Model-Based Systems Engineering (MBSE) relies on formal models for system lifecycle management, supporting model coherence and efficient reuse of components. Modelling languages, particularly SysML v2, provide a standardized framework for complex system modelling, overcoming key limitations of earlier versions. Unlike SysML v1, which was a UML profile and inherited UML’s complexities, SysML v2 is designed to better handle multi-disciplinary, large-scale, and emergent system behaviours. However, the validation of SysML v2 models is still an emerging area, and while some methods are beginning to surface, comprehensive and standardized validation approaches are not yet fully established. This may pose challenges to ensuring correctness and reliability in engineering workflows. This paper presents a systematic, metamodel-based method for validating SysML v2 models, utilising the SysML v2 metamodel as a formal specification. By defining validation rules derived from this metamodel, the method facilitates automated detection of structural and semantic inconsistencies. A practical case study validates the method on SysML v2 models of aerospace, automotive, and software development domains, demonstrating the systematic identification and resolution of errors. This research advances SysML v2 model validation, contributing to broader MBSE objectives by ensuring correct SysML v2 models for complex system development in multi-disciplinary environments.

The use of different techniques and tools is a common practice to cover all stages in the development life-cycle of systems generating a significant number of work products. These artefacts are frequently encoded using diverse formats, and often require access through non-standard protocols and formats. In this context, Model-Based Systems Engineering (MBSE) emerges as a methodology to shift the paradigm of Systems Engineering practice from a document-oriented environment to a model-intensive environment. To achieve this major goal, a formalised application of modelling is employed throughout the life-cycle of systems to generate various system artefacts represented as models, such as requirements, logical models, and multi-physics models. However, the mere use of models does not guarantee one of the main challenges in the Systems Engineering discipline, namely, the reuse of system artefacts. Considering the fact that models are becoming the main type of system artefact, it is necessary to provide the capability to properly and efficiently represent and retrieve the generated models. In light of this, traditional information retrieval techniques have been widely studied to match existing software assets according to a set of capabilities or restrictions. However, there is much more at stake than the simple retrieval of models or even any piece of knowledge. An environment for model reuse must provide the proper mechanisms to (1) represent any piece of data, information, or knowledge under a common and shared data model, and (2) provide advanced retrieval mechanisms to elevate the meaning of information resources from text-based descriptions to concept-based ones. This need has led to novel methods using word embeddings and vector-based representations to semantically encode information. Such methods are applied to encode the information of physical models while preserving their underlying semantics. In this study, a text corpus from MATLAB Simulink models was preprocessed using Natural Language Processing (NLP) techniques and trained to generate word vector representations. Then, the presented method was validated using a testbed of MATLAB Simulink physical models in which verbalisations of models are transformed into vectors. The effectiveness of the proposed solution was assessed through a use case study. Evaluation of the results demonstrates a precision value of 0.925, a recall value of 0.865, and an F1 score of 0.884.

Ensuring the semantic alignment between system models and natural language requirements is a critical bottleneck for the adoption of Model-Based Systems Engineering (MBSE). While SysML v2 offers enhanced expressiveness compared to SysML v1, including: (i) first-class support for behavioral and structural integration, (ii) improved typing and reuse mechanisms, (iii) formal semantics enabling precise model interpretation, existing validation toolsprimarily target syntactic correctness leaving a gap in automated semantic assessment. The present work proposes a deterministic multi-agent framework designed to automate the semantic validation of SysML v2 models. Unlike traditional monolithic Large Language Model (LLM) approaches, the proposed framework decomposes the validation process into a specialized pipeline of five agents: model abstraction, requirement normalization, semantic judgement (grounded in the requirements quality characteristics defined in the ISO/IEC/IEEE 29148:2018 ”Systems and software engineering — Life cycle processes — Requirements engineering”), metric aggregation, and actionable feedback generation. The approach is evaluated through an empirical study using a dataset of 14 SysML v2 models and 303 requirements derived from official specifications. A comparative analysis is performed using two different LLM backends: a highefficiency proprietary model (GPT-4o-mini) and an open-weight reasoning model (Deepseek-R1 ). Results show that the multi-agent orchestrationreliably identifies semantic mismatches, incomplete realizations, and omissions across varying levels of system complexity, ranging from small componentlevel models (M1–M5) to large system-of-systems architectures involving multiple interacting subsystems. The findings show that while the openweight model exhibits a more permissive validation stance, the proprietary backend provides a conservative and highly precise assessment suitable forsafety-critical domains. This work contributes to the automation of auditable model quality assessment, providing a scalable path for iterative model refinement in modern software and systems engineering.

This paper proposes using "Thinging Machines" (TMs) to enhance ER modeling, aiming to capture dynamics and offer adjustable detail. However, the paper needs significant revisions before publication. The explanation of TMs is dense, and the link to ER...