CAE data management has become a critical topic for many CAE departments — after all, tying together which CAD models were used as the basis for an analysis, the meshing technique and resulting mesh, the loads and constraints and the resulting data and visualizations can be a nightmare. Especially if, as we all advocate, a design team simulates early and often to get to the best possible design alternatives.
In the past, CAE was seen as a validation tool: it took too long to create a model and run an analysis for it to be an integral part of the design iterations. Typically, an analysis was run at “freeze points” in the design cycle — all designers did their work to a deadline, this was “frozen” and passed to the CAE team. In a process that could take weeks or months, analysts then provided input on design elements that did not meet requirements. Because it took so long and was so specialized, CAE was outside the normal process.
Today, many analyses are possible as part of the routine work of design. It now takes minutes to hours to do what used to take days and weeks, due to advances in compute speed, distributed computing, optimization in meshing and solvers, and improved user interfaces. What used to be one giant analysis at the end of a project is now many analyses, spread throughout.
So what to do with the gigabytes of data created? After all, storage is cheap, so it is possible to save every last bit of data. Some design teams, unsure of what will be valuable or required in the future, do just that. But is that a good idea? How do you find what you’re looking for — if you even know what key phrase, date or design parameter to use in the search? How do you abstract an entire analysis into the key inputs and outputs to reduce storage volume and speed searches?
Enter CAE data management tools. Also known as Simulation Data Management (SDM), Engineering Knowledge Management (EKM), Simulation Process Management (SPM), Simulation Lifecycle Management (SLM), Technical Data Management (TDM) …
The current class of tools available breaks down into three subclasses:
1. Pure CAE management: tracking meshed models, loads, constraints and results. Examples include Altair Data Manager, ANSYS EKM, and MSC SimManager, among others.
2. PLM infrastructures that manage CAE data along with CAD, tying CAE into the greater design process and managing the resulting data using the PLM system’s change management and other stage gates. Products in this category include Dassault Systemes’ SLM and Siemens PLM’s Teamcenter.
3. PLM infrastructures that manage CAE data as a separate type — not completely integrated into the PLM system, which means that the install is easier and perhaps cheaper, but including change notification to a CAE-specific management system. PTC’s ProjectLink spaces would be a typical example.
These further break down into
a. Toolkits — build what your organization needs and tie it into other enterprise management systems (PLM, ERP, etc.). The most expensive and time-consuming option, but guaranteed to create a solution tailored to the needs of the specific enterprise.
b. Configurable — certain roles, tasks and processes are programmed into the solution to simplify installation and create a quicker payback.
c. Off-the-shelf — intended for individual analysts or small teams, ready to go as is — you do your work but conform to its limitation. Simplest to get up and running quickly.
Each class of tool and level of complexity has advantages and drawbacks, based on how the enterprise sees CAE and its importance as an asset. More on that in the next post in the series.
