Engineering the future of mobility

Laurel: Nand, how does that help safety, when you can use simulation or digital twins, or just have more data to make these vehicles safer?

Y: Simulation forms the basis for offering a digital twin, or systems engineering, in my opinion. So with simulation in the early stages, you can make the right architecture selections and then move on to detailed designs. It allows you to explore the space for optimization in the delivery of that solution. When you combine that with the physical representation of the same simulations and put these two things together, that’s how you increase confidence in your simulation and physical test results, which is called CAE test correlation. That helps deliver systems engineering. So you could say simulation, digital twin, they go hand in hand to deliver or allow systems engineering to go from one end to the other.

Laurel: So Nand, how does systems engineering help scale product development and/or create this industrial efficiency? What is the return on investment?

Y: That’s interesting. Industrial efficiency is one of the biggest end results, how to monetize all these investments. I’ll use a couple of examples you asked about earlier. First, how is a safe vehicle delivered? When running a lot of simulations, one of the goals is to reduce the number of physical prototypes built so that you can rely on that simulation. By definition, this is cheaper because no parts or machinery are consumed to build those prototypes, and that’s a big deal in the auto industry. At the same time, you are doing a lot of innovation. There are some things that haven’t been done in a physical test environment, so you need to go hand in hand and do CAE mapping to build trust. After that point, you are generating another set of data through the simulation. Now in your next program or iteration of the design, it will be much more efficient.

Let me take that even further: how does artificial intelligence fit together with this massive simulation data? There are many cases where you take the data from the simulation and through machine learning train the algorithms on the result of that particular simulation. So if you’re doing aerodynamic analysis and looking at a drag coefficient, it’s computationally intensive. Sometimes those run up to five days to get results. If you’ve trained your algorithms through machine learning and artificial intelligence, you can continue to build your database, for given test conditions, on what the results would be. In the end, when you have another new design scenario, you don’t have to do those five-day simulations. You run it through those algorithms, and that gives you the results in a matter of minutes. You can see a lot of efficiency, both in terms of the time it takes to do it and in terms of computing, which brings down the cost of all those things. This is how you increase your return on investment and extend your product development. You can scale product development for multiple perspectives by doing more with less, with fewer people because with simulations and all these technologies combined, you can do the same amount of work. Or you can save with the same number of people by pushing more products. In the automotive industry, you have sometimes as many as 20 programs running simultaneously, and can be more efficient.

Laurel: Dale, when we talk about ROI and aircraft for aerospace defense, we’re talking about investing in a system and hardware that can last for years. An airplane is not replaced in a year. It has to last a long time. How does ROI affect the way people think about systems engineering?

Valley: That’s a great question. Some of the comments that Nand made captured a lot of that very well. I normally think of this in two ways. One is that when you think about a program, and in the aerospace industry, you go through the development program, you’re working with great teams. Look at the budgets that are used for some of these programs, and they could spend 10 million, 20 million, maybe even $100 million a month. As part of that funding, they are going through the certification process. If you can use simulation to avoid a month or two of delay, that’s a significant amount of money. Many times, if you only had a handful of simulation people working on this problem, the ROI can be 10, 20, 30, 40x. That’s a pretty amazing savings when you go through the process, or maybe it’s a pretty amazing cost avoidance.

The other part, which you mentioned, is being able to support these programs over a product development life cycle of 50 or 60 years. Having the simulation in place to be able to understand how the aircraft is performing once it’s in the field, and by upgrading the digital twin, the simulation, can optimize maintenance cycles, which can mean huge cost savings for operators. operators. . Again, the ROI can be a multiple of 10 or 20 on some of that. Sometimes those costs are hidden, but they are significant savings.

Leave a Comment