Recall a graph showing the timing of costs committed versus cost incurred.
If we overlay two additional curves,
product knowledge versus time and the ability to change a design versus time,
these curves guide us to the following conclusions.
That increased simulation capability and capacity provides more knowledge
earlier in the design process enabling changes to be made before costs are committed.
It's important to maximize impact on cost drivers
and maximize the ability to iterate early in the design process.
And developing and using models allows the product knowledge curve to be shifted left,
front loading product knowledge,
reducing the need for physical prototypes.
To truly use a model-based system to its full potential,
a framework is required.
This framework provides the guiding structure to
implement analysis throughout the lifecycle of the product.
MBSE frameworks can focus on
single components or a series of components that are linked together.
Each part of the model can be checked for errors as well to
determine if a design meets the product development requirements.
An analysis framework connects a variety of
simulation tools -- for example, Ansys, Nastran, Matlab,
and tools built in-house -- in such a way that outputs of
one analysis provide the input for subsequent analyses without human intervention.
A framework will also have some form of execution engine or control
mechanism to iterate the analyses
until a stable solution is reached for a particular design point.
Models and the framework can be
version control and updated without impacting the architecture of the analysis.
In an ideal situation,
MBSE focuses a significant amount of effort on getting the requirements right.
With some analysis frameworks,
these requirements can be automatically checked against
simulation outputs to see if expectations are met.
One example of an MBSE framework is Siemens offering HEEDs.
HEEDs provides a means of computationally exploring
possible design alternatives in search of specific performance goals.
And, in HEEDs, design space exploration is made possible
by automating the construction of the virtual prototype via process automation,
accelerating testing of the virtual prototype via distributed execution,
looking for better design alternatives via an efficient search process,
and ensuring reliable product performance via
insight and discovery tools related to visualization.
When we think about these automated processes,
keep in mind the traditional process of simulation
requires multiple steps and human intervention.
The MBSE process allows for automated design space exploration.
And, with that in mind,
another example of an MBSE framework is the LMS Imagine.Lab.
LMS Imagine.Lab facilitates the simulation of mechatronic systems using an MBSE approach.
Imagine can integrate with LMS Amesim models,
Matlab and Simulink models, or other tools.
LMS Imagine.Lab has three key components.
System synthesis which focuses on system architecture and configuration.
Amesim which is for building and running simulations.
And Sysdm which focuses on system data and configuration management.
These all come together in order to explore our design spaces for our products.
Other example frameworks include Phoenix Integrations ModelCenter.
This creates and automates simulation workflows,
allows the integration of models from standalone application,
and runs multi-disciplinary simulation processes.
The MBSE pack of model center enables integration with Doors or No
Magic to connect requirements to
model-based simulation results closing the loop in system design.
Esteco's ModeFrontier is another simulation workflow automation tool
that focuses on trade space exploration.
And Dassault's iSight is a simulation workflow automation package that enables
optimization and the exploration of design of
experiments in a multi-disciplinary analysis environment using different tools.