Dear Arya,
The fact is that yr question is rather unanswerable in a general form as I'm sure you already knew.

It is unfortunately also often unanswerable in a specific case unless a regulatory situation exists. Similar problem to the infamous microbiological specifications.
One of the most common general approaches is to take a worst case scenario, eg the most difficult species to destroy within the typical flora for the type of product involved but even here, the criteria for "destroy" will vary and are often not spelled out, eg how many log reductions are required ?
Nonetheless, it's an interesting topic to discuss
Rgds / Charles.C
added - Steve's document is interesting but I am a bit cautious regarding the data, eg V.parahaemolyticus is apparently considered the most difficult species to kill for shellfish/crustaceans. I think this is incorrect, seem to remember that V.para is actually considered relatively easy to kill (in the sense of 1D reduction at a given temperature). Most commonly (but not invariably) in EU, I have seen L.monocytogenes selected with minimum 70degC / 2min or equivalent as a reference cooking parameter. Again, you need the actual matrix / data / basis of comparison. However, I suspect this L.mono criterion is not typically used in USA (

).
added2 - one interpretational reason for a difference in approach is possibly explained in the uploaded initial document in this thread -
http://www.ifsqn.com...?showtopic=1080