INCOSE International Symposium最新文献_第5页

Enhancing Model-Based Systems Engineering with Large Language Models 用大语言模型增强基于模型的系统工程

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70067

Khushnood Adil Rafique, Sanan Shah, Šandor Dalecke, Christoph Grimm

{"title":"Enhancing Model-Based Systems Engineering with Large Language Models","authors":"Khushnood Adil Rafique, Sanan Shah, Šandor Dalecke, Christoph Grimm","doi":"10.1002/iis2.70067","DOIUrl":"https://doi.org/10.1002/iis2.70067","url":null,"abstract":"Model-Based Systems Engineering (MBSE) is a cornerstone of modern systems engineering, enabling the design and management of increasingly complex systems. However, creating models with modeling languages such as System Modeling Language (SysML) V2 remains a highly manual and time-intensive process. As system complexity increases and interdisciplinary collaboration becomes critical, streamlining workflows and accelerating the modeling process are more important than ever. This paper investigates the use of two classes of Large Language Models (LLMs)— the general-purpose reasoning model, GPT-4, and the domain specific fine-tuned model, CodeT5, to assist in generating SysML V2 models from natural language descriptions. Our findings reveal that while LLMs can generate initial “skeleton” models, relying solely on AI for entire model generation is not advisable due to critical errors and inconsistencies in the generated outputs. Instead, LLMs serve as effective tools for accelerating the modeling process by producing foundational prototypes that can be refined and improved through human intervention. With fine-tuning techniques for CodeT5 and prompt engineering for GPT-4, we evaluate their performance across key dimensions of syntax accuracy, semantic alignment, and structural correctness. Using a diverse set of metrics and statistical analyses, we identify the trade-offs between task-specific optimization (CodeT5) and broader contextual reasoning (GPT-4). The potential of LLMs as accelerators in MBSE workflows cannot be ignored with the “early” modeling capabilities they present. This work contributes to the growing intersection of AI and MBSE, offering practical insights into the selection and application of LLMs in systems engineering and paving the way for future advancements in the field.","PeriodicalId":100663,"journal":{"name":"INCOSE International Symposium","volume":"35 1","pages":"1523-1543"},"PeriodicalIF":0.0,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

THE COST OF EXPERTISE: PERFORMANCE TRADEOFFS IN LLMs FOR SYSTEMS ENGINEERING 专业知识的成本：系统工程法学硕士的性能权衡

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70078

Paul Wach, Ryan Bell, Brady Jugan, Ryan Longshore, Raymond Madachy

{"title":"THE COST OF EXPERTISE: PERFORMANCE TRADEOFFS IN LLMs FOR SYSTEMS ENGINEERING","authors":"Paul Wach, Ryan Bell, Brady Jugan, Ryan Longshore, Raymond Madachy","doi":"10.1002/iis2.70078","DOIUrl":"https://doi.org/10.1002/iis2.70078","url":null,"abstract":"In the systems engineering (SE) community, generative artificial intelligence (GenAI), such as large language models (LLMs), continues to grow in popularity with applications varying from help with writing requirements, to creating traditional text-based documents, and to generating models. Furthermore, the text-based format of the Systems Modeling Language version 2 (SysMLv2) gives rise to wide ranging possibilities when combined with LLMs. It is common to specialize LLMs on specific tasks, such as SE; which may be achieved by a number of means. Fine-tuning and retrieval augmented generation (RAG) are example methods to teach a model how to act and what the model needs to know, respectively. Typical practice evaluates LLMs relative to one-another through the use of benchmarks. Benchmarks are tasks that use a common scoring-based method where the LLM either answers a set of multiple-choice questions or textual queries to verified, correct answers. A domain specific benchmark for SE, SysEngBench, was developed over the last year. With many in the SE community developing custom language models, the usage of a standard benchmark is essential for relative performance comparison. We have conducted a limited experiment using SysEngBench to compare performance, on SE tasks, of a small set of LLMs (control plus two modified models). The intent of this study was to set a baseline for ranking models according to performance on SE tasks, the results suggest what many in the SE community may find intuitively unexpected.","PeriodicalId":100663,"journal":{"name":"INCOSE International Symposium","volume":"35 1","pages":"1490-1507"},"PeriodicalIF":0.0,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SysML4Sec – Methodology for Security modeling in the context of large-scale product development with multiple design levels SysML4Sec—在具有多个设计级别的大规模产品开发环境中进行安全建模的方法

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70108

Hartmut Hintze, Alice Santin, Marvin Blecken, Daniel Patrick Pereira, Ralf God

{"title":"SysML4Sec – Methodology for Security modeling in the context of large-scale product development with multiple design levels","authors":"Hartmut Hintze, Alice Santin, Marvin Blecken, Daniel Patrick Pereira, Ralf God","doi":"10.1002/iis2.70108","DOIUrl":"https://doi.org/10.1002/iis2.70108","url":null,"abstract":"Increasing functionalities and the rising complexity of aircraft systems used by flight and cabin crews, passengers, maintenance staff, and other stakeholders necessitate a reconsideration of system development methodologies. This paper presents SysML4Sec, a model-based security engineering method that integrates the systems engineering process (SEP) with the security engineering process (SecEP). Both processes operate concurrently and interactively within the same system model across all development levels. We detail SysML4Sec for model-based security engineering using a consistent SysML approach and demonstrate SEP-SecEP interactions in developing a passenger seat connectivity function with the SysML4Sec supporting ‘CATIA Magic Systems of Systems Architect’ tool. The ‘Magic Systems of Systems Architect’ tool provides a single source of truth where a system and security engineers could collaborate to design the system architecture and the associated preliminary security assessments as per the DO 326A aviation standard. It enables to define Assets to be protected, the Threat Conditions (TC), the Threat Scenarios (TS) and Security Measures (SM) which are specified in terms of effectiveness to reduce the risk.","PeriodicalId":100663,"journal":{"name":"INCOSE International Symposium","volume":"35 1","pages":"595-612"},"PeriodicalIF":0.0,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248754","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

On The Importance of Being Able to Hold a Stake 论持有股份的重要性

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70002

Richard Beasley, Andy Pickard, James Nolan, Sarah Sheard

引用次数: 0

Relationship between Adaptability and Resilience 适应性与弹性的关系

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70090

Haifeng Zhu, Ken Cureton, John Brtis, Eileen Patrice Arnold, Scott Jackson

引用次数: 0

Integration of MBSE and Agile Development by Seamlessly Creating Test Plans from Model Simulations in SDV Development 通过在SDV开发中从模型模拟中无缝创建测试计划，集成MBSE和敏捷开发

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70101

Daisuke Hashimoto, Ken Kawamura, Yutaro Ito, Ho Kit Robert Ong

{"title":"Integration of MBSE and Agile Development by Seamlessly Creating Test Plans from Model Simulations in SDV Development","authors":"Daisuke Hashimoto, Ken Kawamura, Yutaro Ito, Ho Kit Robert Ong","doi":"10.1002/iis2.70101","DOIUrl":"https://doi.org/10.1002/iis2.70101","url":null,"abstract":"With the advent of software-defined vehicles (SDVs) such as self-driving cars and battery electric vehicles (BEVs), the automotive industry faces the most drastic change over the last hundred (100) years. Over-the-Air (OTA) updates at high frequencies are expected to enhance SDVs' functionality and the connected car concept aims to create a new value by allowing cars to communicate with their external environment. On the other hand, the complexity of these systems is significantly increasing due to the number of software and the connected car concept required. To ensure the quality and enhance the agility of the SDV development, many automotive manufacturers have tried to introduce Model-Based Systems Engineering (MBSE) and Agile development approaches concurrently, but achieving a balance between these two approaches is a challenge. This paper proposes the methodology and practices to combine MBSE and Agile development by connecting models and simulation to verification planning seamlessly. We use the Systems Modeling Language (SysML) to model and analyze the requirements, logical architecture, test cases, and the System Architecture models to capture engineering information that reflects verification planning and actual verification. Since the requirements and needed verifications evolve continuously in Agile development, the proposed approach can ensure that the requirements, design, and test cases are updated alongside each iteration without sacrificing agility. The proposed approach allows high-level reuse of models and simulation to contribute to verification planning to achieve quality while maintaining focus on agility.","PeriodicalId":100663,"journal":{"name":"INCOSE International Symposium","volume":"35 1","pages":"1873-1885"},"PeriodicalIF":0.0,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A TMBR-based, Semiformal Method for Early Requirements Definition of Training Simulators 基于tmbr的训练模拟器早期需求定义半形式化方法

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70040

Władysław Sowul

{"title":"A TMBR-based, Semiformal Method for Early Requirements Definition of Training Simulators","authors":"Władysław Sowul","doi":"10.1002/iis2.70040","DOIUrl":"https://doi.org/10.1002/iis2.70040","url":null,"abstract":"This paper introduces a semiformal methodology for defining requirements early for training simulators, leveraging the True Model-Based Requirements (TMBR) paradigm. Training simulators, often developed concurrently with their target systems, require solution-agnostic and system-independent requirements. This dual development process is challenged by incomplete knowledge of the simulated system's behavior and design. The proposed method addresses these challenges by adopting a human-centric approach to model key input/output requirements, ensuring the simulator replicates the target system's functional and experiential characteristics.The methodology enables iterative transformations from the simulated system's models to the training simulator's requirements, allowing continuous updates as the target system evolves. Sensory-based interface modeling and signal decomposition are used to accommodate human factors while maintaining unambiguous, verifiable requirements. Stakeholder needs and constraints are integrated, ensuring practical alignment with real-world applications.The findings align with INCOSE SE Vision 2035, promoting adaptable and interconnected systems engineering practices. By integrating human-centric and model-based techniques, the method enhances the rigor and completeness of requirements engineering, contributing to the advancement of Model-Based Systems Engineering (MBSE).","PeriodicalId":100663,"journal":{"name":"INCOSE International Symposium","volume":"35 1","pages":"1126-1143"},"PeriodicalIF":0.0,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Helping Future Nuclear Power Facilities Navigate Predatory & Hostile Environments: Insights from Systems Security Engineering 帮助未来的核电设施在掠夺性和敌对环境中导航：来自系统安全工程的见解

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70111

Adam D. Williams

{"title":"Helping Future Nuclear Power Facilities Navigate Predatory & Hostile Environments: Insights from Systems Security Engineering","authors":"Adam D. Williams","doi":"10.1002/iis2.70111","DOIUrl":"https://doi.org/10.1002/iis2.70111","url":null,"abstract":"Discussions at COP28 and COP29 emphasize that deploying advanced and small modular reactors (A/SMRs) with high safety and security standards can address energy security and climate change challenges. INCOSE's Vision 2035 advocates for systems-theoretic approaches to integrate security throughout the development lifecycle, ensuring resilience in contested environments. The systems security engineering (SSE) domain aims to incorporate security solutions into systems engineering through requirements, trade-space navigation, and systems architecture. Recent dialogues within INCOSE's SSE working group shift the security paradigm toward engineering for functional persistence in predatory and hostile environments. This perspective shifts the emphasis on security to highlight design decisions that enhance situational awareness, preparation, defense, and recovery capabilities and that augment efforts to manifest “security-by-design” for A/SMRs. By applying systems-theoretic principles to integrate security early, frequently, and continuously, A/SMRs can improve their security performance and cost-effectiveness. This proactive approach is essential for navigating the evolving threat landscape while meeting global energy security and climate change objectives.","PeriodicalId":100663,"journal":{"name":"INCOSE International Symposium","volume":"35 1","pages":"1306-1319"},"PeriodicalIF":0.0,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Analyzing Systems Engineering Vision 2035 Through a Cultural Lens 从文化视角分析系统工程2035愿景

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70038

Ahmad Nayef Salih Alsudairi, Azmin Shakrine Mohd Rafie, Serhan Shaman Alshammari, Amini Amir Bin Abdullah, Syaril Azrad Bin Md Ali, Ezanee Bin Gires, Abdullah Algarni

引用次数: 0

Toward Quantitative Assessments of Cybersecurity Countermeasure Efficacy 网络安全对策效能定量评估研究

INCOSE International Symposium Pub Date : 2025-10-09 DOI: 10.1002/iis2.70077

Ben Breisch, Kristin Voss, Will Barnum

引用次数: 0