Product Design Challenges Drive Simulation Evolution

By Jim Cashman, CEO, ANSYS

Around the year 2000, some experts believed that the market for engineering simulation software had leveled out. However, ANSYS leaders saw almost infinite opportunities for growth. By meeting the needs of product developers, ANSYS stayed just ahead of new product design requirements and delivered essential simulation software. Sixteen years and many software advances later, the company still believes in endless possibilities to help customers develop better products through simulation.

Save PDF Subscribe
Product Design Challenges Drive Simulation Evolution

There is some dispute about who said it first, but the great Danish physicist Neils Bohr is often credited with having said “It is very difficult to predict — especially the future.” Even some of the most brilliant, respected pioneers in their fields have had difficulty making predictions. Like Thomas Watson, president of IBM, who said in 1943, “I think there is a world market for maybe five computers.” Or Ken Olsen, founder of Digital Equipment Corporation, who made this statement in 1977: “There is no reason anyone would want a computer in their home.” These were tough calls based on sound reasoning at the time. To be fair, we all make mistakes, and we shouldn’t be too hard on these stellar businessmen.

About 20 years ago, the engineering simulation industry — which uses software to make models of new products whose properties and performance can be tested on a computer, eliminating the need to build and test physical prototypes — experienced a similar period of doubt. Some industry leaders believed that the market had plateaued. They had some solid reasons for believing this: Performing simulation was difficult, generally requiring an engineer with a Ph.D. and a good grasp of the complex mathematics involved, along with enough expensive computing power — and time — to solve complex product design challenges. The limited supply of such high-level engineers and the relatively slow computers of the time led them to believe that engineering simulation had peaked.

Our Contrarian Viewpoint

At ANSYS, we did not share this outlook. In fact, we believed the market had tremendous room for growth. How could you talk about stagnation when only about 1 in 30 engineers used simulation at all? From decades of continual improvements, engineering simulation software was increasingly proving its worth by reducing product development times and costs, enabling companies to get their new products to market faster. It also increased their confidence that the product would work as specified the first time and every time throughout its expected lifecycle. Finally, engineering simulation was moving in the direction of creating a virtual prototype of the complete product, not just each component of the product in isolation. The physical, electrical and thermal interactions of various components when put together in a final product play a large role in determining how the product will function in the real world. To create a virtual prototype of a complete product, you have to be able to simulate ALL of it. As you’ll see below, this factor drove ANSYS to add electronics (and soon other physics) simulation to our offerings; it also gave rise to the concept of considering embedded software an integral component of a product, not simply some control code added as an afterthought. Engineers became convinced that embedded software could be subject to modeling, simulation, verification and certification as much as any physical component of a product.

Our attitude at ANSYS reflected these positive trends and driving forces. We believed that every engineer could benefit from simulation — not just the Ph.D.s, but any engineer involved in product development. After all, simulation is basically the automated solution of the equations they learned during their engineering studies. We just had to continually make setting up and running these automated simulations easier so that it was accessible to every engineer. So we started on a journey that continues to this day to simplify simulation. To do this, we have relied on experts in the field to let us know what is coming so we can strike out a little bit ahead of the pack. It’s like surfing. You want to stay just a little bit ahead of the wave. If you’re way ahead of it, you will miss the ride (in technical terms, new enabling technology and computing advances might make your innovation obsolete). And if you’re far behind the wave (with a “me too” product), you have missed the opportunity entirely.

Making simulation easier has involved, at least in part, incorporating more of the specialized knowledge in the brains of mathematically inclined Ph.D. engineers into the software solution itself. So we hired some of the best mechanical engineering thought leaders — back then, ANSYS was mainly a mechanical engineering simulation company analyzing the structural components of products for strength, toughness, durability, etc. — to work with our software development experts to transfer their knowledge to the software. At the same time, they thought hard about each step in the simulation process to make it more intuitive, creating menu-driven human–machine interfaces to replace arcane code-based input instructions.

Responding to Product Evolution

While implementation of these changes in our software solutions was expanding the simulation market, the nature of the products being developed started to evolve. In the past a mechanical device was made up of mechanical components, but increasingly electronics were being incorporated into the same device. The microelectronics being developed were combined with mechanical components to produce electromechanical devices that were capable of better sensing and control of the machines they were part of. This trend accelerated as electronics continued to get smaller.

But combining the physics of mechanics and electronics made product development more difficult: a complete virtual prototype had to include both types of physics to be valid. The number of new variables involved, and how electronic properties affected mechanical performance and vice versa, rapidly increased the number of prototypes a company had to investigate. The demand for simulation solutions increased, and the idea that simulation had reached a plateau began to fade.

To keep up with these new requirements, ANSYS had to evolve to handle electronics simulations involving electric and magnetic fields, integrated circuits, radio frequency signals, etc. The era of multiple physics simulations was born, and the challenges we faced multiplied. It was not long before other physics were included in the mix: fluidics, heat transfer, and embedded software modeled as another component in a virtual prototype.

Multiple physics simulation is a sequential process: You do a structural analysis and then you do a thermal analysis, maybe followed by a flow analysis. But in the real world, those aren’t individual things. They all happen at once. For instance, if you’re operating something in a hotter environment, the combination of heat and the mechanical vibration stress can cause failures where neither one alone would have had the same effect.

That led to multiphysics — as opposed to multiple physics — simulation. Multiphysics tries to look at all of those combined effects at once. So the structural, thermal and flow analyses take place simultaneously, passing changing variables back and forth automatically between solvers. This process eventually converges on an optimal solution that takes all of these interconnected physical properties into account, producing a holistic approach to managing the multitude of conflicting trade-offs.

If this sounds complex, it is. But once again, we had to focus on making simulation easier so it could be done by more engineers even as the complexity increased. We had to develop a platform that enabled various physics solvers to work together seamlessly, so engineers could perform simulations without excessive training. The resulting ANSYS Workbench platform does precisely that, and it continues to grow as the needs of our customers expand. Workbench helps our customers to build a complete virtual prototype of an entire system, not just individual components.

IoT is a logical evolution of product design

IoT is a logical evolution of product design

"Simulation is really the only way to understand how products are going to behave in an increasingly complex, regulated and demanding business world."

Keeping Simulation Flexible

We also learned that many of our customers had unique software that they had created in-house to give them an advantage over competitors, or because there was no commercial software available to perform a critical task. Whatever the reason, they had to be able to integrate their custom software with ANSYS solutions to optimize their simulations. So we made Workbench flexible, enabling customers to plug in their own software and seamlessly run simulations that combined the power of our physics solvers with their unique expertise.

As simulations became more complex and companies required more innovation to compete, high-performance computing (HPC) and the cloud were necessary to speed up product design and reduce costs while ensuring reliability. We responded by making it possible to use HPC and cloud solutions to produce higher-fidelity simulations of products, which enabled engineers to look at complex interactions and see not only where something failed but why. Simulation began to be used earlier in the design process and throughout it, instead of just at the end as a final verification point. This saved cost, time, computing and personnel resources — it’s cheaper to make changes earlier in the design process than later.

We will make simulation easier

The Way Forward

Recently, some of our customers have started to use our simulation solutions for a purpose we and others had long ago anticipated: to monitor the operation of systems and predict when maintenance will be needed, because unscheduled downtime is considerably more expensive than scheduled downtime. They do this by creating a “digital twin” of the system — a virtual copy of their product that can be used to determine what the expected behavior of the actual product should be. If the real-world system begins to deviate from the digital twin’s behavior, that’s an indication that something is wrong. The digital twin can help you diagnose the problem and schedule the downtime to fix it.

Though it may seem like something out of science fiction, perhaps one day we’ll have enough computing power and sufficiently advanced software to produce a complete virtual prototype of the entire Internet of Things — effectively, a digital twin of the IoT — to monitor its workings and accommodate every new change to the system.

ANSYS is the simulation leader

Regardless of whether we ever get that far, it’s clear that simulation is really the only way to understand how products are going to behave in an increasingly complex, regulated and demanding business world. We’ll continue trying to surf just ahead of the wave, so we can see what’s coming technologically, and constantly improve our software to deal with the changes that our customers will be facing. We’ll make simulation easier even as products get more complex, so every engineer in your organization can use it to save you time and money.

Today, about one in six engineers is using simulation; we’ve long been predicting that by 2030 every engineer will do so. We’ll stick by that prediction, with full knowledge of the perils involved. We look forward to working with you so that we can reach this goal together

"HPC and Cloud solutions produce higher fidelity simulations of products, enabling engineers to look at complex interactions and see not only where something failed but why."

Jim Cashman Bio

Jim Cashman

Jim Cashman is the chief executive officer of ANSYS, Inc. Prior to joining ANSYS, Cashman held positions at PAR Technology Corporation, Metaphase Technology, Inc. and Structural Dynamics Research Corporation. He is chairman of the Pittsburgh Technology Council and a past board member of the Carnegie Museum of Natural History. Cashman’s experience includes senior responsibilities in technology, product and market strategy management, as well as sales, operational and international functions prior to his general management role with ANSYS for the past 15 years. His long-standing vision and well-diversified background are key components of the company’s board structure and effectiveness.

Start a conversation with Ansys

Contact Us