Modernization of legacy systems with generative artificial intelligence: evaluation of the transformation of functional specifications into services
DOI:
https://doi.org/10.47212/rtcAlinin.4.226.10Keywords:
functional specifications, generative artificial intelligence, legacy systems, modernization, servicesAbstract
This study evaluates a generative artificial intelligence approach for transforming the functional specifications of a legacy system into software services, and determines their functional compliance, architectural usefulness, incidents, and estimated productivity compared with traditional development. An observational empirical case study was conducted on a real legacy system, whose population consisted of 1,280 services generated with the Claude Opus 4.6 model from structured specifications. Through purposive sampling by complexity, the 500 services with three or more Story Points were selected. Data were collected through automated tests, manual review, the project's incident tool, and a five-criterion rubric to assess usefulness, applied by the lead architect and settled by consensus among three architects, without computing inter-rater agreement. Times were estimated from the team's historical velocity. Of the 500 services, 425 (85%) met the specifications without adjustments and 75 required iteration; 85 incidents were identified, predominantly architectural (41%), and 430 services (86%) were classified as useful. The estimated time per component was 4 days with the generative approach and 10 days with the traditional one. It is concluded that generative artificial intelligence can accelerate the transformation of legacy systems and produce services that meet the specifications and are useful, provided it is integrated with technical validation, architectural assessment, iteration, and specialized human supervision. Because the study covers a single system and relies on estimated times, the results should be interpreted with caution.
Downloads
References
Bakar, H. K. A., Razali, R. and Jambari, D. I. (2019). Implementation phases in modernisation of legacy systems. 2019 6th International Conference on Research and Innovation in Information Systems (ICRIIS), 1–6. https://doi.org/10.1109/ICRIIS48246.2019.9073628
Chen, M., Tworek, J., Jun, H., Yuan, Q., Ponde de Oliveira Pinto, H., Kaplan, J., Edwards, H., Burda, Y., Joseph, N., Brockman, G., Ray, A., Puri, R., Krueger, G., Petrov, M., Khlaaf, H., Sastry, G., Mishkin, P., Chan, B., Gray, S. (…) Zaremba, W. (2021). Evaluating large language models trained on code. arXiv. https://doi.org/10.48550/arXiv.2107.03374
Crotty, J. and Horrocks, I. (2017). Managing legacy system costs: A case study of a meta-assessment model to identify solutions in a large financial services company. Applied Computing and Informatics, 13(2), 175–183. https://doi.org/10.1016/j.aci.2016.12.001
Diggs, C., Doyle, M., Madan, A., Scott, S., Escamilla, E., Zimmer, J., Rothermel, K. J., Sanchez, D., Robinson, S., Zwyssig, T. and Thaker, S. (2024). Leveraging LLMs for legacy code modernization: Challenges and opportunities for LLM-generated documentation. arXiv. https://doi.org/10.48550/arXiv.2411.14971
Duarte, D. y Chacón, O. (2022). Inteligencia artificial: retos y desafíos de la ética laboral en la sociedad tecnológica, Mundo Fesc, vol 12 no. S3 pp. 265-280 de 2022
Esposito, M., Li, X., Moreschini, S., Ahmad, N., Cerny, T., Vaidhyanathan, K., Lenarduzzi, V. and Taibi, D. (2026). Generative AI for software architecture. Applications, challenges, and future directions. Journal of Systems and Software, 231, 112607. https://doi.org/10.1016/j.jss.2025.112607
Etikan, I., Musa, S. A. and Alkassim, R. S. (2016). Comparison of convenience sampling and purposive sampling. American Journal of Theoretical and Applied Statistics, 5(1), 1–4. https://doi.org/10.11648/j.ajtas.20160501.11
Fávero, L. F., de Almeida, N. R. and Affonso, F. J. (2025). A systematic mapping study on the modernization of legacy systems to microservice architecture. Applied System Innovation, 8(4), 86. https://doi.org/10.3390/asi8040086
Gandhi, S., Patwardhan, M., Khatri, J., Vig, L. and Medicherla, R. K. (2024). Translation of low-resource COBOL to logically correct and readable Java leveraging high-resource Java refinement. Proceedings of the 1st International Workshop on Large Language Models for Code, 46–53. https://dl.acm.org/doi/10.1145/3643795.3648388
Gholami, M. F., Daneshgar, F., Beydoun, G. and Rabhi, F. (2017). Challenges in migrating legacy software systems to the cloud — an empirical study. Information Systems, 67, 100–113. https://doi.org/10.1016/j.is.2017.03.008
Hadjichristofi, C., Tsilimigkounakis, M., Sotiropoulos, G. and Vescoukis, V. (2026). An empirical evaluation of large language models applying software architectural patterns. AI, 7(6), 195. https://doi.org/10.3390/ai7060195
Hasan, M. H., Osman, M. H., Admodisastro, N. I. and Muhammad, M. S. (2023). Legacy systems to cloud migration: A review from the architectural perspective. Journal of Systems and Software, 202, 111702. https://doi.org/10.1016/j.jss.2023.111702
Hou, X., Zhao, Y., Liu, Y., Yang, Z., Wang, K., Li, L., Luo, X., Lo, D., Grundy, J. and Wang, H. (2024). Large language models for software engineering: A systematic literature review. ACM Transactions on Software Engineering and Methodology, 33(8), 220. https://doi.org/10.1145/3695988
Irani, Z., Abril, R. M., Weerakkody, V., Omar, A. and Sivarajah, U. (2023). The impact of legacy systems on digital transformation in European public administration: Lesson learned from a multi case analysis. Government Information Quarterly, 40(1), 101784. https://doi.org/10.1016/j.giq.2022.101784
Kavitha, V., Gowri Laksshmi, K. S., Hemashree, P. and Shobana, R. (2025). Empowering software engineering automation through explainable AI. En Advances in Computers. Elsevier. https://doi.org/10.1016/bs.adcom.2025.07.009
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W.-t., Rocktäschel, T., Riedel, S. and Kiela, D. (2020). Retrieval-augmented generation for knowledge-intensive NLP tasks. Advances in Neural Information Processing Systems, 33. https://doi.org/10.48550/arXiv.2005.11401
Margara, A. and Salvaneschi, G. (2018). On the semantics of distributed reactive programming: The cost of consistency. IEEE Transactions on Software Engineering, 44(7), 689–711. https://doi.org/10.1109/TSE.2018.2833109
Nguyen-Duc, A., Cabrero-Daniel, B., Przybylek, A., Arora, C., Khanna, D., Herda, T., Rafiq, U., Melegati, J., Guerra, E., Kemell, K.-K., Saari, M., Zhang, Z., Le, H., Quan, T. and Abrahamsson, P. (2023). Generative artificial intelligence for software engineering – A research agenda. arXiv. https://doi.org/10.48550/arXiv.2310.18648
Peng, S., Kalliamvakou, E., Cihon, P. and Demirer, M. (2023). The impact of AI on developer productivity: Evidence from GitHub Copilot. arXiv. https://arxiv.org/pdf/2302.06590
Pérez-Castillo, R., De Guzmán, I. G.-R. and Piattini, M. (2011). Knowledge discovery metamodel-ISO/IEC 19506: A standard to modernize legacy systems. Computer Standards & Interfaces, 33(6), 519–532. https://doi.org/10.1016/j.csi.2011.02.007
Ramírez, A. (2025). Globalization, interconnection or sustainability. Revista Temario Científico, 5(1), 1–3. https://doi.org/10.47212/rtcAlinin.3.125.1
Runeson, P. and Höst, M. (2009). Guidelines for conducting and reporting case study research in software engineering. Empirical Software Engineering, 14(2), 131–164. https://doi.org/10.1007/s10664-008-9102-8
Sandborn, P. A. and Prabhakar, V. J. (2015). The forecasting and impact of the loss of critical human skills necessary for supporting legacy systems. IEEE Transactions on Engineering Management, 62(3), 361–371. https://doi.org/10.1109/TEM.2015.2438820
Sergeyuk, A., Golubev, Y., Bryksin, T. and Ahmed, I. (2025). Using AI-based coding assistants in practice: State of affairs, perceptions, and ways forward. Information and Software Technology, 178, 107610. https://doi.org/10.1016/j.infsof.2024.107610
Tawosi, V., Moussa, R. and Sarro, F. (2022). On the relationship between story points and development effort in agile open-source software. Proceedings of the 16th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM '22). https://doi.org/10.1145/3544902.3546238
Tawosi, V., Moussa, R. and Sarro, F. (2023). Agile effort estimation: Have we solved the problem yet? Insights from a replication study. IEEE Transactions on Software Engineering, 49(4), 2677–2697. https://doi.org/10.1109/TSE.2022.3228739
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Yeismer Cornelio Espejo Bohorquez
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
