February 1, 2017
As product complexity expands, it seems engineers are tasked with completing more sections of the product design cycle simultaneously to meet time-to-market goals. However, challenges can still arise as teams push products through a disconnected, sequential workflow.
To pinpoint some of these difficulties, Digital Engineering chatted with some product design and testing service providers to see how engineering teams could address some of the hurdles that can cause a gap between finishing a product design and ensuring it can effectively operate once it leaves the manufacturing line.
Q: What are some of the challenges when bridging the design/test gap?
A: Gregor Mittersinker, co-founder and principal, Loft LLC:
One of the biggest challenges is that you need to have a product at a finished level that you can simulate how it would behave. With the emergence of rapid tooling and rapid prototyping, it becomes much more feasible. A lot of times, we make alphas that fully function and we can field test on.
A lot of clients are scared in engaging on research and testing and repeatability of their ideas [due to] timelines. They have a very aggressive objective of getting to market quickly and therefore they are hesitant in engaging in testing initiatives. However, the key to us helping our clients is the philosophy of “test early, test many times, but don’t have the test become this overarching objective.” This way, testing has become a subsystem of agile development.
A: Tony Norton, executive vice president, Altair ProductDesign:
When we take a design into testing—whether it be physical or virtual—a highly constrained design can often lead to little room for change during testing. And our philosophy to approach this is to do early concept design using physics and simulation as early as possible, and then drive the design using simulation throughout. So in effect, the whole design process becomes a number of tests.
A lot of times when we get involved with clients, they come from a design and test process, and they’re looking to get into a more simulation-driven process. That really smooths out that transition [from design to test], so now instead of getting surprises after or late in the design process, you’re now doing the learning early.
A: Peter Lavelle, application engineer, HBM Prenscia:
Communication, planning and cross-disciplinary knowledge are three areas that engineers should focus on.
Design simulations are increasing complex and accurate, however they still require validation. Therefore, it is important to plan from the outset what design aspects require more detailed investigation in order to understand what data actually needs to be collected.
Many engineers simply want data without taking the time to consider what is useful data. Useful data can be used for cross-correlation between design and test, in order to provide a form of validation of up-front simulations. Therefore, good communication at the planning stage between the design and test engineers will help identify the data that is required, and the tests that will generate it.
Q: Has the increase in product complexity added to the design-test gap?
A: Norton:
Yeah, I would agree with that. As you’ve got a lot of products now that have traditionally been mechanical products and now also have electronic hardware and software. So now instead of having to develop a mechanical design, you’re having to develop electronic and embedded software designs in parallel. The integration of those three disciplines definitely adds complexity, because you can test each one individually, and integration of those three designs is key. A 1D tool can really help with communication between these three disciplines.
A: Mittersinker:
Our human-centered approach has always had testing as a core integrated timeline. When we do agile development, which is a subtask-based initiative, the testing loop then influences consecutive refinement and engineering activities. It’s pretty much the only way we feel that you can develop and design products that deliver on some of the harsher and more extreme use cases that products have to perform on today.
Q: Why do engineers work with a product design firm to bridge the gap?
A: Mittersinker:
One of the key reasons is that companies have core expertise that might be around some sort of core subsystem of a product, but they don’t have the necessary expertise in design for manufacturing, or heat dissipation. As a commercialization house, we can provide that outside expertise.
A: Norton:
Part of it is new technology. A lot of our engagements are what we’d call a technology transfer. We’ll work on a project alongside the customer, and at the end of it, they’ll understand the process we’ve employed and they’ll be able to use it going forward. Additionally, we’re always looking at the new physics and simulation products that are being developed and ways to apply those in the design process.
Another reason people engage with us is just from a cross-functional interaction point-of-view. We can assemble a pretty small and diverse team of people, and sometimes in a larger organization that’s tougher to do. Being able to deploy a small cross-functional team can sometimes really add value to the customer if it’s hard to achieve internally.
A: Lavelle:
Time is a finite resource determined by deadline constraints, which can influence the decision to outsource a piece of work. This can be particularly problematic when something unexpected happens, for which planning has not allowed. This allows the onus of timely delivery to be shifted onto a third party, with the results expected on the date agreed in the contract.
Modern test equipment is certainly not cheap to buy (or run) and the initial investment cost can be prohibitively high. This is especially true for smaller companies, one-off products or tests that are rarely performed. As a result of high investment costs, specialist testing services are provided by a variety of companies, such as HBM Prenscia’s material testing services (AMCT).
Q: What should clients consider when working with a third-party product design firm?
A: Norton:
Expectation setting is key, and defining a clear scope of work. Oftentimes when you bring in an outside firm, it should be a collaboration going forward. [Engineers] should be prepared for questioning to understand why things are done a certain way.
A: Mittersinker:
It has to do with the fact that a lot of times you have to be honest [with the client and service provider]. With this honesty comes a level of trust. I would not take on a client without a very realistic understanding that we [as a partnership] are on the right path. When you have subsystems relying on thermal performance, adhesion or anything that needs a lot of testing, the yield component cost equation becomes very important and [as a provider], you have to be realistic.
A: Lavelle:
Testing providers often see both ends of the spectrum when it comes to test specifications. They can be vague, ambiguous and poorly defined specifications that are open to interpretation. Alternatively, the tests can be over-specified and unnecessarily aggressive, which can quickly increase the time and cost of testing.
Therefore, providers of testing services should always give a critical appraisal of any test specifications, providing advice and liaising with their customers in order to ensure the most appropriate and cost-effective tests are performed.
Subscribe to our FREE magazine,
FREE email newsletters or both!About the Author
Jess LulkaJess Lulka is a former associate editor for Digital Engineering. Contact her via de-editors@digitaleng.news.
Follow DE