Working for manufacturer engineering couple years. Sharing the manufacturing experience, skill, DFx, design, and information. Be care of this blog is personal share and content may not be 100% correctly.
When Workingbear first joined an electronics final assembly factory as a product/process engineer, I often had to attend MRB (Material Review Board) meetings. From time to time, people would come to me asking me to evaluate rejected parts and decide whether they could be waived for use. Otherwise, the production line could face material shortages or even a shutdown.
What I disliked the most back then was dealing with rejected plastic parts. It was always a tough call. Unfortunately, in final assembly, more than 70% of rejected materials were related to plastic components… not exactly a fun situation.
Most other electronic components were relatively easier to handle. Electrical issues could be passed to test engineers. For expired materials, as long as solderability was verified and no reliability concerns were found, they could often be released for use. Meanwhile, SMT engineers would take care of baking components or handling bulk parts.
Note: MRB (Material Review Board) is usually made up of teams from distribution, planning, quality, and process engineering. Sometimes production, R&D, and purchasing are also involved. The goal is to discuss and resolve production issues—especially cases where materials do not meet specifications and cannot be released to the line.
Plastic parts, however, are a different story. Their dimensional variation can be quite large. During injection molding, even small process variations can lead to warpage, shrinkage issues, or out-of-tolerance dimensions. Sometimes these deviations don’t really affect functionality, but designers often assign very tight tolerances—some even labeled as “critical dimensions.” As a result, we often had to take these out-of-spec parts and physically assemble them to see if they were still usable, check whether functionality was affected, and sometimes even go back to discuss tolerance settings with design engineers.
Ideally, if there were enough time and resources, these issues should have been identified and resolved during the DFM (Design for Manufacturability) review before mass production. But in reality, product launch timelines keep getting shorter, and people often use schedule pressure as an excuse—intentionally or not—to skip these upfront checks. So instead, problems are handled on the fly during production.
Dimensional and structural issues with plastic parts are still manageable, because in most cases, the decision ultimately falls on the process engineer—like Workingbear. After all, if functional problems show up later, the process engineer is still the one held accountable.
The real headache is cosmetic issues. Whether they are acceptable or not often depends on subjective judgment. In these cases, we usually work with quality engineers to define standards. For parts with different colors, we use a colorimeter and measure values like L, a, b, and ΔE. But in the end, visual inspection still matters. We check parts under both fluorescent lighting and natural sunlight before making a final decision. Sometimes the measured difference is small, but it just doesn’t look right.
For scratches and contamination, we also define specifications. Scratches are evaluated based on depth, width, and length. For contamination, we define criteria such as spot size, quantity, and spacing. One of the more challenging areas is surface texture. After a mold runs for a certain number of cycles, the texture can become less distinct or fade. In these cases, we usually create samples and have engineers from different teams sign off on a golden sample. When issues arise, we compare against that signed sample. Nowadays, with better digital imaging, we also require photos to be documented in the specification, to avoid discrepancies between the original approved sample and what’s being used later.
There was also an interesting question I ran into back then: why is the edge defect on metal parts called a “burr,” while on plastic parts it’s called “flash”? I always thought “flash” sounded more like something shiny—so shouldn’t it apply to metal instead? Later, a colleague explained it this way: originally, the word “burr” referred to a spiky plant seed. When metal processing first developed, people noticed similar sharp edges on metal parts and used the same term. Plastic parts came later, and to distinguish the two, the term “flash” was used for plastics. I’m not sure if that explanation is 100% accurate, but it makes sense to me.
So, does your factory also have an MRB system? And do your MRB meetings get heated? In Workingbear’s experience—yes, quite often. Everyone tends to have their own point of view.
Here’s something worth thinking about. While MRB meetings are meant to address material issues, not many people go further to dig into the root causes after the meeting. If we apply the 80/20 rule and identify the most frequent 20% of issues from MRB cases, then manage them using SPC methods—and if supplier-related issues are addressed through 8D reports with proper follow-up—we should be able to reduce the frequency of MRB meetings over time, and also reduce the number of times engineers get chased down for waiver decisions.
Leave a Reply