Salvato in:
Dettagli Bibliografici
Autori principali: Herd, Benjamin, Kelly, Jessica, Heinemann, Clarissa, Zacchi, João-Vitor
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2605.22530
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913152848887808
author Herd, Benjamin
Kelly, Jessica
Heinemann, Clarissa
Zacchi, João-Vitor
author_facet Herd, Benjamin
Kelly, Jessica
Heinemann, Clarissa
Zacchi, João-Vitor
contents We present a method for dynamic quantitative assurance that enhances static safety cases with continuous, runtime-driven confidence updates. The method quantifies and propagates confidence across the development lifecycle by integrating design-time evidence and windowed runtime Safety Performance Indicators (SPIs) within a single Subjective Logic (SL)-based assurance case. At runtime, SPI evidence is continuously evaluated, and targeted claims are updated using a rule that increases confidence in the absence of violations and imposes prompt penalties when violations occur. This design prioritizes safety-relevant responsiveness over exact classical Bayesian posterior updates. We demonstrate the method using a simulation-based construction zone assist function, focusing on an ML-based construction cone detection component, and show how confidence evolves as SPI evidence is observed in operation.
format Preprint
id arxiv_https___arxiv_org_abs_2605_22530
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Subjective Logic-based method for runtime confidence updates in safety arguments
Herd, Benjamin
Kelly, Jessica
Heinemann, Clarissa
Zacchi, João-Vitor
Artificial Intelligence
We present a method for dynamic quantitative assurance that enhances static safety cases with continuous, runtime-driven confidence updates. The method quantifies and propagates confidence across the development lifecycle by integrating design-time evidence and windowed runtime Safety Performance Indicators (SPIs) within a single Subjective Logic (SL)-based assurance case. At runtime, SPI evidence is continuously evaluated, and targeted claims are updated using a rule that increases confidence in the absence of violations and imposes prompt penalties when violations occur. This design prioritizes safety-relevant responsiveness over exact classical Bayesian posterior updates. We demonstrate the method using a simulation-based construction zone assist function, focusing on an ML-based construction cone detection component, and show how confidence evolves as SPI evidence is observed in operation.
title A Subjective Logic-based method for runtime confidence updates in safety arguments
topic Artificial Intelligence
url https://arxiv.org/abs/2605.22530