Hey John, thanks for your question!
Technically the DAO can decide to add or remove operators whenever it wants to. Historically, new operators have been added a few times per year so that new independent NOs are added to the set, and to better distribute stake amongst them . There is also a soft target that no NO should have more than 1% of total ETH staked allocated to them via Lido, as explained in the Operator Set Strategy (this is not a ratified policy at the moment, but I think it would be opportune to do so once V2 is out and some considerations around V2 modules are added).
As at end of Q1/23, the majority of operators are above this soft 1% target
As you pointed out, the protocol is currently undergoing the V2 update with which native withdrawals will be enabled to allow users to unstake directly via the Lido contracts. In terms of timing, I think it makes sense to wait to see what effect the unlocking of withdrawals has on total ETH staked via Lido. If withdrawals do not meaningfully affect the total ETH staked, it would make sense to commence a new onboarding round (open up applications) around early June. If they do, it may make sense to wait a bit to see where the stake level settles, but I personally think that an onboarding would be useful regardless, it may just change the timing from early June to something like August.
Also, if possible to answer, what will be the main criteria for assessment(with Shapella upgrade are there additional criteria that LNOSG wants to assess further?) and how many will be selected for the next wave?
There’s no specific post-Shapella criteria that I can think of. In terms of how many would be in the new cohort, it’s difficult to say, as the primary driver is how TVL may be affected by withdrawals. IMO it should at least be enough such # of operators > percentage points of Lido stake share + little bit more to allow for growth before another onboarding round.