RegComp

Supporting regulatory requirements engineering through AI-assisted regulatory change analysis

RegComp

Software-intensive systems operating in regulated domains must continuously adapt to evolving laws, standards, and policies. Regulatory changes often affect multiple software artifacts, including requirements, architecture, source code, documentation, and test assets. Understanding the consequences of these changes is a complex and time-consuming task that is still largely performed manually. The RegComp (Regulatory Compliance) project investigates AI-assisted approaches that support regulatory requirements engineering throughout the lifecycle of software systems. By combining natural language processing, information retrieval, and large language models, the project develops methods for automatically identifying regulatory changes, analyzing their impact on software artifacts, and supporting engineers in making informed implementation decisions. The overall goal is to reduce manual effort, improve traceability, and enable faster adaptation to evolving regulatory requirements.

Project description

The RegComp project develops novel methods for supporting regulatory requirements engineering through automated regulatory change analysis and impact assessment. Instead of focusing solely on verifying compliance, the project investigates how changes in regulations can be systematically analyzed to understand their implications for software systems.

The research combines advances in natural language processing, semantic search, information retrieval, and large language models to automatically identify regulatory changes, characterize their type and scope, and determine which software artifacts are likely to be affected. The developed approaches support engineers in prioritizing implementation activities, tracing the rationale behind regulatory updates, and understanding the relationships between evolving regulations and software assets.

Particular emphasis is placed on explainable AI-assisted decision support that enables engineers to efficiently assess regulatory changes while maintaining transparency and confidence in the generated results. The developed methods are evaluated using real-world industrial case studies to ensure their practical applicability and usefulness in regulated software development.

Research contribution

The RegComp project advances the state of the art in regulatory requirements engineering by introducing AI-assisted methods for automated regulatory change analysis and software impact assessment. The project contributes models, techniques, and tool-supported approaches that enable organizations to better understand how evolving regulations influence software systems throughout their lifecycle.

The research contributes methods for automatically detecting and categorizing regulatory changes, assessing their impact on software artifacts, supporting traceability between regulations and engineering assets, and providing explainable recommendations for software evolution. By integrating artificial intelligence with regulatory engineering practices, the project aims to improve the efficiency, scalability, and transparency of regulatory change management while reducing the manual effort required to maintain compliance in continuously evolving regulatory environments.

Funding

Project duration

01.01.2023 – 31.12.2026

Contact

Project partner

Publications

  • 2026 Investigating Automated Change Impact Analysis in FinTech Regulations Parisa Elahidoost , Hugo Villamizar , Florian Angermeir , Jonathan Streit , Daniel Mendez , Michael Unterkalmsteiner and Tony Gorschek Information and Software Technology Journal, ():, 2026. Details URL DOI BIB
  • 2024 Designing NLP-based solutions for requirements variability management: experiences from a design science study at Visma Parisa Elahidoost , Michael Unterkalmsteiner , Davide Fucci , Jannik Fischbach and Peter Liljenberg , 2024. The 30th International Conference on Requirement Engineering: Foundation for Software Quality (REFSQ). Details URL DOI BIB
  • 2024 Practices, Challenges, and Opportunities When Inferring Requirements From Regulations in the FinTech Sector - An Industrial Study Parisa Elahidoost , Daniel Mendez , Michael Unterkalmsteiner , Jannik Fischbach , Christian Feiler and Jonathan Streit In 9th International Workshop on Empirical Requirements Engineering, 2024. IEEE. Details DOI BIB
click to return to top of page