring of shelves with a video /camera device and image processing tool. Initially, the images of front-end products such as Surf Excel, Kellogg’s Cereals, etc., are captured and stored as reference images. These products are then arranged on the shelves. Subsequently, the image processing tool analyses the images from the video, detects and counts the number of products present.
Whenever a product is out of stock, the application immediately identifies this change and designates it as unavailable. In addition, if there are no products found on a shelf, an alert is automatically sent to the store manager, indicating the need for restocking.
Furthermore, this Smart Shelf-Monitoring system effectively tackles the issue of products being incorrectly placed on shelves. The application matches the stored product images with the captured shelf image to determine the presence of each product. An alert is generated when a product is not detected and is sent to a designated person or store manager. This Real-time Retail Shelf Monitoring aims to improve inventory management and guarantee optimal product availability.
What are the Next Steps of Development in Retail Shelf Monitoring?
- Real-Time Shelf Monitoring: Currently, most retail shelf monitoring systems operate periodically, with store employees manually checking the shelves at regular intervals. However, with the introduction of real-time shelf monitoring systems, retailers can now receive immediate updates regarding the status of their shelves, enabling them to promptly address issues such as out-of-stock situations or other concerns.
- Predictive Analytics: By analysing data from past sales and consumer behaviour, retail shelf monitoring systems can predict which products are likely to be in demand in future. This can assist retailers in optimizing their inventory management processes and ensuring that they consistently have the appropriate products in stock.
- Recommendations: With the help of AI and Machine Learning, retail shelf monitoring systems can analyse consumer data and provide personalized recommendations to shoppers based on their past preferences or based on similar purchase cohorts. This can assist retailers in enhancing the customer experience and boosting sales.
- Integration with E-Commerce: As more and more consumers turn to online shopping; retailers will need to find ways to integrate their shelf monitoring systems with their e-commerce platforms. Computer Vision can be used to keep an eye on stock levels and keep online product listings updated in real time.
Conclusion
Overall, the future of retail shelf monitoring is exciting. This is important in the current world as it enables retailers to enhance the customer experience, optimize inventory management processes, identify customer preference trends, and eventually boost sales. By leveraging Computer Vision algorithms, retailers can elevate their shelf monitoring practices and maintain a competitive edge. Shelf monitoring, through the provision of valuable data and insights, is changing the retail industry and enhancing the shopping experience for consumers. As technology continues to progress, we can anticipate witnessing even more groundbreaking advancements in this field.
Looking to ensure uninterrupted product availability and avoid stockouts in your retail stores? Contact us about how our Retail Shelf Monitoring solutions can help.