I’m sure you have read the news that Aras is drawing a line in the sand in the world of simulation. We have been working on Simulation Data & Process Management (SPDM) internally for some time—we believe that it is key in enabling the Business of Engineering.
Simulation is necessary to develop the next generation of products. The features and capabilities of new products are rapidly changing and driving complexity. Simulation allows a firm to verify their design before they cross the point of no return—full-scale manufacturing. Simulation alone is great…but real value is gained when simulation is wrapped up in a PLM platform that can manage the simulation data AND process.
To help shed light on how value can be obtained with SPDM, I thought some examples would be useful and no list would be complete without the big three:
1) Digital Transformation
You may ask – “Aren’t I transforming my business simply by doing simulation?” Sure, you are. But, that is like saying I’m eating better by getting a Diet Coke with my supersized meal.
The transformative aspects of simulation really kick in when you integrate them into your daily processes. To do that, you need to manage both the process and the resulting data from your simulation. With properly managed simulation data you can start to relate your simulations to the existing product data in your PLM system and business process.
In this new paradigm, there are relationships between the requirements, models, parts, simulations and the simulation processes. Everything is connected, informing your entire product team. Now, you’re transforming how your firm does business! Plus, those models can be reused or used as the basis for future designs, adding additional value and knowledge.
Capturing your simulation data allows you to trace back to see what the results of particular simulation use cases were. Perhaps you are seeing failures in the field and want to see if there was a stress concentration that was not noted. With this ability to reach back, you can verify that the particular version of the simulation (because you are running many simulations) that was used, met the needs.
The other benefit with Aras’ direction is in the process. In the case I cited, you would also be able to evaluate the process used to reach the result. Perhaps the result showed that the solution was spot on and you go about your merry way. However, perhaps the process that was used was improper—whether it be the mesh, the boundary conditions, etc. By capturing the process and maintaining those records, you can now verify all aspects of the simulation.
3) Digital Twin
Once you have a manufactured product—an aircraft with a tail number, for example, you can generate your digital twin. The twin is an aggregation of the CAD data, simulation results, etc. for that particular aircraft. If that aircraft rolls off the line, and it matches the “As-Designed” (eBOM) and “As-Planned” (mBOM) definitions, then the twin is the same as the design data.
Where things get fun is when there is a deviation. Maybe on the 5th aircraft there was a problem with a machine and a number of parts were machined improperly versus the design. Because we have our simulation and processes managed in Aras SPDM, the design team can simply create a variation to the part, adjust the simulation process to match, and run a new simulation. If the results are acceptable, a deviation can be issued and the modified models and simulations saved to PLM. Now the digital twins for aircrafts 1-4 and 5 match their true “As-Shipped” conditions. The customer or MRO team will thank us for accurately capturing this information in PLM.
Those high-level cases are great, they use the buzzwords and make the executives and analysts happy. But, let’s dig a little deeper. Here are four more areas where I feel SPDM with a PLM platform changes the game.
4) Characteristic Data Capture
As I have shown, simulations are great. And managing the data and process is valuable. When you are operating inside of a PLM platform that supports SPDM, the sky’s the limit.
One very simple and effective feature is the capability to extract characteristic data from a simulation and process. In CAD, we capture things like weight, cross-sectional area, and volume and map them to properties in the PLM platform. These are all characteristic of the part geometry. In simulation, we want to capture characteristics of the parts performance. We may want to assess strength, temperature, deflection, creep, thrust, fuel efficiency, drag, etc.
Simulation is much more effective when firms capture characteristic data in PLM item properties. These properties allow you to search for existing products, systems, or parts that meet geometric and performance criteria. For example, while designing the new LHA Aircraft carrier, Huntington Ingalls may want to reuse an existing propeller. They could search their Aras instance for propellers that meet rules for diameter, thrust, and weight. If one meets the need, they are done. If not, they have a number of existing parts that are close to the requirement—AND the CAD, simulation, and processes to use as the basis for a new design.
5) Multi-Fidelity Models
Simulation is not just about expensive, high fidelity 3D models with pretty colors. That is like taking a 747 to cross the road. Effective simulation utilizes both low and high fidelity models to support the entire lifecycle of the product. By managing and supporting the ability to relate them together you can push your digital transformation further.
Low fidelity mathematical models and simulations (MATLAB, Simulink, etc.) are used early in the development cycle. Then high fidelity (ANSYS, LS-Dyna) simulations are used in the detailed phase to get very fine results.
The important factor is that the low fidelity models are not thrown to the curb. They establish the design basis as well as the boundary conditions for the high fidelity models. There are relationships between them that create additional value that SPDM captures and manages.
6) Risk Reduction
Product development is not cheap. The development phase of a product requires decisions that have to balance a complex equation with only partial information. When proposing design decisions, engineers like to give options. But, which option is objectively the best? A subjective, best guess approach is not appropriate when considering decisions that impact the viability of a product and perhaps the firm itself.
With Aras’ platform with SPDM, the engineer can model a number of options at an appropriate level of fidelity and capture the simulation data, results, characteristics, and process. For each successive analysis, the process can be adapted to fit the next design option. With the characteristics and results available, an appropriate selection method like weighted analysis or Design of Experiments can be used to objectively arrive at the best selection. Your firm can feel confident that their Business of Engineering has provided a product that meets market requirements.
7) IoT Feedback
More firms are getting their feet wet with IoT. They are capturing some kind of data from their products in the field. Much is made of the ability to do predictive maintenance with IoT data and that is a worthy pursuit. However, there is more we can do with IoT data.
Engineering 101 says to state your assumptions and solve the problem. With all product designs there are assumptions. As engineers, we simply do not know everything (But, I’ll never admit that I said that) so we have to make assumptions to be expedient.
IoT data provides us with information that we can use to replace our assumptions. Data that captures actual use of the product and can be used in our earlier simulations. With SPDM, we have the simulation models and process that can be re-executed against this new information and saved with the “As-Maintained” structure of the product. These more accurate simulations can be used to inform design optimization projects, next generation designs, or maintenance and performance packages.
But, I’ve saved the best for last…
8) Correlation…Statistical Correlation
Nearly all firms still rely on physical testing to validate their products. Even the ones that are performing simulation at a high level. Most of these firms deal with some regulatory body that requires proof that a requirement has been met. In my past life in aircraft seating, we were required to prove that our seats could withstand a crash that measured 16Gs (16 times the force of gravity). No matter how much simulation we did…we still crashed seats.
Physical testing of this type is a waste. With today’s technology, we can validate our design vs requirements with simulation alone. So why don’t firms do that? Because they have not invested in the ability to statistically correlate their simulations and tests. To do this you need SPDM. Full stop.
If a firm were capturing data for each of their simulations and using that to provide evidence of long term correlation—regulatory bodies will accept that in lieu of full testing. The key is to capture and correlate over time—it takes a commitment AND the SPDM platform to do it.
With SPDM, we are capturing the model data, key characteristics, and the process itself. We can use all of that information in concert to provide trend and performance analysis to the regulatory body in the form of a plan to reduce testing to a minimal level. I don’t think the FAA is going to let Airbus fly a plane that has not had any testing. However, they will let them reduce their testing based on proven simulation capability.
Testing is a true cost center for a firm. By reducing the number of tests in favor of simulation there is a realized benefit to the financial bottom line.
While this list of areas where SPDM shows its value is a good start, I firmly believe that simulation is the way to achieve differentiation and market leadership. I encourage all of our customers to simulate early and simulate often. All of the factors noted above will help you to deliver value to your firm. Are you ready to get started?