Bayesian online robust parameter design for correlated multiple responses. Issue 5 (3rd September 2021)

• Record Type: Journal Article
• Authors: Yang, Shijuan; Wang, Jianjun; Ren, Xiaolei; Gao, Tingyu
• Is Part Of: Quality technology & quantitative management. Volume 18:Issue 5(2021)
• Page Start: 620
• Page End: 640
• Publication Date: 2021-09-03
• Subjects: Seemingly unrelated regression -- Bayesian inference -- direct Monte Carlo -- time series model -- noise factor; Quality control -- Periodicals; Quality control -- Statistical methods -- Periodicals; Industrial management -- Periodicals; Industrial management; Management -- Research -- Methodology -- Periodicals; Qualitative research -- Periodicals; Management; Quality control; Quality control -- Statistical methods; Periodicals; 658.00721
• DOI: 10.1080/16843703.2021.1952545
• Languages: English
• ISSNs: 1684-3703

Abstract

ABSTRACT: In production and manufacturing processes, noise factors are often considered difficult or costly to observe. The emergence of advanced sensor technology has made it easier for some major equipment to obtain large amounts of online monitoring data during the production stage of a product. In this paper, a new Bayesian approach is proposed to extend offline RPD to online multi-response RPD by making full use of this additional information. As new observations of the noise factor are obtained gradually, the settings of the controllable factors are adjusted online to further reduce the influence of noise factor variations on production quality. This approach not only addresses the correlation among multiple responses but also considers the uncertainty of model parameters and the variability of noise factors. A case study and a simulation study demonstrate that the proposed approach is superior to existing methods.