Newsroom
 / Press / 
Simulation
  • 796-0-0
  • 797-1-796
  • 802-3-796
20.11.2019

Dr. Martin Herrmann on simulation as key element in the product development process

Simulation-driven design for mid-sized companies

Simulative methods can make processes faster, more efficient, more flexible and safer. That’s why they are increasingly in demand in manufacturing. For product development, Simulation Driven Design has become a well-established approach. Aircraft manufacturing and the automotive industry have led the way here, but Dr. Martin Herrmann expects a "democratization" of simulation in medium-sized companies as well. The CENIT editorial team asked the expert how companies can make optimal use of Simulation Driven Design, what changes the introduction will entail and how it can best be implemented.

Mr. Herrmann, how would you sum up the benefits of simulation driven design?

Simulation driven design (SDD) lets engineers determine whether a component will be able to fulfill its function – and it gives information while they are still in the early stages of development. You can evaluate different variants, weigh the pros and cons and then select the best solution – “best” in terms of cost effectiveness, or quality, or tolerances, depending on the requirements. The decisive key aspect is timing: The earlier you can conduct and optimize your simulation, the less you have to spend on making the changes that result from it. Once you’ve set up the installation facility, it’s expensive or outright impossible to generate variants.

 

The earlier you can conduct and optimize your simulation, the less you have to spend on making the changes that result from it. Once you’ve set up the installation facility, it’s expensive or outright impossible to generate variants.

What’s the best way to make use of simulation driven design?

It’s best to use SDD at the earliest stage possible, ideally as early as the concept phase in which you determine the basic geometry. The next step is the actual design phase in which you optimize the individual parameters. Up to this point, designers have usually focused on the individual component exclusively. Now they have to assess and verify whether this component will work together with other components. Next, they use additive manufacturing processes to conduct a first practical geometry test – without the high costs of conventional prototyping. Now the designers can make changes, e.g. modifying wall strengths to avoid problems during production. And then they move on to the actual production process, which in many instances can be simulated as well.

Product Development Process with or without Simulation Driven Design

How and where does SDD help the designer?

The designer has to observe the constraints imposed by material and production requirements – and of course the component must be able to fulfill its intended function. Here a major concern is rigidity. Without software-based assistance, the designers have to rely exclusively on their knowledge and experience. SDD, on the other hand, can help designers define the right geometry for the component. It gives them plenty of leeway, but they need support from SDD to make use of this freedom. Ideally, all they need to do is define the installation space, the points of connection to other structures, and the relevant loads. On this basis, the software then calculates the optimal geometry.

Alternatively, the designer can rely on model-based system engineering to define and optimize key parameters even without geometry data. This form of simulation comes in at a higher level of abstraction. Based on a draft system, it analyzes the basic functions of a machine which depend on various parameters. These might be the key characteristics of a drive unit, such as performance at a specific engine speed. Model-based system engineering helps the designer define and optimize these key parameters.

System Engineering Process SDD inclusively

A lot of different systems can come into play here. Doesn’t that make life difficult for the designers?

Our goal is to minimize the number of interfaces: Ideally, the designers will work in the software environment they know best. That way, they also have full access to the shared data pool and are able to use simulations intuitively. This applies to model generation, but also to evaluation and calculation. If the designers hit a snag, they can seamlessly forward the concept to a specialist who uses the same information to develop a more complex model. Both sides haven’t to worry any longer about how to transfer the data –data is simply there. However, in the real world we still see different applications and therefore different interfaces. But at least this has the advantage that users can select the most efficient solutions for their specific purposes.

Our goal is to minimize the number of interfaces: Ideally, the designers will work in the software environment they know best. That way, they are able to use simulations intuitively.

So we still need those simulation experts?

Despite all the advances in integration, designers will only be able to cover certain simulation domains. For complex repetitive tasks, for example, the experts will come in and develop automated processes. These operate like a black box: The designer enters the relevant key figures and immediately obtains the required information. Then the specialists are only needed for matters of detail. Because such experts are few and far between, it’s better to assign them to other, more value-adding tasks.

This process will change the way designers work. How do they feel about this transformation?

That differs from company to company, but mostly the feedback is positive. This is because SDD lets designers investigate new concepts and use virtual prototypes to test them – without breaking the bank. It gives them a reliable decision-making basis that doesn’t rely only on experience. In the past, designers often had to turn to technical handbooks, but those only contained standard use cases. If they needed additional information, they could have internal specialists or external providers run simulations for them, but that usually meant delays and additional costs. Now they can run specific evaluations for specific use cases on their own. That gives them freedom and certainty that the components they design will do what they’re supposed to. Be that as it may: If you really can’t find a way to make simulation work for you, you’re not going to use it.

Does simulation improve component quality?

The short answer is yes. During the concept phase, SDD describes the geometry of components. As soon as you generate a geometry, you also have to define a set of tolerances. A typical case is plastic components. If you use such components, you need to know about the so-called assembly elasticity. If the component keeps breaking in a certain place, you have to reinforce it. But then the question is whether you can still assemble it at all. Simulation lets you reconcile these two parameters at a fairly early stage, giving you a cost-effective result.

How common is pre-production simulation these days?

It’s not yet common across the board, but in some industry sectors like aerospace or automotive we already see plenty of simulation. Today, computer software helps us determine how safe a car will be on the road. Once the design of a car has been finalized, it would be too costly to make changes. And real-world crash tests certainly aren’t an option: You’d have to run each car model against the wall ten times before optimizing it. That would be hugely expensive. Instead, car makers conduct plenty of early simulations to obtain detailed information on deformation behavior. Then they only have to run one or two crash tests for final confirmation. But I’d like to see simulation gain ground among mid-size manufacturers – a democratization of simulation, if you will. This is also a matter of competitiveness: In today’s world, auto suppliers who want to sell a complex component already have to be able to document feasibility during the bidding phase, and they need simulations to do that. If, for example, the designers specify a complex geometry, the tool designers have to be able to show that they can implement it without folds or tears in the material.

I’d like to see simulation gain ground among mid-size manufacturers – a democratization of simulation, if you will. This is also a matter of competitiveness: In today’s world, suppliers often have to document feasibility during the bidding phase, and they need simulations to do that.

What do mid-size manufacturers have to keep in mind when introducing simulation driven design?

The decisive aspect is maintaining a culture of openness. All stakeholders involved have to stand by this strategic decision. Also, the introduction of SDD means fulfilling a number of preconditions. First, you have to evaluate all processes whether they can be automated. In addition, you have to identify the core aspects of your production. If a certain product feature is key to your success, then you have to make sure that it will not be questioned by analyses and optimization potentials – neither during development nor during production. You also have to take stock of the entire workflow: You can’t just focus on the specific components, you have to look at the whole manufacturing process, no matter whether you’re dealing with plastic or metal casting.

Will simulation replace physical testing at some point in the future?

I don’t expect us to reach that point for a long time, if ever. On the contrary, the more precise our simulations become, the more precisely we have to measure and evaluate the results. Our main goal is to provide the designers with a tool that helps them during the drafting process. Software cannot replace testing, it can’t replace the designer, and it certainly can’t replace the creative development specialist.

Many thanks for this interview, Mr. Herrmann

 

 

Martin Herrmann

Martin Herrmann For more than 30 years, Martin Herrmann has been involved in FEM simulation. Before assuming his position as simulation expert at CENIT, he acted as Managing Director of SynOpt (since 2012), a company in which CENIT has held a majority interest since July 2017. As a sales partner of Dassault Systèmes, SynOpt specializes in simulation products from the SIMULIA portfolio.

After completing studies in mechanical engineering and a doctorate in manufacturing technology, Mr. Herrmann began his career in 1989 as development engineer for crash simulation at Daimler AG. From 1992 to 2011, he was Managing Director of a company providing measurement technology for the auto industry and FEM simulation services. From 1994 onwards, he also acted as sales partner of DEFORM software (simulation of processes in metal deformation, chipping and thermal treatment).

CENIT Simulation experts

Via CENIT’s participation in SynOpt, our clients benefit from the company’s Germany-wide reputation as a simulation expert in the fields of structural analysis, metal deformation and chipping. Technologically, SynOpt’s expertise is based on tools including the SIMULA solutions by Dassault Systèmes – an environment with which CENIT is intimately familiar as well.

CENIT provides consulting and support services across all process stages of virtual product development. Our portfolio covers the entire bandwidth from potentials analysis to methodological development and software integration to custom implementation and automation of simulation procedures.