Fuzzy acceptance sampling plans for inspection of geospatial data with ambiguity in quality characteristics

  • Authors:
  • Xiaohua Tong;Zhenhua Wang

  • Affiliations:
  • Department of Surveying and Geo-Informatics and Key Laboratory of Modern Engineering Surveying of State Bureau of Surveying and Mapping, Tongji University, 1239 Siping Road, Shanghai 200092, PR Ch ...;Department of Surveying and Geo-Informatics and Key Laboratory of Modern Engineering Surveying of State Bureau of Surveying and Mapping, Tongji University, 1239 Siping Road, Shanghai 200092, PR Ch ...

  • Venue:
  • Computers & Geosciences
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a method for designing a fuzzy acceptance sampling plan (FASP) for the case where sampling parameters are defined as fuzzy numbers, in particular for quality inspection of geospatial data with ambiguous quality characteristics. In contrast to existing fuzzy sampling plans that concentrate solely on ambiguity in the fraction of nonconforming items, the proposed method includes three cases with different fuzzy sampling parameters. These are the fuzzy fraction of nonconforming items, fuzzy sample rate, and fuzzy fraction of nonconforming items and fuzzy sample rate together. The design of the fuzzy sampling plan is modeled as a fuzzy optimization problem dealing with two cases in terms of lot size based on the fuzzy Hypergeometric and Poisson distributions. The proposed method is implemented to design fuzzy sampling plans for quality inspection of geospatial mineral products in Qinghai Province, China. The results show that (1) the proposed method has the advantage of performing quality inspection for geospatial data products with uncertain parameters; (2) in contrast to a traditional sampling plan having a single OC-curve, the OC-band of a fuzzy sampling plan has the lower and upper bounds; and (3) in contrast to existing fuzzy sampling plans which account primarily for uncertainty in the fraction of nonconforming items, the proposed method completely covers fuzzy sampling parameters.