.png){kind=icon}
.png){kind=icon}
Return
Generalized quick-switch sampling systems indexed by Taguchi capability with record traceability
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cie.2022.108577.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Acceptance sampling plans of variable measurements quantified by process capability index have become compelling because they can validate how well the upstream lots (process) fit the downstream high-yield re-quirements in a more discriminatory and economic way. Recent research has accomplished two contrasting systems of quick-switch sampling (VQSS) based on the Taguchi capability index, namely acceptance criterion -switching VQSS (Balamurali and Usha, 2019, published in Journal of the Operational Research Society) and sample size-switching (Wang and Wu, 2021, published in Computers & Industrial Engineering) VQSS. Both studies, nonetheless, lacked in discerning the system's inspection strengths and weaknesses, nor investigated their joint performance regarding sampling efficiency and sensitivity towards abrupt deviation of the process capability. Moreover, for a sequential lot-inspection process, their simplified mathematical formulation has overlooked the importance of lot-and-sample identification in the supply chain. To forward improvements, we develop a generalized VQSS (GVQSS) system with lot-and-sample records traceability. The GVQSS system ac-tivates the concurrent changes of sample size and acceptance criterion based on the past result for lot disposition. It can also accommodate two existing VQSS systems. Additionally, an R-Shiny interactive web app is constructed for practitioners to easily acquire the optimal criteria of VQSS systems. These advancements enable the GVQSS to deal with different phases of supplier-buyer partnerships. Finally, we demonstrate how to apply the GVQSS system through a real-world case.
Keywords:
Variables sampling system
Taguchi capability index
Generalized quick-switch sampling system
Average run length
Web-based app
Journal
IF:
6.5
Papers:
1.0W
Citations:
3.8W