Assuming we're dealing with the actions of two people based on the preferences of two non overlapping sets of affected people, other than the number of people, which methods are available to resolve weighing conflicts on the same preference?
I know there's a hierarchy of preferences, staying alive trumps avoiding pain, but how does this famously criticized 'mathematical' weighing of seemingly equal preferences work exactly?
Going off of Cody's comment, and speaking as a layman on the subject (the last time I formally studied utilitarianism was in high school), it seems to me that in a sense utilitarianism is a meta-ethic. I would claim that any ethical system can be reconstructed - without changing it's meaning or consequences - to be described in utilitarian terms.
Any functional ethic must attempt not only to identify what is right and wrong, but how right or wrong it is. As such, these are effectively weights on the utilitarian scale determining the correct course of action. Even ethics with absolutes, such as the categorical imperative, can still be re-imagined in this way, and violations of the imperative simply outweigh all other actions unilaterally.
And so, to respond to your question which methods are available to resolve weighing conflicts? I suspect the answer is the field of ethical philosophy.
I agree with Cody and Dimo. I believe the key to Utilitarianism is assessing the possible outcome of your actions, and weighing their various effects as far as you reasonably can. The limits of what would still be a reasonable conclusion ("is it reasonable to conclude that two lives are worth more than one in the given case?") could be established in various ways, such as by another specific system of ethics. There will always be large unknown variables.
The opposite is anything that precludes any weighing at all, such as cast-iron principles ("I will not take another's life, no matter what") or ethics solely based on intentions that are unrelated to results ("if you do it for God, it doesn't matter what you do; if you do it with the intention of pleasing God, you may kill as many as you feel is necessary").
However, it could be said, which Dimo hints at, that all ethics unwittingly work within a Utilitarian framework, rendering the term meaningless: if pleasing God is the highest goal in ethics, because that is my only moral obligation, I can weigh the results of my actions according to that model. Shall I destroy Timbuktu or not? I know God commands me to do it, and so Timbuktu's destruction is the moral thing to do. Not destroying it weight zero.
Thus it would seem reasonable to include in the definition of Utilitarianism some more specific goals, like people's physical well-being as a positive weight, or the fulfilment of their desires. But even these two goals are controversial among Utilitarians.
In practice, whenever I think "your decision is impractical and could profit from a more Utilitarian point of view", I just want to tell people, "think more about the practical results of your actions, instead of these mere principles; you have adopted them seemingly at random, and you are unwittingly giving them far too much weight in your narrow-minded motivation, forgetting the true purpose for which you have invented those principles in the first place".
A short answer is Revealed Preference Theory. You need to know a bit of calculus to understand the math behind it, but the essential claim is that we can observe people as they go about their daily business, and their preferences will be "revealed".
Here's a classic problem: should we allow some factory to pollute a lake? If we let the factory pollute, people will be benefited via cheaper goods, but others will be harmed by the pollution. How do we figure out which is better?
To figure out the cost of a polluted lake, we can study things like how much the price of a lakeside house decreases after the lake gets polluted, the decrease in tourism, etc. From this, we can learn how strongly people prefer to be near non-polluted lakes.
Conversely, we can examine how much people prefer to buy this cheaper good by examining its elasticity. Wikipedia has some stuff on determining demand elasticity. For more detail, I'd recommend reading some of the papers there which actually calculate the elasticity of various products.
The general class that these problems fall into is Cost-benefit analysis. Social ROI is another buzzword, the "Social e-valuator" has a buzzword-filled intro that you might extract some understanding from.
In contrast to the other answerers, I think this area has some of the strongest foundations of any part of philosophy. Not only can we in theory determine the weights, be we do (very frequently!) determine these weights in practice.
