My initial thought was why go to the bother of building up then stripping down to measure high/low spots before machining again.
However if you take a sample of blocks, build them up, run them, then strip down and measure, you'll find the common high/low points for a particular design. With that info, you can then machine the new blocks accordingly.

But, given the extra costs involved, for what basically amounts to something that will only affect the run-in period, and maybe give a marginal increase in component life, I doubt many main stream manufacturers will use it. It's probably of more interest to high performance or truck manufacturers, where everything is about incremental gains for power or fuel economy.