Push Or Pull

At work today Potatohead snuck up on me at my workbench and said he wanted to talk for a minute. I was in the middle of gutting one of our Sig Gens when the fool materialized out of thin air. The unit I was messing with had failed in test, technicians troubleshot it down to the main wiring harness and they called me in to do the dirty work. I had to replace the wiring harness which is a nasty job that takes time. Fifty solder joints have to be removed and cleaned, the instrument card cage has to be pulled, and the front panel must be dropped. Then the new wiring harness can be installed and connected up. It's sort of like doing a spine transplant. I was a busy cog in the great corporate machine at the moment. Reluctantly, I swiveled my chair to face the man and asked what was up.

Garden Tool apparently had some sort of meeting, or briefing session with Potatohead during the managed move of supervisors between Sources and us. Garden Tool identified me to Potatohead as one of the go-to people on the line. That's a compliment I didn't need. Anyway, Potatohead wanted to ask my opinion on a new method he invented for flowing the instruments through the area. He also hit me up about implementing his idea. If I liked it, he was hoping I would spearhead the project for him. I think he was sniffing around for a kiss ass...

Potatohead said in an overconfident tone, "Peasant, what do you think about going from a Push system to a Pull system?" He was referring to the que we have at each station on the product line. The way a que system works is simple, in theory. In reality it's useless. If you know you have X amount of units to build and test by the end of the month, and you know how long each step of the process takes time-wise, you can figure out how many boxes you should have in que at each station every eight hour shift.

For example, the Assembly I station might have say, five boxes in que on dayshift, then Assembly II should have five boxes in que by the time swingshift starts, and so on. Trust me, it's entirely stupid, total bullshit. Only gung-ho process engineers and doofus managers like Potatohead pay any attention to que systems. We are entirely at the mercy of the individual units themselves because they are totally complex. There are nearly thirty PC boards with over 6,000 components in every instrument. According to Murphy what can go wrong, will go wrong. These boxes get finished when they get finished. No sooner.

I complete an instrument at a step in the process, I push it to the next station. I grab another unit and work on it. Like mindless robots we repeat the cycle until everything is pushed out of our que. I like to stay busy so if I'm fortunate enough to wipe everything out at one station I'll hop over to the next and keep going there. So I thought about Potatohead's question for a moment. Push system to a Pull system. Well, what's the difference? None. Boxes are still going to be shuffled around on the shop floor regardless. They won't get completed any faster. I said to Potatohead, "This seems like you're trying to re-invent the wheel. It's the same thing, really. So, I guess I'd say I think it's dumb." I swiveled back to face my eviscerated test instrument and resumed performing surgery on it.