Transforming Rail Madad System: An AI- Driven Approach To Complaint Resolution

The design and implementation of adding Artificial Intelligence to the RAILMADAD program are described in this paper. The system is intended to enhance the grievance reporting process for passengers and then investigate their complaints. Additionally, the system equipped with memory-based RAG chatbot to enhance passenger user experience. To reduce the work fatigue for railway employees, to speed up resolving the issue and implementing corrective measures with the office, we implemented a routing algorithm. After analyzing the sentiment surrounding the complaint, this AI model will rank the issue’s urgency for action. To improve the precision and speed of complaint handling, metadata such as location, issues, and timestamp are extracted. To protect the privacy of the passengers, the complaints filed by the passengers are compiled into a tokenized form. For passengers: Convenience: 24/7 grievance redressal mechanism, without the human intervention. Privacy: Maintain the privacy of the complaints from the passengers. Efficiency: Increases the speed for the resolution of the given complaints. AI Chatbots: Deploy AI chatbots to provide instant acknowledgment and preliminary responses, collecting additional necessary information from the complainant. The proposed model: An bi-directional encoder representation from Transformers model which analyses user’s complaint input sequences, capturing the underlying character level feature and then classifies them into their respective departments to accurate redressal of complaints. Our model outperforms several highline model achieving an accuracy of 95.93 percent and F1-score of 0.96.