'Artificial' Intelligence in the Bindery

Many years ago (too many to admit!), I worked in a hot-type letterpress shop in NYC. The bindery manager wore a short-sleeve shirt with the sleeves cut off, and had biceps that Schwarzenegger would approve of from muscling all those paper lifts around.

He also had decades for experience in handling paper, and in setting up and troubleshooting all of the various buckle folders, guillotine cutters, and saddlestitchers and perfect binders within his domain. He's a part of history now, as are the countless hands-on bindroids that went through multi-year apprenticeships and on to careers in finishing.

As the nature of print and finishing have changed, so has the equipment. Very long runs are less frequent. I remember crews taking two days to makeready a binder that would then run for three weeks straight. Much of finishing today involves shorter-run lengths with many more change-overs for size and format. Automation and "machine intelligence" are crucial. You can't turn out 30 jobs a day (or more) if you need a half hour to set the machine up.

So today's cutters, folders, saddlestitchers and binders are extremely "intelligent." This is a necessity due to the need to reduce makeready time, and to compensate for the lack of "old hand" experience in the bindery. In today's digital print environment, training apprenticeships have all but disappeared, and an operator may be thrown into production with a few days worth of training. The rise of smart machines that can handle hundreds of adjustments on their own was inevitable.

But paper hasn't automated itself, and here's the rub. It's still an organic medium which requires a knowledge of what it likes (and what it doesn't) and how to best make it perform for a variety of operations. We deal with its vagaries all the time on our systems. Making it perform properly in our finishing systems is a delicate balance of proper humidification, static control, coating control, and more. It means deploying a multitude of special rollers, belts, guides, and more to make it behave as we wish.

And this requires knowledge that can only be acquired by learning from someone who has the experience which can be transferred. Artificial (machine) intelligence is not yet up to all of the tricks and techniques that are still required to make the bindery productive.