Sodexo
Optimize Sodexo's payment process using computer vision
The Need
Optimize the payment process to reduce waiting time
Sodexo, a global leader in food services, serves millions of consumers daily across the world. Managing such a large-scale operation comes with numerous operational challenges, and the payment process remains one of the company's biggest pain points. How can we make the payment experience as efficient and seamless as possible for consumers? Together, we took on this challenge—and delivered a solution!
The Challenge
Create a system capable of quickly analyzing meal trays
Sodexo's challenge was to develop a system capable of quickly and accurately analyzing meal trays to speed up payments—this is where Computer Vision comes into play. The system must precisely identify various food items from different angles in real time, reducing wait times and enhancing operational efficiency.
Our Approach
Develop a computer vision tool, a dataset, and a labeling tool to train the models
To build this new system, we collaborated with our technology partner Deepomatic.
- We developed an innovative computer vision tool for the automated analysis of meal trays. This identification system optimizes the tray recognition process, enabling faster payment processing.
- We created a custom-annotated food dataset designed specifically to meet the requirements of meal tray analysis through computer vision. This dataset includes a vast collection of high-resolution images representing various dishes, beverages, and side items in different configurations, lighting conditions, and perspectives.
- Additionally, we developed a tailor-made labeling tool to streamline and optimize the process of generating training data for our computer vision inference models.
The Result
A significant reduction in payment processing time and efficient daily operations
The implementation of this technology has led to a significant reduction in payment processing time, while ensuring efficient daily deployment and evolution of the algorithms through Deepomatic’s platform. Moreover, transitioning to an Edge computing approach has lowered the ecological impact and enhanced user data privacy.
Integrate an Industry 4.0 platform to tackle Ynsect's data management challenges
AI to optimize raspberry aroma production through fermentation by Phytowelt
Modernize the IT System with Vertical Solutions
Acculturation to Generative AI: BDO Teams Train in This New Technology and Define Business Use Cases
Driving Eco-Counter's Growth: A Future-Ready Infrastructure