Having a good factual model of a person would be necessary, and perhaps sufficient, for making that judgment favourably.
For a single person, yes, but it takes a significant investment of time to build an accurate, factual model of a single person. It becomes impractical to do so when making decisions that affect even a mere hundred people.
How would you recommend scaling this up for large groups?
Sociology and psychology. Determine patterns in human desires and behaviour, and determine universal rules. Either that, or scale up your resources and get yourself an fAI.
Another month, another rationality quotes thread. The rules are: