A sampling scheme has been proposed by our statisticians that allows for a statistically based sample size to be created in the OQ which is then applied in the PQ to calculate capability. The advantage of this method is that the sample is taken by combining all 3 lots in the PQ rather than taking samples from each lot individually. The downside is that if it fails all 3 lots fail.
My question is whether FDA would accept this method of combining the lots.