AI-Based Smart Food Ordering and Restaurant Management System with Multilingual Chatbot and Personalized Food Recommendations

Abstract

Online food ordering, restaurant communication, and kitchen stock management are commonly handled through separate tools. This fragmentation causes communication delays, inaccurate information about food availability, and limited support for customers who prefer regional languages. This paper presents the Smart Food Ordering and Restaurant Management System, a web application that combines online menu browsing and ordering, an AI-assisted chatbot for food-related queries, interaction in Tamil, English and Telugu, live kitchen inventory monitoring, rule-based personalized recommendations for children, gym users and customers with dietary requirements, a restaurant administration dashboard, and direct restaurant communication options. The frontend is built with React and Vite and the backend with Node.js and Express.js, with inventory events delivered over WebSocket. Inventory levels automatically classify menu items as available, low stock or out of stock, which prevents orders for unavailable items. The recommendation module ranks dishes with a transparent weighted score rather than a trained model. A working prototype was implemented, its principal interfaces are presented, and a set of test scenarios is defined. The results indicate that a single platform can serve both customers and restaurant staff. Formal user evaluation and performance benchmarking are left for future work.

Introduction

The Smart Food Ordering and Restaurant Management System is a web-based application designed to improve the food ordering experience and restaurant management through digital technology. Traditional food ordering methods, such as phone calls and messaging applications, often cause communication delays, manual errors in inventory management, and difficulties in checking food availability. Existing platforms also provide limited personalized food recommendations, multilingual support, and AI-based assistance.

The proposed system integrates food ordering, restaurant administration, kitchen inventory management, AI chatbot assistance, and personalized food recommendations into a single platform. Customers can browse food menus, search and filter dishes, check real-time availability, add items to their cart, and place orders. The system also supports three languages: Tamil, English, and Telugu, making it accessible to a wider range of users.

The AI chatbot helps customers by answering questions about food items, ingredients, prices, restaurants, nutrition information, and order procedures. The recommendation module suggests suitable dishes for children, gym users, and customers with specific dietary requirements using a transparent rule-based scoring method.

The system follows a three-tier client-server architecture. The frontend is developed using React, HTML, CSS, JavaScript, and Vite, while the backend uses Node.js and Express.js. SQLite and an in-memory data store are used for data management, and WebSocket technology provides real-time inventory updates. Administrators can manage food menus, prices, orders, and kitchen stock through a dedicated dashboard. Delivery partners can also view assignments and update delivery statuses.

The inventory module automatically classifies food items as available, low stock, or out of stock based on their quantities. When stock information changes, customers receive updated availability without refreshing the webpage. Security features such as authentication, password hashing, input validation, role-based access control, and parameterized database queries are also considered.

Conclusion

This paper presented the Smart Food Ordering and Restaurant Management System, a web application that unites online ordering, AI-assisted queries, Tamil, English and Telugu interaction, live inventory status, personalized dietary suggestions and restaurant administration. The inventory rule, cart safeguards and transparent recommendation score illustrate how these functions can be combined without relying on a complex machine-learning pipeline. The prototype shows that a single platform can serve customers and restaurant staff in one place. Future work will proceed along several directions. Persistent storage should replace the in-memory data store across all routes. Recommendations could learn from previous orders and verified nutritional data. Payment gateways such as UPI and card processing, together with live delivery tracking and cloud deployment, would extend the system into a complete service. Finally, a structured user study measuring ordering time, error rates and satisfaction would provide quantitative evidence of the benefits claimed in this paper.

References

[1] World Health Organization, “Healthy Diet,” 2024. [Online]. Available: https://www.who.int/news-room/factsheets/detail/healthy-diet [2] Meta Open Source, “React Documentation,” 2026. [Online]. Available: https://react.dev/ [3] Vite, “Vite Documentation,” 2026. [Online]. Available: https://vite.dev/ [4] OpenJS Foundation, “Node.js Documentation,” 2026. [Online]. Available: https://nodejs.org/docs/latest/api/ [5] OpenJS Foundation, “Express.js Documentation,” 2026. [Online]. [6] Available: https://expressjs.com/ [7] Mozilla Developer Network, “Using the Fetch API,” 2026. [Online]. Available: [8] https://developer.mozilla.org/en-US/docs/Web/API/Fetch_API/Using_F etch [9] Mozilla Developer Network, “WebSocket API,” 2026. [Online]. Available: [10] https://developer.mozilla.org/en-US/docs/Web/API/WebSockets_API [11] OWASP Foundation, “OWASP Application Security Verification Standard (ASVS),” 2025. [Online]. Available: https://owasp.org/www-project-application-security-verification-standard/ [12] OWASP Foundation, “OWASP Top 10: Web Application Security Risks,” 2025. [Online]. Available: https://owasp.org/www-project-top-ten/

Copyright

Copyright © 2026 Haritha Devi R S, Akshitha Shree M, Bala Shunmugi A. This 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.