This "enhancement factor" of 5 to 20 and observed values up to 100 was reported in Paste 2006. Shearing in the upper zones of the thickener was then emphasised as a target area to enable this improvement in thickener performance.
I believe the current research is focused on the rake. If the same values are reported, then the actual mechanism (reason) for this enhancement seem to still be unsolved or unpublished.
This enhancement by the rake mechanism was reported in Ushers PhD thesis about 20 years ago and that might not be the first published observation. (I did not try to find when and where it was first mentioned). I really hoped we would progress faster, but as a lawyer friend one said of one of his high profile cases "if this case is handled properly, it could carry on indefinitely"

I had to slap myself when you said 20 years. My PhD thesis, Usher (2002) notes observations of performance enhancement due to shear which is well summarized in an AIChE paper, Usher and Scales (2005). I believe that Dorr-Oliver anecdotally noted such effects from raking years before any of us were born, but much of this knowledge was kept in house. As you have suggested, we have demonstrated that this enhancement can occur through aggregate densification in un-networked zones in the thickener as well as when raking a networked bed.

Heuristic factors abound in all aspects of thickener and drive sizing calculations even after extensive pilot test work has been conducted. Fudge factors and rules of thumb remain popular partly because they work but mainly because nobody wants to be the first to buy a prototype solution with lower "Service factors" designed from a new fandangled measurement in a pilot plant.

I fully agree with you that irrelevant experience is used to size the thickener and drive. I say the experience is irrelevant because thickening is currently being pushed into territories where high rate thickening experience is not sufficient. The so called k-factor was utilised as a sales gimmick to confuse the market and try to hide IP while still give a subjective, meaningless and confusing number for perceived comparison. Unfortunately this is what the market was trained do and every time the thickener vendors say "tumble" the buyer "rolls over"; it is difficult to unlearn the folly and relearn the correct.

The norm has been to oversize the thickener drive for pump failures and bogged units. While I thought that this never occurred in the modern plants, I have just recently learn that it occurred twice in a very new coal plant. It would be good to relate the drives offered by the vendors to pilot work or even bench scale work if that is possible? I was thinking of writing a paper to compare thickener drives offered on the Escondida project to the pilot work that we did on site. Any comments? Lastly, there must be some merit in the pilot rake speeds - they are always much faster than the installed units. The speed changes the compression structure promoting better expression - any further comments? Maybe a good time to have a general tele-conference with like minded thinkers