An Agentic AI Framework for Intelligent Support Ticket Classification, Routing, and Automated Resolution

In this paper, we have described the Agentic AI Framework for Intelligent Support Ticket Classification, Routing and Automation. We have focused on solving the ongoing issues of inefficient triaging, mis-routing, prolonged resolution times and poorly grounded automation of enterprise IT service desks. By splitting the support ticket life cycle into a system of eight specialized state-managed AI agents, our framework integrates deep semantic understanding, multi-factor priority assessment, workload-based ticket routing and RAG-based knowledge base retrieval. The framework overcomes several key problems of the previous approaches by adding a verification step performed by the Verification Agent in conjunction with an independent Human Escalation Agent routing critical and uncertain incidents to human technicians with the pre-assembled diagnostic context. With this design, we achieve maximum automation without compromising safety and auditability of the process. By proposing a methodology, mathematical formulation and benchmarking standards — and by situating the design against the most recent 2024–2026 agentic-AI, retrieval-verification, and explainable-routing literature [4], [6], [7], [12]–[14] — we lay the basis for the future deployment of autonomous agentic AI in enterprises.

[1]  A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, "Attention is all you need," in Advances in Neural Information Processing Systems (NeurIPS), vol. 30, 2017, pp. 5998–6008. [2]  J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, "BERT: Pre-training of deep bidirectional transformers for language understanding," in Proc. 2019 Conf. North American Chapter of the ACL (NAACL-HLT), 2019, pp. 4171–4186. [3]  N. Reimers and I. Gurevych, "Sentence-BERT: Sentence embeddings using Siamese BERT-networks," in Proc. 2019 Conf. Empirical Methods in Natural Language Processing (EMNLP), 2019, pp. 3982–3992. doi: 10.18653/v1/D19-1410. [4]  Y. Gao, Y. Xiong, X. Gao, K. Jia, J. Pan, Y. Bi, Y. Dai, J. Sun, M. Wang, and H. Wang, "Retrieval-Augmented Generation for Large Language Models: A Survey," arXiv preprint arXiv:2312.10997, 2023. [5]  P. Lewis, E. Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal, H. Küttler, M. Lewis, W.-t. Yih, T. Rocktäschel, S. Riedel, and D. Kiela, "Retrieval-augmented generation for knowledge-intensive NLP tasks," in Advances in Neural Information Processing Systems (NeurIPS), vol. 33, 2020, pp. 9459–9474. [6]  A. Asai, S. Wu, Y. Wang, A. Sil, and H. Hajishirzi, "Self-RAG: Learning to retrieve, generate, and critique through self-reflection," in Proc. Int. Conf. on Learning Representations (ICLR), 2024. [7]  C. Niu, Y. Wu, J. Zhu, S. Xu, K. Shum, R. Zhong, J. Song, and T. Zhang, "RAGTruth: A hallucination corpus for developing trustworthy retrieval-augmented language models," in Proc. 62nd Annual Meeting of the Association for Computational Linguistics (ACL), 2024, pp. 10862–10878. [8]  Q. Wu, G. Bansal, J. Zhang, Y. Wu, B. Li, N. Zhu, L. Jiang, X. Zhang, S. Zhang, J. Liu, A. H. Awadallah, R. W. White, D. Burger, and C. Wang, "AutoGen: Enabling next-gen LLM applications via multi-agent conversation," arXiv preprint arXiv:2308.08155, 2023. [9]  T. Guo, X. Chen, Y. Wang, R. Chang, S. Pei, N. V. Chawla, O. Wiest, and X. Zhang, "Large language model based multi-agents: A survey of progress and challenges," in Proc. 33rd Int. Joint Conf. on Artificial Intelligence (IJCAI), 2024. [10]             H. Chase et al., "LangChain / LangGraph: Building stateful, multi-actor applications with LLMs," Software Documentation and Framework, 2024. [Online]. Available: https://github.com/langchain-ai/langgraph