Understanding the Concept Validation Phase in the NPI Framework
Last week, a client asked me: "Why do we always trip up during the concept validation phase? We think the technology is feasible, but then a bunch of problems arise during mass production." That's actually quite representative. Today, let's chat about what exactly goes on in the concept validation phase, and why it determines the life or death of new product introduction.
To put it simply, this is a critical phase for "testing the waters with small investments." Think of it like a developer testing the geological structure on a small plot of land before building a skyscraper—there’s no way you’d just start construction after pouring hundreds of millions into it right away. In manufacturing, this phase needs to complete three key tasks: technical risk identification, cross-disciplinary collaboration, and minimal resource investment. Someone might ask: How is this different from later engineering verification (EVT)? Simply put, the former verifies whether something can be manufactured at all, while the latter checks if it can be stably mass-produced.
Inside the POC Phase
Technical feasibility studies are not just theoretical exercises I’ve seen too many teams confidently present simulation data only to crash and burn during prototyping. True experts know that DFM (Design for Manufacturability) must be involved from the start of the POC. For example, when an electric vehicle company was validating solid-state batteries, the process team directly required electrolyte layer thickness tolerances to be controlled within ±0.05mm—not just for precision, but to prepare for later roll-to-roll processes.Prototype development isn’t about showpieces Here’s a funny case: A smartwatch team rushed to create a 3D-printed prototype with features like automatic heating in the strap. During review, the process engineer shot it down: “The cost of mass-producing this would be enough to build ten real cars!” So remember, rapid validation should focus on core parameters, such as medical device tubing burst pressure requirements being ≥300psi—forget the flashy extras for now.
Smart ways to handle cross-departmental debates I recommend trying the "Iron Triangle Review Mechanism." On one industrial robot project, R&D, process, and procurement sat together, and the SQE (Supplier Quality Engineer) pointed straight at the blueprint: “This servo motor model has been discontinued for two years—you’re still doing verification on it?” It saved at least three months of detours. Now you understand why cross-functional collaboration mechanisms are so important in the POC stage, right?
Lessons Learned the Hard Way
An EV battery company made a big mistake during the POC phase last year. They tested energy density and passed, but ignored degradation curves under extreme temperatures of -30°C. Then complaints started rolling in six months after mass production... This clearly shows that verification reports must include multidimensional data, such as pressure-flow curves in hydraulic system validations like this example. Visualized data is king.Speaking of which, I suddenly remembered a story: A medical equipment factory recently introduced digital twin technology and simulated 100,000 bending cycles during the POC phase, catching material fatigue defects early on. Definitely worth learning from, since digital twin technology in POC applications can save huge prototyping costs.
Industry Secrets You Need to Know
Honestly, many companies still see the POC phase as nothing more than "making a sample to take a look." Those who truly understand the game set clear technical readiness level (TRL) evaluation criteria from the start. For example, aerospace engine parts must reach TRL4 to pass, which is completely different from consumer electronics metal parts requiring SPC simulations.Finally, here's a question for everyone: Do you know how DFM and DFMEA should work together during the POC phase? I suggest checking out some specific examples of DFMEA failure mode analysis—it’ll definitely be more interesting than reading a manual. Oh, and next time you run into a technical decision dilemma, try scoring using the TRL assessment method. Trusting data beats guessing hands-down.
(There’s a lot of industry insider knowledge in this article—definitely worth bookmarking to read again later! If you feel anything wasn't explained clearly, hit me up in the comments!)