"Martin Rapier recently published his high-level operational rules"
Well, sort of recently. IIRC I actually write them in 2002, it just took a while to filter through to the Journals publishing cycle;) They've been through various iterations and have finally mutated into 'Panzergruppe' which now seems to be fairly stable.
If looking for Commadn Decision levels of exposition, then look elsewhere.
I've used these (or variants) to do a number of large scale operational games including Market Garden, Goodwood, Crusader, Gazala and Mansteins dash to the Dvina.
Original set and game commentary on my homepage:
link
as is Panzergruppe.
PDF link
IIRC these are linked from freewargamesrules.co.uk too.
We also did the 1967 War in Sinai using Megablitz, but that took all day rather than an evening.
"It's a six turn game?"
That seemed logical at the time:) Real battles last for days, weeks, months, and a lot of stuff happens overnight IRL (like resupply etc). Hard to fit an entire war into an evening with shorter turns.
Arnhem was a ten turn (day) game, but there were five Corps equivalents engaged. Goodwood I went with 12 hour turns, but there were 14 divisions engaged as I included the Canadians (who are often overlooked).
I've only ever really done AiW with elements of batalions size or greater, althuogh I did do some skirmishes with platoon bases.
I wouldn't claim these are the ultimate operational wargames rules, but they work for us. It is intended to be played with the strength points hidden, which makes quite a difference to game play. Casualties are function of combat outcomes and scale with target density, rather then being the determinant of combat outcomes. ie if you lose badly you take heavy losses, if you lose badly with a lot of units, you take really heavy losses. You don't work out the casualties to see who has 'won' though, so it is a bit more like a boardgame I guess. This means that a small number of very cohesive and effective units can tear huge holes in large numbers of less cohesive ones, rather like real life, although I've toned down some of the effects for game play purposes. Even in Sinai 67 the IDF can only expect to inflict a 4:1 loss ratio on average, more if they manage to concentrate. The relationship between target density and casualties isn't quite right (absolute losses should go up but percentage losses decline) but in such a simple combat model it was easier to go with a linear relationship.