ISO/TS 16733 Event Tree Analysis and scenario selection 1

[Ciaci]

I am trying to understand how, once all the branches of the Event Tree Analysis are defined, it is possible to decide if a fire scenario should be analysed with a simulation. I don't know an 'a priori' criterion to quantitatively assess the damage (deaths or money), and I don't know which probability should be the threshold to discard a scenario labelling it as "not credible". I don't have the ISO/TS 16733, but I found some reports and books citing this code.
Do you know which is the criterion to select a scenario to assess its consequence with a numerical calculation or to discard it?

 

[BrianPetersen]

You are not going to get through this or any other risk assessment without making judgement calls. Hopefully it's someone with experience making those judgements, who can explain the rationale behind them.

It would sure help if the rest of us knew what sort of "fire" you are envisioning here.