Unified Grievance System: An AI-Assisted Web Platform for Complaint Management, Department Assignment, and Transparent Resolution Tracking

Abstract

The Unified Grievance System is a web-based prototype intended to improve the reporting and handling of civic and residential-community complaints. The platform provides a common interface through which residents can submit complaints, attach supporting images, view progress, and receive status updates, while administrators and staff can review, assign, and manage cases. The current implementation uses a React-based client, Supabase authentication and storage services, a FastAPI backend, and a Gemini-based AI analysis path intended to assist with complaint interpretation and department assignment. The system is designed around issue categories such as potholes, garbage overflow, broken street lights, and water leakage, with department options including Electrical Department, Water Department, General Maintenance, Road Maintenance, Sanitation. Administrative views support complaint summaries, category and monthly-trend analytics, resident management, and configuration of selected system settings. The paper describes the system architecture, workflow, implementation choices, AI-assisted routing approach, and a proposed evaluation methodology. The prototype is still under development and AI analysis behaviour has not been validated on a labelled dataset. Therefore, no classification accuracy, response-time improvement, or user-satisfaction result is claimed. The next steps are to stabilize AI integration, complete a complaint activity timeline, test role-based access and notifications, and evaluate the system using realistic complaint records.

Introduction

The Unified Grievance System is a web-based application designed to improve complaint management in residential societies and colonies. Residents often face problems such as overflowing garbage, damaged roads, broken streetlights, water leakage, and maintenance issues. Traditional reporting methods can result in fragmented records, delayed responses, unclear responsibilities, and difficulty tracking complaint resolution.

The proposed system provides a centralized platform where residents can submit complaints and monitor their status, while administrators can review complaints, assign them to the appropriate departments, manage residents, update progress, and analyze complaint trends.

A key feature is AI-assisted complaint analysis and department assignment. The system uses the Gemini model family to interpret complaint descriptions and recommend suitable departments, such as the Water Department, Electrical Department, Sanitation, Road Maintenance, General Maintenance, and Parks and Gardening. However, administrative oversight remains important, and AI recommendations are not intended to replace human decision-making.

The application follows a layered architecture comprising a React frontend, Supabase for authentication and file storage, a FastAPI backend, and an AI analysis component. Its main features include complaint submission, status tracking, department assignment, administrative dashboards, resident management, complaint summaries, category distribution, monthly trends, and records of status updates, remarks, and resolutions.

The project addresses important gaps in existing grievance systems by connecting the entire complaint lifecycle—from submission and review to assignment and resolution—in one platform. It also emphasizes the need to evaluate functional correctness, access control, AI recommendations, notifications, and usability through systematic testing.

The research reviews existing civic-reporting platforms, such as FixMyStreet, and studies on automated citizen-request classification and AI-assisted grievance management. Rather than claiming a new machine-learning architecture or superiority over existing systems, the project focuses on integrating these capabilities into a working prototype for residential communities.

Conclusion

The Unified Grievance System is a web-based platform developed to simplify and centralize complaint reporting and management within residential communities. The system provides a common interface through which residents can report civic issues, submit supporting images, record complaint locations, and monitor the progress of their complaints. By bringing residents, administrators, and staff together on a single platform, the system aims to make grievance management more organized, transparent, and accessible. The platform integrates a React-based frontend, Supabase authentication and storage services, a FastAPI backend, and a Gemini-based Artificial Intelligence component. These technologies work together to support essential functions such as complaint registration, user authentication, complaint management, department assignment, status tracking, and administrative monitoring. The system also incorporates features intended to assist administrators in identifying duplicate complaints, analyzing reported issues, and recommending appropriate departments for complaint resolution. One of the key contributions of the project is the integration of AI-assisted analysis into the complaint management workflow. Instead of relying entirely on manual interpretation, the system is designed to use AI to analyze complaint descriptions and assist in determining the appropriate department. This can help organize incoming complaints and reduce repetitive administrative work. However, the reliability of these AI-assisted functions must be evaluated through systematic testing before their effectiveness can be established. The system also emphasizes transparency and accountability by allowing complaints to be tracked throughout their lifecycle. Residents can follow complaint progress, while administrators can manage complaints, coordinate departments, monitor resolution status, and oversee user activities. Role-based access control helps separate resident and administrative functions, while authentication and database-level security measures contribute to protecting user information and maintaining controlled access to system resources. The development of this project demonstrates how modern web technologies and AI-assisted features can be combined to address practical problems in residential community management. By digitizing complaint registration and tracking, the platform provides an organized alternative to fragmented reporting methods, such as verbal complaints or informal communication channels. It also establishes a foundation for maintaining structured complaint records that can support future analysis and administrative decision-making. Although the current prototype implements the core complaint management workflow, certain aspects require further development and validation. These include comprehensive end-to-end testing, evaluation of AI-generated analysis, verification of access permissions across different user accounts, implementation of a complete complaint activity timeline, and improvements to notification reliability. Further enhancements could include external SMS integration, advanced complaint analytics, geographical visualization of recurring issues, multilingual support, and improved AI-assisted prioritization based on complaint severity. In conclusion, the Unified Grievance System provides a practical foundation for a more structured, transparent, and responsive grievance management process in residential communities. Its modular architecture allows additional features and integrations to be introduced as requirements evolve. With further testing, security verification, and functional improvements, the system has the potential to support more efficient complaint handling, better coordination between residents and administrators, and improved visibility into the resolution of community issues.

References

[1] mySociety, “FixMyStreet,” online platform and project description. Available: https://www.fixmystreet.com/ [2] N. Kim and S. Hong, “Automatic classification of citizen requests for transportation using deep learning: Case study from Boston city,” Information Processing & Management, vol. 58, no. 1, article 102410, 2021, doi: 10.1016/j.ipm.2020.102410. [3] V. Devi Nandhini et al., “AI-Powered Grievance Management System for Petition Analysis, Prioritization, Routing and Transparent Tracking,” conference paper, 2026, doi: 10.5281/zenodo.19408619. Verify publication metadata and peer-review status before formal submission. [4] S. C. Rajkumar et al., “A zero-shot LLM framework for multimodal grievance classification, urgency scoring, and abuse detection in civic feedback systems,” Scientific Reports, vol. 16, article 2299, 2026, doi: 10.1038/s41598-025-32079-7. [5] “Smart Complaint Routing Systems for Public Issues Using Multimodal Artificial Intelligence,” IEEE conference publication, 2026, doi: 10.1109/I3CTCON68242.2026.11508062. [6] React, “React Documentation.” Available: https://react.dev/ [7] Supabase, “Supabase Documentation.” Available: https://supabase.com/docs [8] FastAPI, “FastAPI Documentation.” Available: https://fastapi.tiangolo.com/ [9] Google, “Gemini API Documentation.” Available: https://ai.google.dev/gemini-api/docs [10] P. Cholke, A. S. Mishra, V. N. Mehar, M. R. Chikhale, H. S. Marathe, and D. Y. Nhalade, “AI-Driven Complaint Management System: An Automated Framework for Prioritizing, Tracking, and Resolving Public Service Complaints Efficiently,” in Proc. 2026 Int. Conf. on Emerging Smart Computing and Informatics (ESCI), 2026, doi: 10.1109/ESCI68015.2026.11493166. DOI [11] “Smart Complaint Routing Systems for Public Issues Using Multimodal Artificial Intelligence,” in Proc. 2026 IEEE Int. Conf. for Convergence in Computing Technology (I3CTCON), 2026, doi: 10.1109/I3CTCON68242.2026.11508062. DOI [12] R. Kumar B. N., T. S., S. R., S. P. R., and S. R., “CivicFix: Smart Complaint Routing for Urban Solutions,” International Journal of Advanced Research in Computer and Communication Engineering, vol. 14, no. 12, Dec. 2025, doi: 10.17148/IJARCCE.2025.141222. IJARCCE [13] A. N., G. Yogeshwaran, U. D. R., M. Kumar C., and S. R., “Smart Civic AI: Intelligent Complaint Management System,” International Journal of Advanced Research in Computer and Communication Engineering, vol. 15, no. 5, 2026, doi: 10.17148/IJARCCE.2026.155196. IJARCCE [14] “An Intelligent Government Complaint Prediction Approach,” Big Data Research, vol. 30, Art. no. 100336, Nov. 2022, doi: 10.1016/j.bdr.2022.100336. ScienceDirect [15] “Proactive Complaint Management in Public Sector Informatics Using AI: A Semantic Pattern Recognition Framework,” Applied Sciences, vol. 15, no. 12, Art. no. 6673, 2025. MDPI [16] B. S. Menon and D. R. S., “Civi-Snap: A Deep Learning Powered Geotagged Civic Issue Detection and Monitoring Framework,” International Journal of Creative and Open Research in Engineering and Management, vol. 2, no. 4, 2026. ResearchGate [17] M. M. Mayuranathan, M. Dharshini, A. Aruneswaran, and N. G. Bose, “AI-Driven Predictive Analytics for Citizen Complaint Trends in Smart Cities,” International Journal of Novel Research and Development, 2026. IJNRD [18] M. Abusalam, M. Sartaz, and S. Ghosh, “AI-COMPLAINT: An Intelligent Web-Based Complaint Management System Utilizing Natural Language Processing for Automated Complaint Classification and Smart Department Routing,” Journal of Science Engineering Technology and Management Science, vol. 3, no. 6, pp. 147–154, 2026, doi: 10.64771/jsetms.2026.v03.i06.pp147-154. JSETMS [19] G. Khlaf, “An Automated Complaint Analysis and Classification System using Machine Learning and Natural Language Processing,” Zenodo, May 2026. Zenodo [20] R. A. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” in Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779–788, doi: 10.1109/CVPR.2016.91. [21] T.-Y. Lin et al., “Microsoft COCO: Common Objects in Context,” in Computer Vision – ECCV 2014, LNCS, vol. 8693, Springer, 2014, pp. 740–755, doi: 10.1007/978-3-319-10602-1_48. [22] M. Abadi et al., “TensorFlow: A System for Large-Scale Machine Learning,” in Proc. 12th USENIX Symp. on Operating Systems Design and Implementation (OSDI), 2016, pp. 265–283. [23] F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011. [24] M. C. Stonebraker and L. A. Rowe, “The Design of POSTGRES,” in Proc. ACM SIGMOD Int. Conf. on Management of Data, 1986, pp. 340–355. [25] FixMyStreet, “FixMyStreet: Report, view and discuss local street problems,” [Online]. Available: https://www.fixmystreet.com [26] CivicPlus, “SeeClickFix 311 CRM: Citizen Request Management,” [Online]. Available: https://www.civicplus.com/seeclickfix-311-crm/

Copyright

Copyright © 2026 Mr. Vaibhav Shrivastava, Devyani Dongre, Nikita Yahake, Tanvi Tilwani, Parul Mishra, Pratiksha Gaidhane, Keshar Bingewar. 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.