The requirements are good to know. They must not be too hard to meet given 69FireBird is willing to attempt this with a PLC. However, the time it takes to implement a flying shear or cutoff in a PLC can more than offset the extra cost of a motion controller no matter what the requirements are. There are basic requirements that must be met. I think these requirements tend to be underestimated. Then there is the support issues. PLCs don't have real time graphs and event logs that work at the millisecond or sub millisecond level. PLCs also don't ramp up or down using nice 5th order polynomials with a ton of math.
Is worth a look I think, we have a project coming up where we are using them also, going to be a learning curve, but with the right tools, should be an enjoyable one.