Reproducibility:

The method should be such that it gives reproducible results of separation. It is usually expressed in terms of standard deviation or coefficient of variation based on replicate measurements. Present at this point, it might be significant to point out that the separation conditions should not be such that a slight variation in the conditions may alter the result significantly. Flexibility in the conditions of separation is always preferred. This is particularly important if the process is to be scaled up on a commercial scale.