Authors : Sahil Sanas; Arya Naik; Aditya Veerkar; Pooja T. Kohok

Volume/Issue : Volume 10 - 2025, Issue 11 - November

Google Scholar : https://tinyurl.com/32ayts2z

Scribd : https://tinyurl.com/y4vamv44

DOI : https://doi.org/10.38124/ijisrt/25nov074

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 30 to 40 days to display the article.

Abstract : Legacy COBOL and C systems are still widely used in industries, yet they are increasingly costly, insecure, and incompatible with modern platforms. Traditional modernization methods, although effective, often require significant time and resources and carry high risks of disruption. This paper proposes a scalable alternative that leverages Large Language Models (LLMs) within a structured multi- agent framework, guided by the “7Rs of Modernization.” The frame- work comprises three agents: an Analysis Agent that interprets and maps legacy code, a Coder Agent that generates modern equivalents, and a Review Agent that validates correctness, security, and compliance through iterative feedback. By automating much of the migration process, the proposed approach enables faster, more transparent, and less risky mod- ernization. It helps enterprises transition from outdated systems to modular, secure, and cloud-ready solutions, offering a cost- effective and future-proof pathway to digital transformation.

Keywords : LLM, Legacy Code, Modernization, 7Rs of Mod- Ernization, Multi-Agent Framework, Software Migration, Reliabil- Ity, Performance.

References :

1. K. Greff, R. K. Srivastava, J. Koutn´ık, B. R. Steunebrink, and J. Schmidhuber, “LSTM: A search space odyssey,” IEEE Transactions on Neural Networks and Learning Systems, vol. 28, no. 10, pp. 2222–2232, Oct. 2016.
2. S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, Nov. 1997.
3. G. Bandarupalli, “Code Reborn: AI-Driven Legacy Systems Modern- ization from COBOL to Java,” arXiv preprint arXiv:2504.11335, Apr. 2025. Available: https://arxiv.org/abs/2504.11335
4. J. Fitzpatrick, “Case Study: Converting C Programs to C++,” C++ Report, vol. 8, no. 2, p. 40, 1996.
5. S. Froimovich, R. Gal, W. Ibraheem, and A. Ziv, “Quality Eval- uation of COBOL to Java Code Transformation,” arXiv preprint arXiv:2507.23356, Jul. 2025. Available: https://arxiv.org/abs/2507.23356
6. L. Solovyeva, et al., “Leveraging LLMs for Automated Translation of Legacy Code: A Case Study on PL/SQL to Java Transformation” arXiv:2508.19663v1 [cs.SE] 27 Aug 2025. Available: https://arxiv.org/ html/2508.19663
7. K. Cheng, X. Shen, Y. Yang, T. Wang, Y. Cao, M. A. Ali, H. Wang, L. Hu, D. Wang, “CODEMENV: Benchmarking Large Language Models on Code Migration,” arXiv preprint arXiv:2506.00894, 2025. Available: https://arxiv.org/abs/2506.00894
8. A. T. V. Dau, H. T. Dao, A. T. Nguyen, et al., “XMainframe: A Large Language Model for Mainframe Modernization,” arXiv preprint arXiv:2408.04660, 2024. Available: https://arxiv.org/abs/2408.04660
9. C. Ziftci, S. Nikolov, A. Sjo¨vall, B. Kim, D. Codecasa, M. Kim, et al., “Migrating Code At Scale With LLMs At Google,” arXiv preprint arXiv:2504.09691, 2025. Available: https://arxiv.org/pdf/2504.09691
10. J. Ala-Salmi, et al., “Autonomous Multi-Agent Modernization of Legacy Web Applications,” arXiv preprint arXiv:2501.19204, 2025. Available: https://arxiv.org/abs/2501.19204
11. C. Diggs, M. Doyle, A. Madan, S. Scott, E. Escamilla et al., “Leveraging LLMs for Legacy Code Modernization: Challenges and Opportunities for LLM-Generated Documentation” arXiv preprint arXiv:2411.14971, 2024. Available: https://arxiv.org/abs/2411.14971