Extensions and developments on the Schilling and Dodge mixed dependent acceptance sampling plans

The Schilling and Dodge (1969) [2] formulation of mixed dependent acceptance sampling plans   has a number of favorable properties; so we decided to build on their results and extend them to better levels. One genuine important weakness of their procedure is that it is not safeguarded against departure (of the parent distribution) from normality when accepting a batch based on the first sample by ‘average-testing ’ basis. One argues that rejection occurs only after re-sampling and, that too, on the ‘quantity-based attribute-testing’ basis according to the setup of their plans. Acceptance, according to their scheme, can possibly occur on the average-testing basis based on first sample. And as a consequence of the reliance of the variables plan (especially the σσ-plans) on normality assumption, and lack of distributional robustness, the acceptance decision may not be very solidly based. With the developments suggested in this paper we propose to modify the plan in a way to strengthen this weak area.