Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM64673
Volume:
10 |
6 |
Jun 2026
Pages:
1-7
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department of Computer Engineering
Sinhgad College of Engineering
Department of Computer Engineering
Sinhgad College of Engineering
Department of Computer Engineering
Sinhgad College of Engineering
Department of Computer Engineering
Sinhgad College of Engineering
Department of Computer Engineering
Sinhgad College of Engineering
Enterprise organizations increasingly require intelligent automation systems capable of managing diverse business processes efficiently. Recent advances in Large Language Models (LLMs) and Multi-Agent Systems have enabled the development of intelligent platforms that support automated decision-making across multiple operational domains. This paper presents an Orchestration-Driven Intelligent Multi-Agent Framework for Enterprise Automation and Decision-Making. The proposed system integrates four specialized agents: an HR Agent for recruitment analysis, a Finance Agent for financial assessment, a Customer Agent for sentiment and feedback evaluation, and a TrackX Agent for employee performance monitoring. A centralized orchestration layer is introduced to coordinate workflow execution, task routing, context management, and response aggregation. Unlike traditional multi-agent architectures that rely on direct agent-to-agent communication, the proposed framework employs a token-efficient communication strategy based on context summarization, selective context transfer, and memory compression. These mechanisms reduce redundant information exchange and improve communication efficiency within the system. The framework is implemented using Python, LangChain, and Gemini 1.5 Pro. Structured JSON-based communication and orchestration-driven workflows enhance scalability, maintainability, and modularity. The proposed architecture provides a practical methodology for developing enterprise-scale intelligent systems while addressing communication overhead and token consumption challenges associated with Large Language Model based multi-agent environments.
KEYWORDS
This paper proposes an Orchestration-Driven Intelligent Multi-Agent Framework for enterprise automation and decision-making across domains such as recruitment, finance, CRM, and employee performance monitoring. The framework uses a centralized orchestration layer to coordinate workflows, manage context and memory, route tasks, and optimize communication between specialized agents. Unlike traditional multi-agent systems that rely on direct agent-to-agent communication, this approach centralizes control to reduce complexity and improve maintainability. A major contribution is the Token-Efficient Communication Methodology, which employs context summarization, selective context transfer, and memory compression to minimize token consumption in LLM-based systems. By optimizing information exchange and centralizing workflow management, the framework offers a scalable and cost-effective solution for enterprise AI applications. Future work includes experimental validation, adaptive orchestration, retrieval-augmented intelligence, distributed agent networks, and advanced token optimization techniques to further enhance scalability and real-world deployment.
[1] J. S. Park, J. O'Brien, C. J. Cai, M. R. Morris, P. Liang, and M. S. Bernstein, "Generative Agents: Interactive Simulacra of Human Behavior," in Proc. 36th Annual ACM Symposium on User Interface Software and Technology (UIST '23), San Francisco, CA, USA, 2023, pp. 1-22. [2] C. Adornetto, A. Mora, K. Hu, L. I. Garcia, P. Atchade-Adelomou, G. Greco, L. A. A. Pastor, and K. Larson, "Generative Agents in Agent-Based Modeling: Overview, Validation, and Emerging Challenges," IEEE Transactions on Artificial Intelligence, 2025. [3] S. Sriram, P. Kumar, and R. Nair, "Privacy-Preserving Multi-Agent Framework Using Local LLMs for Secure Enterprise Task Automation," in Proc. International Conference on Machine Learning and Applications (ICMLA), 2024, pp. 112-119. [4] L. Fan, W. Jiang, W. Qiu, and M. Zhang, "A Review of Multi-Agent Modeling and Simulation Methodologies: Applications, Challenges, and Future Directions," IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 51, no. 9, pp. 5510-5526, Sep. 2021. [5] LangChain Inc., "LangChain: Building Applications with LLMs Through Composability," LangChain Documentation, 2024. [6] W. McKinney, "Data Structures for Statistical Computing in Python," in Proc. 9th Python in Science Conference (SciPy 2010), Austin, TX, USA, 2010, pp. 56-61. [7]Plotly Technologies Inc., "Collaborative Data Science: Plotly Python Graphing Library," Plotly Documentation, 2024. [8]Automation Anywhere, "Multi-Agent Systems: Building the Autonomous Enterprise," Automation Anywhere Whitepaper, 2025. [9] IBM, "What is a Multi-Agent System?," IBM Think, 2024. [10] X. Zhou et al., "SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents," in Proc. International Conference on Learning Representations (ICLR 2024), Vienna, Austria, 2024.
Publication Date
6/8/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
4
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...