MBSE-AI Integration 2024 Abstracts

Area 1 - MBSE-AI Integration
Full Papers Short Papers

Area 1 - MBSE-AI Integration

Full Papers
Paper Nr: 6
Title:

On the Formal Robustness Evaluation for AI-based Industrial Systems
Authors:

Mohamed I. Khedher, Afef Awadid, Augustin Lemesle and Zakaria Chihani
Abstract: The paper introduces a three-stage evaluation pipeline for ensuring the robustness of AI models, particularly neural networks, against adversarial attacks. The first stage involves formal evaluation, which may not always be feasible. For such cases, the second stage focuses on evaluating the model’s robustness against intelligent adversarial attacks. If the model proves vulnerable, the third stage proposes techniques to improve its robustness. The paper outlines the details of each stage and the proposed solutions. Moreover, the proposal aims to help developers build reliable and trustworthy AI systems that can operate effectively in critical domains, where the use of AI models can pose significant risks to human safety.
Download
Paper Nr: 7
Title:

AI Systems Trustworthiness Assessment: State of the Art
Authors:

Afef Awadid, Kahina Amokrane-Ferka, Henri Sohier, Juliette Mattioli, Faouzi Adjed, Martin Gonzalez and Souhaiel Khalfaoui
Abstract: Model-based System Engineering (MBSE) has been advocated as a promising approach to reduce the complexity of AI-based systems development. However, given the uncertainties and risks associated with Artificial Intelligence (AI), the successful application of MBSE requires the assessment of AI trustworthiness. To deal with this issue, this paper provides a state of the art review of AI trustworthiness assessment in terms of trustworthiness attributes/ characteristics and their corresponding evaluation metrics. Examples of such attributes include data quality, robustness, and explainability. The proposed review is based on academic and industrial literature conducted within the Confiance.ai research program.
Download
Paper Nr: 8
Title:

Bringing Systems Engineering Models to Large Language Models: An Integration of OPM with an LLM for Design Assistants
Authors:

Ramón María García Alarcia, Pietro Russo, Alfredo Renga and Alessandro Golkar
Abstract: Although showing remarkable zero-shot and few-shot capabilities across a wide variety of tasks, Large Language Models (LLMs) are still not mature enough for off-the-shelf use in engineering design tasks. Organizations implementing model-based systems engineering practices into their product development processes can leverage on ontologies, models, and procedures to enhance LLMs applied to engineering design tasks. We present a methodology to integrate an Object-Process Methodology model of a space system into an LLM-based spacecraft design assistant and show a performance improvement, as compared to a conventional LLM. The benchmark is evaluated through subjective expert-assessed and an objective cosine-similarity-based criteria. The results motivate additional efforts in integrating Model-Based Systems Engineering practice into LLMs as means to improve their performance and reduce shortcomings such as hallucinations and black-box, untraceable behavior.
Download
Short Papers
Paper Nr: 5
Title:

MBSE-Enhanced LSTM Framework for Satellite System Reliability and Failure Prediction
Authors:

M. A. Alandihallaj, Mahya Ramezani and Andreas M. Hein
Abstract: This paper investigates the integration of Artificial Intelligence (AI) and Model-Based Systems Engineering (MBSE) in the field of satellite system reliability. We employ Long Short-Term Memory (LSTM) networks, an AI technique, to predict the failure probabilities of various subsystems. These LSTM models are integrated into an MBSE framework, enhancing the accuracy of system-wide failure prediction and operational decision-making. The approach involves training LSTM networks on simulated datasets representing a range of operational scenarios for each subsystem. The outputs from these networks are then aggregated using a weighted approach to determine the optimal disposal time, aiming to extend the satellite’s operational lifespan. The performance of the system is evaluated a simulated real mission scenario. This research highlights the potential of AI-MBSE integration in advancing satellite system design and maintenance strategies.
Download
Paper Nr: 9
Title:

AI Engineering for Trust by Design
Authors:

André Meyer-Vitali
Abstract: The engineering of reliable and trustworthy AI systems needs to mature. While facing unprecedented challenges, there is much to be learned from other engineering disciplines. We focus on the four pillars of (i) Models & Explanations, (ii) Causality & Grounding, (iii) Modularity & Compositionality, and (iv) Human Agency & Oversight. Based on these pillars, a new AI engineering disciple could emerge, which we aim to support using corresponding methods and tools for “Trust by Design”.
Download
Paper Nr: 10
Title:

Compliance by Design for Cyber-Physical Energy Systems: The Role of Model-Based Systems Engineering in Complying with the EU AI Act
Authors:

Dominik Vereno, Katharina Polanec and Christian Neureiter
Abstract: In the evolving landscape of intelligent power grids, artificial intelligence (AI) plays a crucial role, yet its integration into critical infrastructure poses significant risks. The new EU AI Act, regulating such high-risk applications, introduces stringent requirements such as risk management and data governance. This study aims to harness the potential of model-based systems engineering (MBSE) for enabling compliance by design in smart grids, ensuring adherence to regulation from early development stages. Through a detailed analysis of the AI Act’s seven requirement for high-risk applications, the paper aligns them with established MBSE practices. The findings reveal MBSE as an effective tool for ensuring compliance, leading to three strategic recommendations: integrating mature disciplines into holistic MBSE approaches, establishing a broadly accepted AI modeling formalism, and creating a standardized model-based compliance assessment process. In conclusion, MBSE is a key enabler for creating dependable and safe AI applications, offering a positive outlook for future smart grid developments that are innovative yet compliant by design.
Download
Paper Nr: 11
Title:

Facilitating User-Centric Model-Based Systems Engineering Using Generative AI
Authors:

Elias Bader, Dominik Vereno and Christian Neureiter
Abstract: The increasing complexity of cyber-physical systems requires model-based systems engineering (MBSE) in an effort to sustain a comprehensive oversight. However, broader adaptation of these models requires specialized knowledge and training. In order to make this process more user-friendly, the concept of user-centric systems engineering emerged. Artificial intelligence (AI) could help users overcome beginner hurdles and leverage their contribution quality. This research investigates the feasibility of a large language model in the systems engineering context, with a particular emphasis on the identification of potential obstacles for similar tasks. Therefore, a GPT model is trained on a dataset consisting of UML component diagram elements. In conclusion, the promising results of this research justify utilizing AI in MBSE. Complex relationships between the UML elements were not only understood, they were also generated using natural-language text. Problems arise from the extensive nature of the XMI, the context limitation and the unique identifiers of the UML elements. The fine-tuning process enabled the LLM to gain valuable insights into UML modeling while transferring their base knowledge, which is a promising step toward reducing complexity in MBSE.
Download
Paper Nr: 13
Title:

On Some Artificial Intelligence Methods in the V-Model of Model-Based Systems Engineering
Authors:

Stephan Rudolph
Abstract: The enhancement of the standard V-Model of Model-Based Systems Engineering (MBSE) with methods from Artificial Intelligence (AI) is currently in the research focus of many universities and engineering companies. In the need to find new means to deal with the steadily increasing complexity in modern systems engineering and design of complex systems, traditional MBSE methods, most notably the standard V-Model of MBSE, is a candidate to be enriched with several AI methods. The work presented summarizes the experience gained with several AI methods in a machine-executable version of the V-Model of MBSE based on an graph-based design language approach to design automation and tries to summarize the resulting shift in the engineering burden and effort as well as the observed gains in design time and quality. Based on the results from engineering practice and some theoretical foundations underlying engineering as a branch of the natural sciences per se, an outlook is attempted on some necessary aspects in future developments of AI for engineering applications.
Download
Paper Nr: 12
Title:

On the Artificial Reasoning with Chess: A CBR vs PBR View
Authors:

Zahira Ghalem, Karima Berramla, Thouraya Bouabana-Tebibel and Djamel E. Zegour
Abstract: In the quest to advance artificial reasoning, this article delves into the contrasting realms of Case-Based Reasoning (CBR) and Pattern-Based Reasoning (PBR). Drawing inspiration from human thinking behavior in tackling novel problems. The study centers on the chess domain, exploring the intricacies of representation, generalization, and reasoning processes. It illuminates the fundamental trade-off between computational efficiency and decision quality in (PBR) systems. This comprehensive examination provides valuable insights into the adaptability of reasoning systems and the role of abstract knowledge bases in enhancing performance.
Download