How will IoT change the retail market? – Vision Campus


Welcome to the Vision Campus! The Internet of Things -or IoT for short- is the intelligent connectivity of physical devices through the internet. Today IoT already enables our homes to be networked in a smart way. In a few years, traffic will flow on the roads autonomously and even entire cities will be connected through IoT technologies. Of course, the Internet of Things will also
radically change the world of retail. New technical achievements, such as object
recognition or customer tracking with embedded vision system, open up new opportunities for
retailers to respond to customers’ wishes. Here is an example: Imagine you are a manager of a retail chain with 200 branches. Of course, you want your merchandise in your shelves to be always available in sufficient quantities and perfectly oriented. In addition, promotions should be presented in the most attractive way possible. To guarantee this ideal concept today in 200 markets, you need many personnel, which means enormous costs and there would still be no uniform results across all stores. This is why an IoT embedded vision solution could be a good idea for a retail manager. Such a system could monitor whether the shelves are filled optimally and the prices and promotional campaigns are correct. Of course, it is also possible to track the
customers in the stores and how they interact with the shelves. Do customers confidently reach for something on a shelf or are they inspired by a promotion? Do they quickly find what they’re looking
for? An IoT embedded vision solution also poses
challenges to the retailers. A critical aspect in the implementation of
such a system is the interaction between software and hardware components. The system architecture must be reliable,
high-performance and scalable to avoid reaching limits even in a growing company. In addition, IoT vision solutions, with all
its vision sensors, generate large data volumes that also create challenges in terms of data
transfer and data security. What does this mean for our retail manager? Before he sets up a new vision system for
his branches, he should be clear about which data they want to receive and how it should
be handled. Since our manager has opted for an IoT embedded
vision system for the stores, there are now two options for data collection and processing: either decentralized processing of data at the edge of the network
or processing of data in the cloud. Let’s say the manager wants to know how many
bottles are left on the shelves. For this, a picture of the shelf is not adequate; what is needed is the information from this picture, that is, the number of missing bottles. For this, the image from the camera must be processed and the required information extracted. Because sending full images to the cloud would require a large network bandwidth and slows down the system, it is more efficient to process
the images at the edge of the network and send only the extracted information. After the pre-processed information from all 200 branches has been sent into the cloud platform, services running in the cloud are
analyzing the information with big data algorithms according to the managers` requirements. Here a fully automated check is performed to see where the representation of the merchandise can be optimized at the branches. The data of the stores is then visualized
in the form of reports or dashboards. The manager gets the analysis in real time. This creates a data-based foundation for decisions on how to optimize the customers’ shopping experience. However, simple tasks for the branches, such as the restocking or orientation of the merchandise on a shelf can also be forwarded to the branch employees in an automated way. To summarize: in the long term, a IoT embedded vision solution is not just only very cost-efficient, it also offers an entire universe of new possibilities for business planning, forecasts and customer analysis. It thus contributes significantly to the success
of the business. Thanks for watching!

, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

Post navigation

Leave a Reply

Your email address will not be published. Required fields are marked *