I am unsure about part 3 regarding the 10 timeslots for the papers.
It's a definite one-way model with different judges rating the same paper each paper receiving 3 ratings. There is virtually no agreement among selected pairs except for 1 pair and no agreement among all of them.