Work

Single and Sequential Sampling Plans for Multi-Attribute Products and Multi-Class Lot in Reliability Test

Year: 2019

Type: article

Abstract: Reliability tests are always conducted to evaluate and improve the reliability of critical engineering equipment. Considering that the one hundred percent inspection is too expensive and time-consuming, an effective sampling plan is required to assess the reliability of engineering products. Generally, a product usually has more than one quality characteristic, and sometimes it is necessary to classify both the product and the lot into multiple classes. However, in the previous research, very few scholars have considered the multi-attribute multi-category products and multi-class lot simultaneously. This paper aims to fill in this gap by proposing single and sequential sampling plans for multi-attribute products and multi-class lot. Specifically, in the case of two-attribute three-category products and three-class lot, a single sampling plan and a sequential sampling plan are presented, respectively. The extended operating characteristic functions are derived by using the finite Markov chain imbedding approach. Two designing methods are proposed and the corresponding optimization models are constructed for each sampling plan by analyzing the operating characteristic surfaces.

Source: IEEE Access

DOI: 10.1109/access.2019.2923316

Authors: Xian Zhao, Siqi Wang, Leping Sun

Institution: Beijing Institute of Technology

Language: English

Topic: Advanced Statistical Process Monitoring

Subfield: Statistics, Probability and Uncertainty

Field: Decision Sciences

Domain: Social Sciences

Open Access status: gold

Estimated APC paid: $1,750

Funder: National Natural Science Foundation of China