Lee Taylor
Registered
Hi,
It’s not the little surface mount comments that are being missed. These are placed by machine and inspected thought Automated Optical Inspection. It’s the larger conventional though hole components.
Indentifying human error as a root cause
- Thread starter Lee Taylor
- Start date
Hi,
It’s not the little surface mount comments that are being missed. These are placed by machine and inspected thought Automated Optical Inspection. It’s the larger conventional though hole components.
Are you certain it's the Operators? Do you absolulety know they are given the correct quantities of parts in the first place. It's difficult to fit/inspect for parts which were never supplied in the first place.
Johnnymo62
Haste Makes Waste
Thanks for your service!
Lee Taylor
Registered
Hi, Im 100% sure the operator has missed fitting the component. Even if they werent given the correct quantities there would be an empty location on the PCB that should be checked prior to going for inspection.which either the operator should notice or inspection should spot the missing device
I bet he never did that again, lesson learnt. Accept our operarors keep repeating the same mistakes which I trying to prevent happening
Hi Lee, if the operator is missing insertion of a component, the PCB assembly/CCA should not move to the next oepration, the condition of a missing component can be shared real time with the operator...also if it is a repeat occurrence, then what some call a OPL One Point Lesson, i.e. a visual cue/sheet at the work station to heighten operator awareness
If they were given too few components, they may think they must've installed everything needing installed. If they were given too many, they can't use the number of parts issued to guide them on whether everything is installed.
Again, I don't know what the board in question looks like (component density, number of components, etc.) but on some boards I've seen it would be almost impossible for the human eye to determine if everything was placed with 100% accuracy in a real-world setting beyond just a few boards. Don't criticize the operators and inspectors until you've tried to do it yourself. If you can't do it 100% why would you think anyone could?
Have you studied the kitting process to have data on it? "Knowing" comes from data. Unless you have the kitting process under control, then what you're asking the operators and inspector to do is a practical impossibility. For example, did you notice how many spelling mistakes and punctuation mistake you made in your post?
Lets stay on topic here...this isn't grammar class.
A number of ways have been proposed in this chain so far, from kitting, to automated inspection at the assembly station, to quick tests (electrical) for continuity...
Have you tried any or all of these? What were the specific results (assuming poor if we're still going here) from each technique tried?
I suggest that checking IS inspection...before going for inspection...
A number of ways have been proposed in this chain so far, from kitting, to automated inspection at the assembly station, to quick tests (electrical) for continuity...
Have you tried any or all of these? What were the specific results (assuming poor if we're still going here) from each technique tried?
It wasn't a lesson in the English language. It was to demonstrate that errors are made and not picked up, because they slip by us. It's the process. But thanks for missing the point.
Matt's Quality Handle
Involved In Discussions
I'm going to echo the kitting, or some other kind of parts accounting system. You don't go into the complexity (number components per board, number of different resistors/capacitors, etc.), but accounting by quantity can help.
This is a bit of a twist, but during surgeries, all sponges come in sealed packs of 20. When done with a sponge, it is thrown on the floor. The patient is not sewn up until all 20 sponges are positively accounted for.
I used this for another problem, where parts were making it to the customer unassembled. An operator loaded a large part and a small part, and an assembly cell riveted them together, then the operator placed in FG bin (in bins of 16). Several times, unriveted parts made it to the customer. I had the HMI programmers put on a simple counter that reset at the start of the shift. Before going to breaks/lunch, the operator had to reconcile the count. Total piece count had to equal the good parts + any scrap (usually a misfire of the riveter).
Maybe something similar, buy a scale with weigh counting capabilities. Scales that can weigh count are pretty cheap, so periodically, the operator must account for the resistors installed vs. planned. If there's a mismatch, all production is held until inventory is confirmed.
This is a bit of a twist, but during surgeries, all sponges come in sealed packs of 20. When done with a sponge, it is thrown on the floor. The patient is not sewn up until all 20 sponges are positively accounted for.
I used this for another problem, where parts were making it to the customer unassembled. An operator loaded a large part and a small part, and an assembly cell riveted them together, then the operator placed in FG bin (in bins of 16). Several times, unriveted parts made it to the customer. I had the HMI programmers put on a simple counter that reset at the start of the shift. Before going to breaks/lunch, the operator had to reconcile the count. Total piece count had to equal the good parts + any scrap (usually a misfire of the riveter).
Maybe something similar, buy a scale with weigh counting capabilities. Scales that can weigh count are pretty cheap, so periodically, the operator must account for the resistors installed vs. planned. If there's a mismatch, all production is held until inventory is confirmed.
