AI-Driven Agile Systems Engineering Approach for Managing Cross-System Interactions
Bibliographic record
Abstract
The continuous advancement of embedded modules modernization, alongside the growing trends of electrification and connectivity, has driven a transformation in product development processes across various industries to meet the expectations of the evolving market demands. However, integrating and managing the complexity of emerging experiences with existing systems remains challenging, as issues such as incomplete and ambiguous requirements gathering can compromise the quality of the user experience and put the success of the entire project at risk. Although recent studies have explored uncertainty and adaptability in systems development, current solutions still lack a robust methodology that can identify and manage cross-system interactions. This work explores the integration of the agile development process with Artificial Intelligence (AI) to abstract the complexity of interactions between new and existing systems, ensuring the completeness of the requirements that govern these interactions. The proposed approach leverages system use cases from an end-to-end perspective to hypothesize interactions with other systems under consideration, using Large Language Models (LLMs) to identify and verify these hypotheses against requirements supporting the use cases, thereby ensuring requirements completeness. An automotive industry case study evaluated this approach for three use cases of a system, revealing that 75% of LLM-identified hypotheses for one of the use cases were opportunities for improvement in the system requirements.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".