There are n queues, each with a different number of items (i.e.
fradulent transactions) -- these queues are populated once a day. This
is a two-fold question: a) what items in each queue are the highest
risk, and b) given that there are y agents who can work on z
items/hour, how deep should/can each queue be worked on? I apologize
that this question is a bit unclear, but this is all the information I
have. I would appreciate any responses to this problem ASAP in regards
to how this problem would be approached and a solution, if possible.
Just a note: the frame of this question is based on online payment
fraud activity and the approach on how such activity is mitigated
(with great consideration to minimizing resources and time).