A Lightweight Cross-Platform Smart Mess Discovery and Management Application Using Flutter

Abstract

Mess and tiffin services across urban India serve tens of millions of students and working professionals, yet the operational infrastructure governing discovery, communication, and nutritional awareness within these services remains largely manual, fragmented, and informationally opaque. This paper presents a lightweight cross-platform mobile application that unifies three functionalities absent from any single prior system: Haversine-based geospatial proximity discovery, Firestore-driven community communication, and Gemini API-backed nutritional inference. The system is engineered using Flutter and Dart for cross-platform delivery, Firebase for serverless backend services, the Google Maps SDK for geolocation, and the Gemini API for dynamic per-item nutritional estimation. A multi-attribute filter predicate combines dietary type, price ceiling, and meal-slot constraints with the proximity result set to produce a refined candidate list in O(n) time. Empirical evaluation against datasets of 100, 500, and 1,000 mess records yields search query latencies of 162?ms, 225?ms, and 308?ms respectively, confirming sub-linear growth consistent with the O(n?k) theoretical model. Outdoor GPS positioning achieves 5–10?m accuracy with a 2.1?s fix time; indoor Wi-Fi-assisted positioning yields 15–30?m accuracy with a 1.8?s fix time. Mess retrieval success reaches 100% within a 5?km search radius, with zero incorrect filter retrievals across all test scenarios. The combined contribution addresses an underserved gap in scalable, resource-light food-service management for institutional and residential environments.

Introduction

This work presents a smart mobile application for managing and discovering informal mess and tiffin services in urban India, where millions of students and migrant workers rely on unorganized food providers lacking digital infrastructure.

Currently, mess services operate through word-of-mouth, handwritten menus, and informal pricing, with no reliable system for discovery, comparison, or real-time updates. Existing food apps (like restaurant aggregators) do not support the specific needs of small mess operators, such as daily menu updates, subscriptions, and lightweight operation on basic smartphones.

To solve this, the paper proposes a Flutter + Firebase-based cross-platform system with three core features:

• Geospatial mess discovery: Uses the Haversine formula to calculate real-time distance between users and nearby mess facilities, enabling location-based search with filters like price, diet, and radius.
• Real-time communication system: A mess-specific chat module using Firebase Firestore allows instant communication between users and operators.
• AI nutrition assistant (Bhojan AI): Uses the Gemini API to estimate calories and macronutrients from menu descriptions without needing a fixed food database.

The system is built using Flutter for cross-platform mobile support and Firebase for authentication, real-time database, notifications, and serverless backend functions. It supports both consumer and administrator roles, where operators can update menus and users receive instant updates via push notifications.

From literature review, existing systems fail due to lack of real-time updates, poor support for small mess operators, no AI nutrition features, and no unified platform combining discovery, communication, and dietary analysis.

The key problems identified are:

• Difficulty in discovering nearby mess services
• Lack of instant communication between users and operators
• No nutritional transparency of daily meals

The proposed system addresses these through a three-tier architecture (Flutter frontend, API layer, Firebase backend) with real-time synchronization and secure role-based access control.

Search and location modules are optimized using:

• Keyword-based matching (O(n) complexity)
• Distance calculation using Haversine-based geolocation filtering

Overall, the system provides a lightweight, scalable, and unified digital platf

Conclusion

This paper has presented a cross-platform mess management application that resolves three concurrent limitations of existing food-service discovery tools: geospatial opacity, communication latency, and nutritional information absence. The system delivers Haversine-based proximity discovery with O(n) complexity, multi-attribute dietary and price filtering with O(n) predicate evaluation, text search with O(n?k) complexity, and a Gemini API-backed nutritional assistant that eliminates static-database constraints. Empirical evaluation across 100–1,000 records confirms sub-400?ms query latency, sub-30?m location accuracy, and zero filter errors, meeting the performance requirements of real-time meal-decision contexts. The Flutter-Firebase implementation achieves cross-platform parity from a single codebase, significantly reducing deployment overhead for small-scale institutional deployments. The Firestore real-time listener architecture propagates menu and announcement updates within 200–500?ms under 4G conditions, satisfying the temporal requirements of time-sensitive mess communications. Future work will address four open directions: (1) offline-first caching using Firestore’s local persistence for low-connectivity environments; (2) integration of image-based food recognition to extract nutritional data from menu photographs; (3) a subscription and payment module enabling cashless mess enrollment; and (4) scaling validation beyond 1,000 records to characterize latency behaviour at 5,000–10,000 entries, corresponding to city-level deployment.

References

[1] T. Sharma, A. Singh, and N. Tyagi, “Real-time chat application development using Flutter and Firebase,” Int. J. Comput. Appl. (IJCA), vol. 183, no. 15, pp. 1–6, 2021. [2] S. Patil and D. Bhadane, “Daily Mess: A mobile application to uplift mess services for students,” Int. J. Res. Appl. Sci. Eng. Technol. (IJRASET), vol. 10, no. 4, pp. 892–897, 2022. [3] A. Kumar, P. Joshi, and S. Mehta, “AI-based food recommendation and pre-booking system for institutional canteens,” in Proc. IEEE Int. Conf. Artif. Intell. Mach. Learn. (ICAIML), Bangalore, India, 2022, pp. 112–118. [4] R. Nair and K. Menon, “AI-based nutrition value estimation for daily dietary planning,” in Proc. IEEE Int. Conf. Adv. Comput. Commun. Syst. (ICACCS), Coimbatore, India, 2023, pp. 245–251. [5] R. Sharma, S. Kumar, and A. Gupta, “Smart food service and monitoring system using IoT and machine learning,” in Proc. IEEE Int. Conf. IoT Intell. Syst. (ICIIS), Jaipur, India, 2022, pp. 67–74. [6] H. Verma, A. Dwivedi, D. Rastogi, A. K. Rai, A. Kanaujia, and B. Mall, “Food delivery application using Flutter and Firebase,” Int. Res. J. Modernization Eng. Technol. Sci. (IRJMETS), vol. 7, no. 12, pp. 1540–1547, Dec. 2025. [7] A. Gujar, S. Narkar, T. Joshi, P. Pawar, and S. Taral, “A local discovery food application using Flutter,” Int. Res. J. Eng. Technol. (IRJET), vol. 12, no. 3, pp. 1023–1028, Mar. 2025. [8] B. H. Hanumant, T. G. Balkrishna, S. A. A. Kumar, T. N. Santosh, and A. Shinde, “SmartMess: An online mess food ordering platform,” Int. J. Innov. Res. Technol. (IJIRT), vol. 11, no. 5, pp. 214–219, Oct. 2024. [9] S. S. Pawar et al., “Meals on Wheels: A platform to connect tiffin services with customers using mobile technology,” in Proc. IEEE Int. Conf. Innov. Technol. (InCITe), Bhubaneswar, India, 2023, pp. 318–323. [10] A. Kumar et al., “AI-powered smart waste bins for improving hostel mess waste management,” in Proc. IEEE Int. Conf. Comput. Intell. Sustain. Technol. (InCITS), Nashik, India, 2024, pp. 88–94. [11] S. K. Sharma, R. Gupta, and P. Verma, “Mess management system: Digitizing hostel food services,” Int. J. Sci. Res. Comput. Sci. Eng. Inf. Technol., vol. 10, no. 6, pp. 451–458, Nov.–Dec. 2024. [12] Y. Zhang et al., “AI nutrition recommendation using a deep generative model and ChatGPT,” Sci. Rep., vol. 14, no. 1, p. 16625, Jul. 2024. [13] R. Gupta et al., “Smart AI chatbots for tailored nutrition and fitness recommendations,” in Proc. IEEE Int. Conf. Innov. Comput. Commun. (ICICC), Greater Noida, India, 2024, pp. 1–6. [14] P. Singh et al., “E-canteen management system based on web application for institutional food services,” in Proc. IEEE Int. Conf. Emerg. Technol. (ICET), Islamabad, 2022, pp. 201–206. [15] A. Deshmukh and S. Patil, “Cashless mess transaction system using RFID and mobile interface,” Int. J. Res. Appl. Sci. Eng. Technol. (IJRASET), vol. 11, no. 2, pp. 678–683, 2023. [16] T. Bray, “The JavaScript Object Notation (JSON) data interchange format,” IETF RFC 8259, Dec. 2017. [17] G. Sellers and J. Kessenich, “Flutter architecture and performance optimization in cross-platform mobile development,” in Proc. ACM Int. Conf. Mobile Softw. Eng. Syst. (MOBILESoft), Melbourne, Australia, 2023, pp. 44–51. [18] S. Vinoski, “Convenience over correctness: Firebase Firestore real-time database architecture,” IEEE Internet Comput., vol. 26, no. 3, pp. 74–78, May–Jun. 2022. [19] C. Senabre Hidalgo and B. Bria, “Geolocated service discovery on mobile platforms: A review of proximity algorithms,” IEEE Access, vol. 11, pp. 34512–34527, 2023. [20] R. Sinnott, “Virtues of the Haversine,” Sky Telescope, vol. 68, no. 2, p. 158, 1984. [21] M. Fowler, NoSQL Distilled: A Brief Guide to the Emerging World of Polyglot Persistence. Addison-Wesley, Boston, MA, 2012. [22] Google LLC, “Gemini API documentation: Gemini 1.5 Flash model,” Google AI for Developers, 2024. [Online]. Available: https://ai.google.dev/gemini-api/docs

Copyright

Copyright © 2026 Atharva Pachpute, Srushti Mhaske, Sahil Wanjare, Digvijay Patil. 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.