Control: You either have it or you don’t. 

Understanding if processes are in control is just one of the jobs that quality inspectors take on. They employ both inspection and testing, so that customer requirements are met and assembly operations stay compliant to regulations. Good things happen when there are controls in place and everything is running as expected.

While there are many different reasons why inspection is necessary, the ultimate objective is the same: find defects before they escape. When it comes to controlling manual assembly, inspection typically means physical and visual, non-destructive inspection done “in-process.”

Humans have done this type of inspection for hundreds of years; however, with the advent of visual inspection automation, high-volume manufacturers can now reduce inspection labor costs, lower inspector fatigue and improve accuracy. As promising as this is, not all manufacturing lines are a perfect economic fit for this technology. If your production volumes are high enough and assembly processes stable enough (low variability), optical inspection tools and vision systems are great for identifying missing components, misalignments, discoloration, etc. But lacking that volume and stability, the most practical and economical options are more manual. While they can be time-consuming, manual inspections are still very effective.

Manual inspection means humans are still in control

With manual inspection, you get the benefit of human reasoning. Standard operating procedures (SOPs) instruct your inspectors on how to conduct inspection, how many to do, where to look, what to measure. And they follow it. But, what happens when they see a pattern or notice something they’ve never seen before? Undoubtedly, they’ll ask why, whereas an automated system merely rejects what it can evaluate – scrapping units, or ignoring the anomaly altogether. 

Which brings us back to control. Are your assembly processes in control? 

The answer to that question lies in whether or not you know the source of errors, variability and nonconformance. In addition, control rests on procedures that feed that knowledge back to the process of assembly. When inspectors find a defective product, they decide to reject it, rework it or scrap it, but they never immediately decide to eliminate the defect.  

Vision systems, optical inspection and even manual processes, while important for detection, won’t help you eliminate defects – only processes that tell you why will help. When you know why something went wrong, you tighten standardized work, improve process steps, source better materials and ultimately, improve defect rates. 

Today’s AI technology has the ability to help control manual assembly and complement both manual and automated inspection processes. Systems like Drishti, with video-enabled data collection, have the ability to demonstrate why a defect occurred in manual assembly. It helps close the loop on inspection processes, so they can deliver their true value: addressing — and fixing — errors before they escape. 

Read how Drishti also drives holistic traceability on manual assembly lines by providing indisputable proof.