Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{198824,
        author = {Anish  Sapkal and Aman Patre and Abhishek Saini and Vibha Patel},
        title = {Smart Anna: An AI-Powered Intelligent Cafeteria Management System Using YOLOv11-Based Real-Time Occupancy Detection and Digital Ordering},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {12189-12203},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=198824},
        abstract = {---Cafeteria operations in university pose several problems like crowding of cafeteria at peak time, long wait times of students, payment process using cash, and lack of use of data analytics for management of the cafeteria by the cafeteria staff. Traditional methods use only human observation and manual methods of management. There are ample opportunities for improving operations using advanced technology. In this paper we introduce Smart Anna which is an ecosystem for managing University cafeteria using edge computer vision, cloud computing and a responsive web interface, providing following functionalities: Real Time Occupancy Monitoring of Cafeteria, Digital Food Ordering and payment using UPI, and Dashboard for analysis of operations by the admin. Two specific models for human head and table availability detection using YOLOv11 object detection model are trained, deployed to an edge device and used to detect objects in a video stream from the cafeteria in real time without uploading any video data to the cloud thus preserving privacy. Count data is generated once every second and transmitted to a Fast API based back end server that processes this information and computes the occupancy status that is served to a react based student user dashboard that polls this server once every 3 seconds. Both models perform at around 29 fps in real-time detection of human heads and tables resulting in mAP@50 performance of 86.4% and 88.2%. The end-to-end latency from occupancy change to dashboard update is under 4 seconds. Usability study performed over a period of 3 days involving 15 students volunteers showed that 80% of subjects avoided unnecessary cafeteria visits and all found the UPI based payment process easier than cash based transactions. Our work shows that AI powered smart campus solutions are feasible from technical and financial perspectives and can provide significant benefits to students and other stakeholders in universities.},
        keywords = {YOLOv11, Computer Vision, Object Detection, Real-Time Occupancy Monitoring, Cafeteria Management, Edge AI, FastAPI, React, Supabase, UPI Payments, Smart Campus, IoT, Deep Learning},
        month = {April},
        }