Creating a Logical System of Ethics

Sklansky's way of trying to use probability and math in conjunction with ethics makes a lot of sense to me.
However, I think what is missing is that in order for these techniques to work in a practical way, a large group of people would need to embrace that methodology.
On top of that, this group would need to assign some basic numerical values to common situations (ex a guilty man going free) in order to prioritize correctly.
Thus, when dealing with Sklansky's death penalty situation (discussed in 2+2 magazine), he arbitrarily comes up with values for murderers going free, innocent men being executed etc.
My contention is that a group of people (logicians, philosophers, policy makers) needs to somehow agree on the values given for each scenario. Example: What is the numerical value given to a woman dying if she does not have an abortion, vs the numerical value assigned to the potential life of a fetus, vs an individuals right to make a choice? Can you quantify choice in this situation?
These kinds of numerical values could be applied in many different ways and then there would be an organic, logical way to make ethical decisions based on the available knowledge.
The numerical assignments would constantly change as new information became accessible.
But without the agreement on the basic concepts, that way of working out ethical problems using probability, odds and so forth can never work.

g