Watch the full interview.
Check out the entire one-on-one conversation via QuickChat video with Rehana and Florian to learn more.
Product quality depends on decisions made long before manufacturing begins. Engineers need to understand how design choices affect manufacturability and how products perform in the field. Connecting requirements, design information and downstream quality feedback can help them identify problems earlier and improve future designs.
Rehana Begg, editor-in-chief of Machine Design, spoke with Dr. Florian Harzenetter, senior director and global advisor for industrial customers at PTC, about how simulation and connected product data can help engineers make better decisions before production begins. Harzenetter also explains the role of digital threads and AI in connecting field feedback to the design process, helping engineers improve quality throughout the product life cycle.
When a quality problem shows up in the field, how often does the root cause actually trace back to the design stage?
Quality is one of the predominant topics that's hurting people because it is really a major cost source. But on the other hand, it's very difficult to fight.
It's really hard to track quality issues back to their root cause. Typically, they end up somewhere in service, somewhere in quality. They don't come back to engineering, and they don't come back much to the production departments.
Plug in to our podcast.
If you want to learn more about how engineers can identify product quality problems sooner, listen to the full interview with Rehanna and Florian on our podcast.
What does “prove the design virtually” actually look like today for a machine designer? And how much of the physical prototyping can it honestly replace?
It can replace a lot. We can do a lot in the digital world. But the first thing is: How do we review our designs?
Engineers need to find the design that makes the product work. There's a lot of simulation capabilities and engineering know-how available that helps them do that.
But it becomes more complicated when it comes to the topic of quality, because quality is not a function of the product. It's sort of a byproduct of the design and the decisions about how a product is being made. And it requires a lot of feedback.
If an engineer decides on a specific design, he's not only defining what the design is; he's implicitly defining also how that design is being manufactured, which materials are used, how those materials will behave in the future, how the manufacturing methods are used and how easy or complicated it is to really manufacture the product to the specification that the engineer sets out.
A review cycle with production would be great to have more closely, because typically in production, people can quickly tell you what the problem is with this design and how to work around it.
What does it take to establish a single, authoritative product record as the default source of product truth across the enterprise?
The single source of truth is what we're looking for, and it's good for many things, especially quality.
The digital thread starts with requirements and coming to quality. Those requirements should set out measurable expectations for the product. How long should it last? What is the intended use of the product? How many hours will it run? What kind of loads will it be exposed to?
And that needs to be broken down into the parts and pieces of the product, to the software, to the hardware, to the electronics, to really understand what it is that each of those parts needs to contribute to that.
When we have those requirements, when we have them broken down to the part, we then need to check whether really those benchmarks and thresholds are being reached and whether the product as a total then behaves as it was defined in the previous stage.
How can field failure reports feed back into design decisions without forcing engineers to replace the systems they already use?
First of all, you need the connection — that's what we call the digital thread. We need traceability from requirements to design to manufacturing to service, and to the actual service incident or service order that the customer had to replace a faulty part. A lot of companies have those data sources in some shape or form, but they're not connected well enough.
In the age of AI, we have an opportunity if we have that digital thread. We bring it together — something we would state at PTC as the intelligent product life cycle — so you can learn from all of those data sources, whether they are from our products or external products. You have them connected and get that feedback together so the designer for the next iteration, or for the correction of maybe that quality cost-heavy part, really understands what's happening there.
When software can propose the design, where does the engineer's judgment stay in the loop? And what will engineers in the near future actually spend their time on?
I would hope they spend much more time on designing the actual products.
We see there is a lot of engineers' time lost in the process. I hear from customers that about 5% to 7% of an engineer's time is spent transcribing data from one system into another.
When designers work on their actual product design, they still, in many cases, need to create drawings that cost them another 20% to 25% of their time, simply transferring the information from 3D to 2D.
It would be even better if we had full transparency and full fidelity of information throughout the process. And we would have means to filter that down to the needs of the respective role that is working with that information, instead of transferring it into different file formats and different representations.
To learn more about PTC, please click here.
Watch the Video QuickChat.
Listen to the Podcast.
Sponsored by:
