I imagine that there are 'decision support' tools that suggest and model outcomes, and then flight planning take that output and construct, in that beautifully elegant piece of computational hardware called 'their brains', a plan. It's a hugely complex logistical dance, and there are likely to be some factors that can be accurately (read reliably, safely and profitably) modelled by computers alone, and some that simply can't.

Crew flight time limitations for example - Commander's Discretion cannot (or at least certainly should not) be assumed by home base in terms of ad hoc extensions to hours worked. And my understanding of how it happens in practice, when and if it DOES happen is that the first thing the skipper does is take an assessment of both hard hours worked, and ideally, although somewhat more subjective, current fatigue levels, of the crew (flight deck AND cc - remember first and foremost their brief is pax / cabin safety), to determine what can sensibly be done on a discretionary / best efforts to aid planning team in keeping the show moving. So that sort of thing probably doesn't fit well in a spreadsheet as there will, I presume, be considerable variance depending on skipper, situation, crew etc. - the only 'knowns' are just how much maximum latitude that gives the skipper within regs / SOPs.

And that's just one example of a hard to model variable - there will doubtless be tons of others....

Staffers - please point out if this is all baloney, but that's my understanding in very broad terms...