OntoGSN: An Ontology for Dynamic Management of Assurance Cases
Tomas Bueno Momcilovic, Barbara Gallina, Ingmar Kessler, Dian Balta · 2025
Read OntoGSN: An Ontology for Dynamic Management of Assurance Cases by Tomas Bueno Momcilovic, Barbara Gallina, Ingmar Kessler, Dian Balta (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-11023.txt.
Introduction
Tomas Bueno Momcilovic and co-authors turn assurance cases into a queryable web of facts, so safety arguments can be kept live as systems change. Their 2025 paper deals with the structured cases that firms build to trust a product's safety or strength. Hand-kept cases rot fast when the system shifts, and false trust can follow.
Their kit, OntoGSN, rests on an ontology, a formal map of the standard notation for such cases. It mirrors the community standard rule by rule, adds a bridge to a popular case tool, and ships a library of ready queries. Graphs built this way can be filled, checked, and updated by machine. A test interface shows the workflow.
Catalog record
| CREATOR | Tomas Bueno Momcilovic, Barbara Gallina, Ingmar Kessler, Dian Balta |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-11023
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
OntoGSN: An Ontology for Dynamic Management of Assurance Cases Tomas Bueno Momcilovic, Barbara Gallina, Ingmar Kessler, Dian Balta Abstract Assurance cases (ACs) are a common artifact for building and maintaining confidence in system properties such as safety or robustness. Constructing an AC can be challenging, although existing tools provide support in static, document-centric applications and methods for dynamic contexts (e.g., autonomous driving) are emerging. Unfortunately, managing ACs remains a challenge, since maintaining the embedded knowledge in the face of changes requires substantial effort, in the process deterring developers—or worse, producing poorly managed cases that instill false confidence. To address this, we present OntoGSN: an ontology and supporting middleware for managing ACs in the Goal Structuring Notation (GSN) standard. OntoGSN offers a knowledge representation and a queryable graph that can be automatically populated, evaluated, and updated. Our contributions include: a 1:1 formalization of the GSN Community Standard v3 in an OWL ontology with SWRL rules; a helper ontology and parser for integration with a widely used AC tool; a repository and documentation of design decisions for OntoGSN maintenance; a SPARQL query library with automation patterns; and a prototypical interface. The ontology strictly adheres to the standard’s text and has been evaluated according to FAIR principles, the OOPS! framework, competency questions, and community feedback. The development of other middleware elements is guided by the community needs and subject to ongoing evaluations. To demonstrate the utility of our contributions, we illustrate dynamic AC management in an example involving assurance of adversarial robustness in large language models. 1 Introduction Assurance cases (ACs) represent a widely accepted artifact for documenting, building and maintaining confidence in a property of a system, such as the safety of a rocket [ 42 ] or the security of a software application [ 21 ] . ACs are defined as a combination of "arguments and evidence intended to demonstrate that a system meets its assurance requirements" (p. 128, [ 3 ] ). They provide an interface for developers and auditors to verify and validate that there is enough evidence to conclude that the targeted property is achieved. Originally conceived for the safety certification of products, ACs have since proven valuable for assuring more opaque properties and systems such as frontier large language models [ 16 ] . However, while much literature has been published on how to create a good AC (cf. [ 30 ] ), managing them effectively is still a substantial challenge. Assurance arguments are handled as standalone graphs of statements and documents, separate from active processes or the evidence itself. The tool ecosystem is accordingly centered on visualizing links of natural language statements and documents 1 1 1 For examples, see: https://scsc.uk/index.php/tools . The increase in autonomy, agency, and machine learning has motivated a proliferation of ACs for systems in more dynamic and data-intensive settings [ 2 ] , but they remain largely separate from the software they are assuring. To directly support developers in managing their ACs, we present the OntoGSN ontology and the supporting middleware toolkit. As a comprehensive formalization of the Goal Structuring Notation (GSN) 2 2 2 GSN is one of the widely used metamodels for creating assurance cases and arguments, and has been chosen for its availability of examples in practice. Other well-known standards include the Structured Assurance Case Metamodel [ 24 ] , and the Claim-Arguments-Evidence framework (cf. [ 6 ] ). , the aim of OntoGSN is to provide the middleware to domain experts for easier integration of data streams, domain ontologies, and software processes that affect the elements of an AC. The middleware serves as a complement to existing tools, enabling direct integration with a commonly used software application, and the use of Semantic Web reasoners such as Pellet. Where a graphical user interface for assurance case management is not needed, OntoGSN can be used standalone in an ontology manager or integrated in a graph database. Our contributions include: 1. A 1:1 representation of the GSN Community Standard v3 [ 3 ] in OWL and SWRL; 2. A repository and documentation for maintaining OntoGSN; 3. An ontology and parser for aligning OntoGSN with the Assurance and Safety Case Environment (ASCE) [ 1 ] ; 4. A library of SPARQL queries as templates for common automation settings based on competency questions; 5. Jupyter notebooks as prototypical user interfaces integrating all components. The implications for practice are twofold: on one hand, engineers can directly apply OntoGSN in their practice, and build upon off-the-shelf tools. On the other hand, they can tailor OntoGSN to the context of their assurance tasks through the proposed argument-pattern, module, confidence-argument, and dialectic extensions. 2 Related Work The use of ontologies for managing assurance cases is not a novel idea, but current implementations are limited and publicly unavailable. The Structured Assurance Case Metamodel v1.1 [ 23 ] , for example, originally provided mechanisms to link external OWL domain ontologies. The ECSEL AMASS project [ 29 , 35 , 34 ] was an open-source certification platform that supported ontologies based on Open Services for Lifecycle Collaboration [ 15 ] , which connected assurance cases with artifacts, pioneering the benefit of ontology-driven AC management (cf. [ 14 ] ). Denney (2020) [ 12 ] explored the benefits of domain-specific ontologies in their AdvoCATE tool [ 11 ] , which supports GSN cases. More recently, MISRA (2024) [ 20 ] defined custom domain ontologies within SysML to shape assurance cases for automated driving. Unfortunately, the results of these related initiatives have either been deprecated or publicly unavailable. Similar issues apply to the tools themselves. Software such as ASCE [ 1 ] and AdvoCaTE [ 12 ] are widely used by assurance experts, and provide useful graphical user interfaces. However, these tools closed-source and often proprietary; being specifically geared towards ACs, their adoption by a wider audience and thereby extensibility may be limited. Additionally, the logic and coverage of performed evaluations (i.e., tests or rules) in these tools is not clear. Related academic resources show promise in addressing some of these issues, and supporting more dynamic ACs development and management. Castillo-Effen et al. [ 7 ] , for example, introduce an algorithmic framework and domain-specific language for managing ACs in the aerospace domain with data; Beyene & Carlan [ 5 ] similarly introduce a the CyberGSN language for enabling features such as digital signatures for cybersecurity ACs. Saswata et al. [ 25 ] provide a triplestore-based framework (i.e., graph database) with patterns that can be instantiated into ACs using domain ontologies. Others, such as Asaadi et al. [ 2 ] , give architectural and development lifecycle recommendations for data-driven applications that are tool-agnostic. We see the positioning of OntoGSN as highly complementary in following ways. First, the related work contributes with focused representations of GSN that fulfill a particular purpose. We instead provide a complete, comprehensive, and source-driven formalization of GSN that can be integrated into any framework, distilled into smaller models for a particular purpose, or aligned with other AC metamodels. Second, the OntoGSN ontology (alongside supporting tools) is public, free and open-source, thereby allowing easy extension of facts and rules from the domain, and implementation of AC-driven assurance before deciding for more advanced tools such as ASCE. Third, and