Multiple Dependent State Sampling Plans Using Probabilistic Quality Regions (PQR) for Lot Acceptance Based on Measurement Data

Abstract

Usually Sampling Plans are designed with quality levels like AQL, LQL, and MAPD. This paper introduces a new method for designing Sampling plan based on range of quality instead of point-wise description of quality. So this method can be adopted toward elementary production process where the stipulated quality level is advisable to fix at a later stage. This paper proposes a variable multiple dependent (or deferred) state sampling plan through Probabilistic Quality Region (PQR) for the inspection of normally distributed quality characteristics. The decision about the acceptance of the lot is based on the states of the preceding lots (dependent state plan) or on the states of the forthcoming lots (deferred state plan). The lot acceptance probability is derived and the two- point approach for determining the plan parameters is described. The advantages of this new variables plan over conventional sampling plans are well discussed. Tables are constructed for the selection of parameters of this plan under the specific values of producer’s and consumer’s risks, indexed through acceptable quality level and limiting quality level, when for standard deviation is known or unknown.