The project was commissioned by the Russian branch of an international French retailer.
The client is a leader in the Russian retail sector and a longstanding partner of ICL Services, having used the company's IT services and products for many years.
The self-service checkout (SSC) format is actively gaining popularity in the client's stores. It speeds up shopping and reduces queues without adding extra pressure on the staff. However, SSCs also lead to increased losses for stores due to scanning errors and theft.
The client had previously explored various ways to reduce losses. Tightening physical control negatively affected customer experience; hence, the client sought a digital solution that would be unnoticeable to shoppers.
The proposed solution was a smart video analytics system where a neural network, analyzing camera data, would alert security about anomalies at the SSC.
The ICL Services team was chosen for the project. In addition to the excellent test results, the client saw the advantage in the years of flawless collaboration with the company on checkout systems and the team's readiness to transfer rights to the developed software and hardware suite to the client.
Key Challenges
- Maximize the detection rate of anomalies during product scanning using a video analytics system.
- Develop and implement customer pathways together with the client, taking into account staff actions in the SSC area and security in the monitoring room and at the store exit.
- Introduce the system with minimal disruption to customer experience.
Implemented our solution
The project was launched in December 2022 at the client's two "pilot" stores in Moscow.
The initial step involved coordinating the solution architecture and IT infrastructure security requirements with the client. In line with the approved architecture, a network firewall was installed to safeguard client data, enabling us to simultaneously use video from the stores' own cameras and SSC data.
The project continued to progress across multiple fronts:
1. acquiring and analyzing video footage from the client's cameras;
2. video markup and AI training;
3. integration with self-service checkout software;
4. development of customer pathways.
During the implementation stage, the ICL Services team, in collaboration with the security service, determined the optimal settings for video surveillance cameras for both video analytics system and security service operations.
The system was launched in phases for different types of anomalies:
— initially, the AI was trained to identify discrepancies between the number of items in the cart and those on the receipt;
— later on, to detect non-payment;
— we implemented a system for identifying specific products, which allowed for the tracking of incorrect scanning and product misplacement.
A 'competition' between AI and actual experts was set up as the final acceptance test. During the test, AI was able to detect a substantial number of anomalies the staff had missed.
Following the pilot stage, we put together a strategy for future system development and technical enhancements.
Products and technologies
Computer vision
- Integration with self-service checkouts
- Firewall
- PostgreSQL database
Results
- The system is currently active in two hypermarkets, where it analyzes customer behavior at self-service checkouts in real time, helping to prevent losses.
- The solution reduces losses in the self-service checkout area without the need for additional staff costs.
- An analysis conducted by the client revealed that the implemented solution has no impact on customer satisfaction.