Simulation-Driven Product Life Cycles Require CAE Portfolios with a Breadth of Fidelity

Altair is focused on providing simulation software that guides product development and life cycle. (Image courtesy of Altair.)

The buzz surrounding computer-aided engineering (CAE) is changing. Though it’s still true today that “simulation is more than just validation—it’s design,” this creed of simulation-driven design neglects to acknowledge the way in which simulation can inform manufacturing, operations and more.

Today, simulation is more than just design and validation; it’s the whole System V. Simulation vendors have noticed this trend. Many, like Altair, have positioned their portfolios to support the needs of the System V.

“A lot of the design process and product life cycle is driven by simulation technology,” confirmed Uwe Schramm, CTO at Altair. “Our software not only helps design products; it also helps to understand the manufacturing process and how that influences the outcome of the products. It helps operations, too, as companies can make decisions [based] on simulation instead of inspection.”

How Simulation Portfolios Position Themselves for the System V

Using simulation to drive innovation throughout the development cycle requires portfolio breadth. (Image courtesy of Altair.)

The key feature that engineers should look for when simulating the complete development, operations and manufacturing life cycle of a product is the breadth of a vendor’s portfolio.

Your standard finite element analysis (FEA) is great for validation, and potentially some design as well. However, FEA will prove too slow once you reach the point of operations or early development.

“Basically, start off with low-fidelity system models, in our case, solidThinking Activate,” suggested Schramm. “These virtual prototypes explore the general physics of the system to assess how the general principles are working together. Then as you design, you add more and more fidelity as you need it. By the time you are at the verification stage, you are using high-fidelity models.”

Creating a Virtual Factory Leads to As-Manufactured Optimized Designs

Engineers using virtual prototypes, virtual factories and the digital twin can simulate the whole system development process as well as a product’s operations. (Image courtesy of Altair.)

Once product designs are optimized, engineers can start modeling virtual factories to digitally manufacture them.

These simulations can help engineers iterate and determine the optimal manufacturing setup and processes for a product.

These manufacturing simulations can even help engineers discover imperfections in final products before they cause a problem in the real world. To assess the seriousness of these imperfections, they can be fed back into the validation and design models. If the imperfections prove to be an issue, further assessments and optimizations can be made to the design.

“You might find the actual spot where the product will break based on manufacturing imperfections,” noted Schramm. In other words, engineers can reduce the need for physical prototypes as they better capture the validation of as-manufactured products digitally using simulation. In fact, some industries even allow these simulations to supplement the testing required for certain certifications.

IoT and Digital Twins Inform Design, Operations and Manufacturing

Linking real-world data to the digital twin via the Internet of Things is key to optimizing product performance and future designs. (Image courtesy of Altair.)

So, you have a part, or an operation, but now you need to ensure the day-to-day use and manufacturing of the product.

Unfortunately, the high-fidelity simulations you have been working on during the validation and late design stages are too slow to adequately inform the quick pace of product operations and daily manufacturing.

This is where Schramm suggests the use of data-driven simulation, machine learning, response surface models (RSM) and digital twins.

The difference between data-driven simulation and model-based simulation is that data-driven simulation is based on machine learning and data collection, while traditional model-based simulation is based on equations and finite elements.

For data-driven simulation, Altair offers HyperStudy. The software takes in the data and runs it through a series of machine learning algorithms, such as least squares, neural networks and smart fit functions.

“The software runs through a few machine learning algorithms, and then picks the best model,” said Schramm. “This simulation can then be plugged into a system simulation as a reduced-order model. This tool can then simulate production for predictive maintenance to supplement design decisions or to inform operations.”

You might also find yourself dusting off that old systems model that was used in the early stages of a product’s development. These low-fidelity systems models should be updated to reflect what has been learned when optimizing the product design and manufacturing process.

“You might need to build a different model with loading conditions based on real operations,” said Schramm. “Then you pick your load cases you need to verify. From here, you build statistical data models.”

Engineers will also find it useful to connect these digital twins and low-fidelity models to their physical products using the Internet of Things (IoT). A recent addition for Altair is its IoT and analytics platform, Carriots. This platform can help ensure that models are being fed with up-to-date information.

Once everything is up to date, you should have a fast and responsive simulation that can inform operations and day-to-day manufacturing in near real-time.

Of course, the development cycle never really ends. As soon as one product is launched, it’s time to start working on the next one. Design and validation teams will find that they also have use for the digital twin. This process is called closed-loop engineering.

“Now, with the IoT and closed-loop engineering, you can put real data into reliability-based design, and with compute models, you can do extra reliability-based design to look at a family of simulations to increase the reliability of a product,” explained Schramm.

Regardless of the Development Stage, Optimization Is Key

Simulation-based optimization is often thought of in early development like topology optimization. However, simulation can optimize all stages of a product’s development. (Image courtesy of Altair.)

Optimization is key to the quick product development and evolution of your products during each point of the development process.

“In all phases [design, validation, manufacturing and operations], you want to support your decision-making using the best algorithms that help you narrow down the decision-based requirements for that phase,” said Schramm.

Manual iteration at any stage can act like a roadblock and stagnate your innovation. Many can attest to having had this experience during design phases, but one place where optimization and day-to-day manufacturing might be a godsend is during operations, when at any moment things might not go as planned. At these times, answers are needed immediately, and optimization tools can deliver them quickly.

No one wants downtime in manufacturing or product optimization, but it does occur. When it does, you need optimization and simulation on your side to arrive at a solution as soon as possible. As Schramm said, “We link optimization into all of our simulation algorithms. You don’t just want predictive, you want prescriptive to see how you can influence your design if you are not meeting your requirements.”

For more on Altair’s technology, check out its portfolio here.

Altair has sponsored this story. All opinions are mine, except where quoted or stated otherwise. —Shawn Wasserman