I went on with my extensive online search for better comparable data but unfortunately I couldn't find any other publication other than the one from Flight Global already mentioned above.
From the many forums with pilot talk and flight sim optimizations etc it seems the data for the 767 is generally ok but the ones for 757 and A310 are too high.
The problem comparing fuel consumption in AWS is that it is stated as an average, considering a full flight profile without indication of flight time used for calculation.
A better number for comparison would be: Fuel consumption per hour during cruise
So: If the numbers provided by AWS for the B757 are for a 2 hour sector and the numbers for the B767 are for a 10 hour sector, then sure the numbers would
add up. The same applies to the A310. If AWS took a short sector as average then the numbers would indeed appear higher due to the climb penalty and the
disproportionate extra weight for carrying reserves.
So in order to be able to compare apples with apples, a standard fuel consumption per hour in cruise would be a lot more helpful, especially for longer range a/c.
I am tempted to order an a/c each of A300/A310/B757/B767 and put them on an identical 3000nm flight to see the fuel consumption results. If anyone has several of these birds in fleet and would like to share their results that would be really nice.
Cheers!
Mike