I won't comment to the day-to-day activities on de-coking operation for any given specific application.

It makes sense generally to de-coke one at a time, since it's a batch process, though I'd expect the intent could be that the third is an emergency spare and is generally unused.

However, we're obligated to assume worst case scenario in the absence of better information, so you have no guarantee which 2 they'll favor at any given point in the future, or even abandon the sparing philosophy if one even exists.

Dialogue with the end user / operators is important.

I wouldn't step out on any limb to say duration and frequency, since every facility is like a snowflake, it seems.
These folks offer a range of 10-24 hours.

With respect to the question of considering pressure vs temperature vs pressure and temperature: If you can demonstrate that the differences between the 3 cases are substantially different/worse from simply combining P&T, then it's one of those situations where it becomes an obligatory assumption in the absence of better data.

I would expect by and large the results to be comparable, though. There will be a minority of cases where pressure or temperature will work to cancel out one-another's ill-effects that might be picked up by looking at SUS -> P, P -> P&T, P&T -> T, T -> SUS versus SUS -> P&T.

At the end of the day, CAESAR will report total permissible cycles for each of these fatigue cases, and those permissible cycles can be what's ultimately approved, in lieu of total life cycle usage.