Agented AI for Intelligent Workflow Automation: An Agent Framework Powered By Large Language Models

Sant Kumar Maurya; Siddharth Kumar

doi:10.70454/JRICST.2026.30406

Authors

Sant Kumar Maurya Department of AIT-CSE, Chandigarh University, Mohali, Punjab Author
Siddharth Kumar Department of AIT-CSE, Chandigarh University, Mohali, Punjab Author

DOI:

https://doi.org/10.70454/JRICST.2026.30406

Keywords:

Agentic AI, Large Language Models, Multi-Agent Systems, Workflow Automation, Retrieval-Augmented Generation, Human-in-the-Loop, Explainable AI, Enterprise Automation, Planning Agent, Tool Calling

Abstract

Enterprise workflow automation has progressed from rigid, rule-based Robotic Process Automation (RPA) toward cognitive systems capable of perception, reasoning, and autonomous action. While Large Language Models (LLMs) have demonstrated strong capabilities in natural language understanding and generation, single-model deployments remain limited in multi-step planning, long-horizon memory, tool orchestration, and verifiable decision-making, which are essential for real-world enterprise workflows. This paper introduces AIWA-Agent, an Intelligent Multi-Agent Workflow Automation Framework that integrates a Planning Agent, a Reasoning Agent, a Memory Agent, a Retrieval-Augmented Generation (RAG) module, a Tool-Calling Agent, an Execution Agent, and a Validation Agent under a Human-in-the-Loop and Explainable AI (XAI) supervision layer. Unlike prior single-agent LLM assistants and loosely coupled multi-agent toolkits, AIWA-Agent formalizes agent interaction through seven governing mathematical constructs Workflow Automation Efficiency , Agent Collaboration Score , Task Completion Accuracy, Workflow Latency , Resource Utilization , LLM Confidence Score , and Agent Decision Score enabling quantitative, explainable orchestration rather than ad hoc prompt chaining. We formalize the framework's architecture, mathematical model, and methodology, and evaluate it through a structured experimental protocol comparing AIWA-Agent against Catgut (single-agent), Autopsy, Crewel, Lang Graph, and Autopen baselines using illustrative, internally consistent performance data across accuracy, latency, collaboration, and resource metrics, together with descriptive statistics, correlation, regression, ANOVA, and reliability analysis. Results indicate that the proposed coordination and validation layers yield measurable gains in workflow efficiency and task success while preserving interpretability. The findings offer a reproducible blueprint for enterprises pursuing agented automation with governance and explain ability by design.

Downloads

Download data is not yet available.

References

[1]. A. Aswan, N. Shaker, N. Parmer, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, "Attention is all you need," arXiv:1706.03762, 2017. Doe: 10.48550/arXiv.1706.03762.

[2]. J. Wei, X. Wang, D. Schulman’s, M. Busman, F. Xia, E. Chi, Q. V. Le, and D. Zhou, "Chain-of-thought prompting elicits reasoning in large language models," in Advances in Neural Information Processing Systems (Neurons), vol. 35, pp. 24824-24837, 2022.

[3]. S. Yao, J. Zhao, D. Yu, N. Du, I. Shoran, K. Narasimhan, and Y. Cao, "React: Synergizing reasoning and acting in language models," in Proc. Int. Conf. Learning Representations (ICLR), 2023, arXiv:2210.03629. Doe: 10.48550/arXiv.2210.03629.

[4]. S. Yao, D. Yu, J. Zhao, I. Shoran, T. Griffiths, Y. Cao, and K. Narasimhan, "Tree of thoughts: Deliberate problem solving with large language models," in Advances in Neural Information Processing Systems (Neurons), vol. 36, 2023.

[5]. T. Schick, J. Dived-Yu, R. Desist, R. Railing, M. Lamely, E. Hambros, L. Zettlemoyer, N. Concede, and T. Siloam, "Tool former: Language models can teach themselves to use tools," in Advances in Neural Information Processing Systems (Neurons), vol. 36, pp. 68539-68551, 2023.

[6]. N. Shinn, F. Casino, E. Berman, A. Goliath, K. Narasimhan, and S. Yao, "Reflex ion: Language agents with verbal reinforcement learning," arXiv: 2303.11366, 2023. Doe: 10.48550/arXiv.2303.11366.

[7]. Y. Sheen, K. Song, X. Tan, D. Li, W. Lu, and Y. Zhan, "HuggingGPT: Solving AI tasks with Catgut and its friends in Hugging Face," arXiv:2303.17580, 2023. Doe: 10.48550/arXiv.2303.17580.

[8]. P. Lewis, E. Perez, A. Pintos, F. Petron, V. Capuchin, N. Goal, H. Kittler, M. Lewis, W. Yam, T. Rocktaschel, S. Riedel, and D. Kiel, "Retrieval-augmented generation for knowledge-intensive NLP tasks," in Advances in Neural Information Processing Systems (Neurons), vol. 33, pp. 9459-9474, 2020.

[9]. Y. Ago, Y. Xing, X. Ago, K. Jiao, J. Pan, Y. Bi, Y. Dai, J. Sun, M. Wang, and H. Wang, "Retrieval-augmented generation for large language models: A survey," arXiv:2312.10997, 2024. Doe: 10.48550/arXiv.2312.10997.

[10]. L. Wang, C. Ma, X. Fang, Z. Zhang, H. Yang, J. Zhang, Z. Chen, J. Tang, X. Chen, Y. Lin, et al., "A survey on large language model based autonomous agents," Frontiers of Computer Science, vol. 18, no. 6, art. 186345, 2024. Doe: 10.1007/s11704-024-40231-1.

[11]. Z. Xi, W. Chen, X. Goo, W. He, Y. Ding, B. Hong, M. Zhang, J. Wang, S. Jin, E. Zhou, et al., "The rise and potential of large language model based agents: A survey," Science China Information Sciences, vol. 68, no. 2, art. 121101, 2025. Doe: 10.1007/s11432-024-4222-0.

[12]. Q. Wu, G. Banal, J. Zhang, Y. Wu, B. Li, E. Zhu, L. Jiang, X. Zhang, S. Zhang, J. Liu, A. H. Awadallah, R. W. White, D. Burger, and C. Wang, "Autopen: Enabling next-gen LLM applications via multi-agent conversations," in Proc. 1st Conf. Language Modeling (COLM), 2024, arXiv:2308.08155. Doe: 10.48550/arXiv.2308.08155.

[13]. K.-T. Tran, D. Dao, M.-D. Nguyen, Q.-V. Pham, B. O'Sullivan, and H. D. Nguyen, "Multi-agent collaboration mechanisms: A survey of LLMs," arXiv: 2501.06322, 2025. Doe: 10.48550/arXiv.2501.06322.

[14]. V. Canola, V. Hassidic, A. R. Sultana, D. Ghost, D. Hangar, and B. Skidder, "A review of trustworthy and explainable artificial intelligence (XAI)," IEEE Access, vol. 11, pp. 78994-79015, 2023. Doe: 10.1109/ACCESS.2023.3294569.

[15]. M. Marsha, K. Lam, J. Wood, A. Alchemy, and J. Kaila, "Explainable artificial intelligence: A survey of needs, techniques, applications, and future direction," arXiv: 2409.00265, 2024. Doe: 10.48550/arXiv.2409.00265.

[16]. E. Cambria, L. Maladroit, F. Mercuric, N. Nubian, and A. Saves, "XAI meets LLMs: A survey of the relation between explainable AI and large language models," arXiv: 2407.15248, 2024. Doe: 10.48550/arXiv.2407.15248.

[17]. X. Wu, L. Xiao, Y. Sun, J. Zhang, T. Ma, and L. He, "A survey of human-in-the-loop for machine learning," Future Generation Computer Systems, vol. 135, pp. 364-381, 2022. Doe: 10.1016/j.future.2022.05.014.

[18]. [18] S. Afrin, S. Roxana, and R. Aram, "AI-enhanced robotic process automation: A review of intelligent automation innovations," IEEE Access, vol. 12, pp. 186534-186552, 2024. Doe: 10.1109/ACCESS.2024.3513279.

[19]. J. G. Enriquez, A. Jimenez-Ramirez, F. J. Dominguez-Mayo, and J. A. Garcia-Garcia, "Robotic process automation: A scientific and industrial systematic mapping study," IEEE Access, vol. 8, pp. 39113-39129, 2020. Doe: 10.1109/ACCESS.2020.2974934.

[20]. T. Chakraborti, V. Isahagian, R. Khalid, Y. Khazaeni, V. Muthusamy, Y. Risk, and M. Univac, "From robotic process automation to intelligent process automation: Emerging trends," in Business Process Management: Block chain and RPA Forum (BPM 2020), Springer, 2020, pp. 215-228. Doe: 10.1007/978-3-030-58779-6_15.

[21]. V. Tope and S. Tube, "AI agented workflows and enterprise APIs: Adapting API architectures for the age of AI agents," arXiv: 2502.17443, 2025. Doe: 10.48550/arXiv.2502.17443.

[22]. D. Finland, F. Fournier, L. Lemonade, I. Skarbovsky, and A. J. E. Swevels, "How well can large language models explain business processes?" arXiv:2401.12846, 2024. Doe: 10.48550/arXiv.2401.12846.

[23]. M. Chui, E. Has an, R. Roberts, A. Single, K. Same, A. Sukharevsky, L. Yee, and R. Zimmer, "The economic potential of generative AI: The next productivity frontier," McKinsey Global Institute Report, 2023.

[24]. K. Gurnard, A.Jangra, H. Baber, and M. M. Islam, "An analytical review on the impact of artificial intelligence on the business industry: Applications, trends, and challenges," IEEE Engineering Management Review, vol. 52, no. 2, pp. 45-58, 2024. Doe: 10.1109/EMR.2024.3355973.

[25]. Open-air, J. Achaia, S. Adler, S. Agawam, et al., "GPT-4 technical report," arXiv: 2303.08774, 2023. Doe: 10.48550/arXiv.2303.08774.

[26]. H. Touring, L. Martin, K. Stone, P. Albert, A. Almahairi, Y. Babied, N. Bashlykov, et al., "Llama 2: Open foundation and fine-tuned chat models," arXiv:2307.09288, 2023. Doe: 10.48550/arXiv.2307.09288.

[27]. W. X. Zhao, K. Zhou, J. Li, T. Tang, X. Wang, Y. Hour, Y. Min, B. Zhang, J. Zhang, Z. Dong, et al., "A survey of large language models," arXiv:2303.18223, 2023. Doe: 10.48550/arXiv.2303.18223.

[28]. Z. Jib, N. Lee, R. Frisked, T. Yu, D. Su, Y. CSU, E. Ishii, Y. J. Bang, A. Mayotte, and P. Fung, "Survey of hallucination in natural language generation," ACM Computing Surveys, vol. 55, no. 12, pp. 1-38, 2023. Doe: 10.1145/3571730.

[29]. Crewel Inc., "Crewel: Framework for orchestrating role-playing, autonomous AI agents," Gather repository, https://github.com/crewAIInc/crewAI, 2024.

[30]. Lang Chain Inc., "Lang Graph: Building language agents as graphs," Gather repository and documentation, https://github.com/langchain-ai/langgraph, 2024.