From Model-based to Model and Simulation-based Systems Architectures
1 like234 views
The document discusses the integration of model-based systems engineering (MBSE) and simulation techniques to improve the quality and effectiveness of system architecture designs. It highlights the Arcadia method, the limitations of traditional MBSE practices, and emphasizes the importance of simulation in bridging gaps in engineering processes. Key best practices for successful implementation include maintaining a digital thread, separating concerns, and fostering a cultural shift within organizations.
From Model-based to Model and Simulation-based Systems Architectures
- 1. www.thalesgroup.com OPEN This document may not be reproduced, modified, adapted, published, translated, in any way, in whole or in part or disclosed to a third party without the prior written consent of Thales - © Thales 2020 All rights reserved. REF xxxxxxxxxxxx rev xxx - date Name of the company / template : 87211168-DOC-GRP-EN-005 From Model-based to Model-and-Simulation-based Systems Architectures Achieving quality engineering through descriptive and analytical models Pierre Nowodzienski pierre.nowodzienski@thalesgroup.com Thales Coporate Engineering Juan Navas juan.navas@thalesgroup.com Thales Coporate Engineering
- 2. 2 OPEN Do our MBSE practices and tools suffice to reduce the risks of our engineering programs? Sure it helps, but there are still scenarios that can slip through the cracks! YES! Yes… but cultural change is just beginning
- 3. 3 OPEN What we call simulation here ▌Simulation – a cognition process that predicts effects, including any form of deliberation based on a model. Including: Workshops in which experts deliberate about the expected behavior of the system Automatic validation rules Computer-assisted simulation of executable models Develop and describe system architecture Simulate the system architecture Predictions Objectives d Architecture (By the way, this pattern can be applied to many engineering activities other than Architecture)
- 4. 4 OPEN How, and under which conditions, simulation techniques can be articulated with MBSE to ensure proper quality architecture designs?
- 5. 5 OPEN Agenda ▌Our MBSE approach ▌Limitations ▌How simulation fills the gaps ▌Case study
- 6. 6 OPEN Our MBSE approach: ARCADIA method ARCADIA drives you to adopt certain perspectives to ensure the consistency and completeness of your architecture design Perspective Objective Operational Analysis What the stakeholders need to accomplish System Needs Analysis What the system has to accomplish for the stakeholders Conceptual Architecture How the system will work to fulfill expectations Finalized Architecture How the system will be developed and built NEED & CONTEXT PERSPECTIVES SOLUTION PERSPECTIVES TOP DOWN BOTTOM UP
- 7. 7 OPEN ARCADIA Aspects Purpose the system’s why Behavior the system does Structure the system is performs fulfills fulfills justifies The reason to exist of the system: the services it provides in different contexts of usage, so to fulfill stakeholders expectations The form – the entities that are considered during the different contexts of usage of the system, as well as the constituents of the system The function – the expected behavior of the entities in the context of usage, of the system, and of its constituents
- 8. 8 OPEN ARCADIA matrix – our “safety belt” when designing complex solutions Purpose Function Modes & States Structure Interfaces Operational Analysis What the stakeholders need to accomplish What is Marie expecting? What is she willing to pay for? How and when interactions between Marie and Inspector occur? What is done when Marie discover too late that a zone is damaged? Who does Marie interact with? What information is exchanged between Marie and the Inspector? System Needs Analysis What the system has to accomplish for the stakeholders What services shall my solution provide? What actions are expected from the system? What are expected from an external entity? What are the operational modes of my solution? (manual, semi- automated, …) Who will my solution interact with? What are the external interfaces of my solution? Conceptual Architecture How the system will work to fulfill expectations What is the contribution of the constituents to the services my solution shall provide? What actions are expected from a constituent? What are the operational modes of a constituent? Are they consistent with system modes? What is the high- level, conceptual decomposition of my system? What are the interfaces between the system constituents? Finalized Architecture How the system will be developed and built What is the contribution of the HW and SW components to the services my solution shall provide? What actions are expected to be implemented by the SW team in the next increment? How to ensure that the SW and HW constituents are available in a given mode? What are the SW and HW components of the drone? (inertial measurement unit,, HD camera, …) What are the connections between the components? What data is conveyed? NEED PERSPECTIVES SOLUTION PERSPECTIVES ASPECTS Examples of questions and answers that emerge when applying Arcadia:
- 9. 9 OPEN (Some) MBSE limitations with regards to engineering goals SHARE SECURE AUTOMATE Full consistency of design requires cross-checks between aspects, but they may be too specific or cannot be automated easily In Systems-of-Systems and AI- powered systems contexts, static descriptions do not suffice to understand emerging behaviors Completeness of IVV strategies often requires automated checks, and hence executable models Tradeoffs and optimizations of design often require frequent trial and error cycles, which require executable models Multi-concerns analysis often require data out of the architecture model and powerful visualization techniques Completeness of IVV strategies based on architecture elements often requires automated checks and hence executable models Agility drives to frequently show customers how the system provides the expected value, which is not always feasible with static models
- 10. 10 OPEN Filling the gaps with Simulation SHARE SECURE AUTOMATE Unambiguous and quantitative simulation results with extensive visualization means Enable to access and understand emerging behaviors and performances Interactive simulation as a powerful communication mean Improve concurrent and collaborative engineering Early and continuously Integrate, Verify and Validate the solution under design Executable models enable to secure the test cases before real system is available. Automate early test execution Automate model coverage analysis Automate document generation for behavioral aspect Automate code generation for functions behavior Enables wider solution space exploration and alternatives comparison Automate the exploration, with parametrized model in an optimization loop to converge towards the best solution. OPTIMIZATION
- 11. 11 OPEN Ensuring a Digital thread is essential System architecture definition Shared data • Interfaces • Parameters • Functions definition • Modes & States • … Data and information Executable model for simulation Engineering concern
- 12. 12 OPEN Data and information Data and information Data and information Data and information Ensuring a Digital thread is essential Engineering conce Engineering concern<<< Engineering concern Engineering concern System architecture definition Shared data • Interfaces • Parameters • Functions definition • Modes & States • … Engineering concern Data and information Executable model for simulation Executable model for simulation Executable model for simulation Executable model for simulation Executable model for simulation
- 14. 14 OPEN Case Study example System Analysis: Functional Chain and Modes machines Navigation modes management Transition triggers
- 15. 15 OPEN Case Study example System Analysis: Functional Chain and Modes machines Navigation modes management
- 16. 16 OPEN Best Practices & Key Takeaways ▌Separation of concerns Descriptive and analytical model serve different objectives and may have different life-cycles; keep them separated but consistent ▌One analytical model per concern type Agility imposes fast question-analysis-answer loops, which are difficult to reach with one big simulation model (unless a perfect digital twin already exists!) ▌Frame simulation activities in your MBSE methodology Simulation models shall contribute to the objectives defined by the global modelling & simulation strategy, and shall be consistent with architecture artefacts ▌Ensuring a Digital Thread is essential Digital continuity and built-in consistency is a prerequisite for success ▌Drive the cultural shift Adapt organizations, processes and practices; do not be cheap on awareness and training efforts ▌Benefits on project effectiveness and engineering quality are field-proven The feedback from the frontline shows a negative correlation between the amount of system requirements covered by analytical models and the effort required on tests