Criteria to Systematically Evaluate (Safety) Assurance Cases
Bibliographic record
Abstract
An assurance case (AC) captures explicit reasoning associated with assuring critical properties, such as safety. A vital attribute of an AC is that it facilitates the identification of fallacies in the validity of any claim. There is considerable published research related to confidence in ACs, which primarily relate to a measure of soundness of reasoning. Evaluation of an AC is more general than measuring confidence and considers multiple aspects of the quality of an AC. Evaluation criteria thus play a significant role in making the evaluation process more systematic. This paper contributes to the identification of effective evaluation criteria for ACs, the rationale for their use, and initial tests of the criteria on existing ACs. We classify these criteria as to whether they apply to the structure of the AC, or to the content of the AC. This paper focuses on safety as the critical property to be assured, but only a very small number of the criteria are specific to safety, and can serve as placeholders for evaluation criteria specific to other critical properties. All of the other evaluation criteria are generic. This separation is useful when evaluating ACs developed using different notations, and when evaluating ACs against safety standards. We explore the rationale for these criteria as well as the way they are used by the developers of the AC and also when they are used by a third-party evaluator.
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 | 0.080 | 0.382 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.025 | 0.009 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.008 | 0.007 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.003 | 0.002 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
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".