unless you have prior experience, do a test fill and get your QA/QC crew to observe and test for density. once you do that you can specify the rolling pattern and method
Test strips are commonly used for asphalt placement to determine rolling requirements; however, the process can be used for base materials as well, though you can expect more variability with the base material than with asphalt.
Do a search on the following string and you'll find lots of references to test strips....
Agreed on the comments. As far as asphalt, which has changed considerably over the years, my observation historically has been as follows using a nuc gauge on backscatter. Mat laydown by the machine with a vibrating screed was what I'd ballpark at essentially 88%. Each vibratory roller pass picked up 1 to 2% density. Our State spec is 93% minimum compaction. Watching the growth curve generated during the test strip, one can get a feel for the number passes. If you don't see density between a minimum of 2 to 4 vibratory passes (I count "1 up = 1 pass", "1 down = 2 passes", ...), something is wrong This is where science gives way to art. You have to watch for overlapping passes, tearing at seams, shoving, travel speed, ... Way more to a good product than just passing density.
Earthwork is even more subjective, but watch for the sheep's foot pads starting to walk above the lift. You now also have moisture as an issue and I you'll be looking at the overall depth of the fill, lift thicknesses, ... gravel will be a little more straight forward. Most of the time in my area, proof rolling with loaded tandem is using to proof roll the subgrade and base before proceeding to paving.
