EUVIMEDEUVIMED entdecken
Crossref

A Multi-Agent AI Framework for Automated Requirement-to-Test Case Mapping in Software Quality Assurance

Shailjakant Shukla, Manav Bansal, Arpit Chhabra

International Journal of Drug Delivery Technology · 08.06.2026 · 16 (48s)

DOI: 10.25258/ijddt.16.48s.109 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Drug Delivery Technology, online erschienen am 08.06.2026. Verfasst von Shailjakant Shukla und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The complexity of today’s software systems has become critically important to ensure sufficient test coverage based on changing requirements. Conventional methods of requirements traceability and test case generation are mostly manual, time-consuming, and subject to human error, often resulting in coverage gaps which translate into post-release bugs. In this paper, a new Multi-Agent Artificial Intelligence (AI) system that could be used to automate the entire pipeline of requirement-to-test mapping is presented, and the way it would revolutionize software quality assurance practices is discussed. The proposed framework uses three collaborating intelligent agents: a Requirement Parsing Agent (RPA) that applies transformer-based Natural Language Processing (NLP) to extract semantic objects of the heterogeneous requirement documents a Traceability Mapping Agent (TMA) that builds and maintains dynamic bidirectional traceability matrices using a knowledge graph backbone and a Test Generation Agent (TGA) that synthesizes executable test cases of the traced requirement nodes using large language model (LLM) guided templates. Experimental analysis of three opensource software projects indicates that the proposed system can meet an average requirement coverage rate of 94.7, a precision of traceability link of 91.3 and a 67 percent reduction in the manual test authoring effort as compared to the baseline approaches. The framework is checked in the framework of IEEE 29148 and ISO/IEC 25010 quality standards. Findings suggest that the system performs remarkably better compared to the current automated tests systems such as NLP-based traceability recovery techniques and traditional test generation systems. This study offers a reproducible, time saving and produces standards-conforming, open-source architecture of intelligent software testing both in an Agile and traditional SDLC setting.

Zitieren

APA

Shukla, S., Bansal, M., & Chhabra, A. (2026). A Multi-Agent AI Framework for Automated Requirement-to-Test Case Mapping in Software Quality Assurance. International Journal of Drug Delivery Technology, 16(48s). https://doi.org/10.25258/ijddt.16.48s.109

Vancouver

Shukla S, Bansal M, Chhabra A. A Multi-Agent AI Framework for Automated Requirement-to-Test Case Mapping in Software Quality Assurance. International Journal of Drug Delivery Technology. 2026;16(48s). doi:10.25258/ijddt.16.48s.109

BibTeX
@article{Shukla2026,
  author = {Shukla, Shailjakant and Bansal, Manav and Chhabra, Arpit},
  title = {A Multi-Agent AI Framework for Automated Requirement-to-Test Case Mapping in Software Quality Assurance},
  journal = {International Journal of Drug Delivery Technology},
  year = {2026},
  volume = {16},
  number = {48s},
  doi = {10.25258/ijddt.16.48s.109}
}
Kostenlos starten →