When a manufacturing line is running well, it's easier to mess it up than it is to make it run even better.
If you process is running at 90% of its capability, you have to be very careful about intervening... The reason for this is there's a lot more to lose than there is to gain at such high performance level.
For example, you may be tempted by the potential ROI of some new technology that will take you from 90% to 92% of your process capability. After all, a 2% improvement for every battery produced could have a significant impact on margins. But you'd have to execute flawlessly to realize this ROI.
Contrast this with operations experiencing significant downtime and consistent quality issues – perhaps running at 50% capability. In this case, it’s easier to introduce change because the problem is obvious and more costly. People also understand that moving the needle from 50% to 85-90% is easier than going from 90% to 92-100%. And the payoff is bigger.
I’m not saying you shouldn’t be reaching for small yet meaningful improvements. But I think the priority should be finding improvements with big upside potential and little downside potential.
I know many battery manufacturers have the same bottleneck: Finished plates that won’t fit in the battery case. The result is increased scrap and rework, which leads to production lines running far below their capability.
This bottleneck can be removed with a relatively simple intervention. We designed our benchtop mini laser thickness gauge for this purpose. It will tell you whether your plates meet spec before they go through the assembly process. It’s quality control upfront, so you can avoid wasting material and applying paste to out-of-spec plates.
But the best part is it’s plug-and-play. There’s no need to retrofit anything. All you have to do is enter your plate types and calibrate the device, which takes about 5 minutes.
Once you address key bottlenecks and get your processes running at around 90%, then you can move on to getting as close as possible to 100% using automatic sensors and control systems.
But first do no harm.