Title:

Dodge Skip Sampling: A Cost-Effective Plan for a Statistically Based Sample Size Reduction

Author(s):

Lloyd Nino Unabia, Jennifer Calipes, Allegro Microsystems Phils., Inc.

Gist:

This paper discusses the use of Dodge Skip Lot sampling in the reduction of sample size in the monitoring of stable characteristics. Lead Fatigue testing was used as the case study for this paper. As opposed to continuous lot sampling that's normally done, Dodge skip sampling requires the inspection of just several representative lots at predefined sample sizes. Thus, it is basically monitoring quality using the lot as the inspection unit, instead of individual units from a lot. Dodge skip sampling consists of the following steps: 1) determination of a single sampling plan called the reference sampling plan; 2) use of the reference sampling plan to perform normal lot-by-lot inspection of units endorsed for acceptance; 3) shifting to skip-sampling if no failures are encountered; and 4) return to the reference sampling plan if a lot is rejected. Characteristics of Dodge skip sampling include the following: 1) it does not consider lot size in the sampling; 2) it emphasizes homogeneity of lots; 3) it is only used on lots with good quality history. Dodge skip sampling is recommended as a means of minimizing sampling size only for processes wherein stability is guaranteed.

Published in :

ASEMEP 2001 Proceedings; June, 2001