![]() |
市場調查報告書
商品編碼
1927574
基於模型的系統工程 (MBSE) 解決方案與軟體/系統建模工具:針對人工智慧時代的抽象與架構MBSE Solutions & Software/System Modeling Tools: Abstraction & Architecture for the AI Era |
||||||
當今嵌入式、邊緣和人工智慧系統的高度複雜性要求我們重新專注於工程最佳實踐。企業必須找到推動創新並成功管理變革的方法。基於模型的系統工程 (MBSE) 是一種核心方法論和工具。 MBSE 是一套成熟的實踐和技術,並且不斷發展以滿足下一代設計需求。
本報告分析了基於標準語言的建模 (SLBM) 工具(例如 SysML/SysML v2、Modelica 等)和基於專有語言的建模 (PLBM) 工具(例如 SCADE、Simulink)的市場趨勢和新興趨勢。本報告還深入探討了影響 MBSE 解決方案和軟體/系統建模工具市場的新興趨勢和技術、標準和法規、工程趨勢以及競爭策略。
人工智慧正在重新定義工程組織的需求和機會。軟體和系統建模工具的使用者已成為人工智慧的早期採用者,其應用場景多種多樣,從將人工智慧工作負載整合到終端設備和系統中,到在自身的工作流程中利用人工智慧。雖然許多供應商正在透過嵌入式人工智慧增強應用生命週期管理 (ALM) 工具,但在人工智慧系統開發領域,建模和基於模型的系統工程 (MBSE) 尤其適用於以下兩個明確的應用情境。首先,SysML 工具非常適合用來幫助工程組織設計先進的系統架構,並為安全關鍵型專案的文件和可追溯性奠定基礎。基於專有語言的工具,例如 MATLAB/Simulink 和 SCADE,支援對需要複雜演算法和對複雜環境及運行因素做出即時回應的系統進行設計、開發和模擬。我們相信,系統複雜性的不斷增加、安全關鍵功能需求的日益嚴格以及企業對效率的需求,將在未來幾年內推動對高級建模工具和基於模型的系統工程 (MBSE) 原則的需求。
The complexity of today's embedded, edge, and AI systems demands new attention to engineering best practices. Organizations must identify the approaches required to drive innovation and manage change. Chief among those methods and tools is MBSE, a proven set of practices and technologies evolving to meet the needs of next-generation design requirements.
This report analyzes the market and emerging trends for standard language-based modeling (SLBM) tools (e.g., SysML/SysML v2, Modelica, etc.), as well as proprietary language-based modeling (PLBM) tools (e.g., SCADE, Simulink). It includes detailed discussion of emerging trends and technologies, standards and regulations, engineering behaviors, and competitive strategies that are impacting the market for MBSE solutions and software/system modeling tools.
This research program is written for those making critical business decisions regarding product, market, channel, and competitive strategy and tactics. This report is intended for senior decision-makers who are developing embedded technology, including:
VDC launches numerous surveys of the IoT and embedded engineering ecosystem every year using an online survey platform. To support this research, VDC leverages its in-house panel of more than 30,000 individuals from various roles and industries across the world. Our global Voice of the Engineer survey recently captured insights from a total of 600 qualified respondents. This survey was used to inform our insight into key trends, preferences, and predictions within the engineering community.
The overall market for MBSE and software/system modeling tools reached $B in 2024 and will reach $B in 2029, a CAGR of % over the forecast period, driven by strong growth within the embedded solution market. We believe this growth could accelerate even further in the coming years, as a function of both organic market need as well as further evangelism by the growing roster of PLM, EDA, and ALM companies all working to integrate more MBSE and SysML v2 solutions across their portfolios.
AI is redefining the needs of and opportunities for engineering organizations. Already, software and system modeling tool users are early adopters of AI across a range of use cases from end devices/systems integrating AI workloads to using AI within their own workflows. While many vendors are enhancing their ALM tools with AI-infused intelligence, there are two distinct use cases for AI system development for which modeling and MBSE are well suited. For one, SysML tools are ideal to help engineering organizations architect advanced systems and establish an underpinning for documentation and traceability for safety-critical projects. Proprietary language-based tools, such as MATLAB/Simulink and SCADE, can help organizations design, develop, and simulate systems with advanced algorithms and needs for real-time response to complex environmental, operational factors. We believe that the combination of advancing system complexity, safety-critical functionality requirements, and corporate mandates for efficiency will drive increasing need for sophisticated modeling tools and MBSE principles for years to come.
Code generation has been a key area of extension and value add for modeling tool vendors for over a decade. In practice, however, legacy solutions fell short due to shortcomings of architectural abstractions and the realities of fragmented hardware ecosystems. Despite generative AI coding capabilities only recently becoming widely commercially available, users of modeling tools have eagerly adopted these solutions at a disproportionately high rate, with % using the technology - a rate twice that of the industry overall.
Developers across both enterprise and embedded domains report significant reservations regarding the trustworthiness of AI-generated code. Across organization types, engineers identified code quality, security, compliance, and license infringement as leading concerns. Embedded engineers cited code quality as the absolute highest concern due to the importance of software performance in embedded system function. Software must run exactly as intended, regardless of deployment environment. Tool providers should restrict model training databases to ensure that AI generates reliable code based on tested documentation and examples, which will also help end users reduce licensing risks. In tandem, solution providers should offer model training and refinement as a service to further ensure a level of specialized code quality that generic LLM-based solutions cannot provide.
To address compliance concerns, modeling tool vendors should partner with requirements management, test, and software composition analysis (SCA) providers. Engineering organizations must effectively manage and trace
requirements to meet standards such as DO-178C and ISO 26262. IBM DOORS, Jama Connect, and Polarion from Siemens all help engineers track compliance from design to code to test. Similarly, SCA tools from vendors such as Black Duck, CodeSecure, Mend, Revenera, Sonatype, and Snyk track violations from known repositories to ensure that open source and AI-generated code do not violate existing licenses. In the same way that application lifecycle management, software testing, and SCA have converged in recent years to form single-platform solutions, AI code generation solutions and extensions fit directly within the software tooling landscape. A fully combined solution featuring modeling, requirements management, code generation, and software verification and validation would give customers a single dashboard or source of truth for code generation analytics, quality, induced risks, and impact on development time.