Developing an integrated optimization inspection scheme with a flexible sampling mechanism for quality determination based on the process loss index

Abstract

Purpose: This study examines the quick switching sampling (QSS) system. This well-established sampling scheme incorporates two single sampling plans (SSPs) with adaptive transition rules between normal and tightened inspections. The QSS system dynamically adjusts inspection stringency in response to fluctuations in product quality, implementing normal inspection when quality meets satisfactory standards, and tightened inspection when quality deterioration is detected.

Design/methodology/approach: Traditional acceptance sampling plans often evaluate product quality based on process yield, which overlooks subtle variations within specification limits. To address this limitation, a novel performance metric, the process loss index L_e, was developed to quantify quality loss. This index is calculated as the ratio of expected quadratic loss to the square of half the specification width. Utilizing this index, two models of QSS sampling schemes were constructed by solving nonlinear optimization mathematical models and evaluated using general metrics. The efficacy and characteristics of these schemes were investigated, compared, and discussed.

Findings: The results highlight the potential of QSS systems to enhance the effectiveness of quality control while maintaining stringent quality standards. Besides, the proposed plan demonstrates superiority over the conventional plan in terms of adaptability, particularly with sample size adjustments, when switching to a stricter inspection plan in response to deteriorating lot quality and improved efficiency.

Originality/value: This study presents a novel approach to quality control by integrating the process loss index into the QSS system, offering a fresh perspective on sampling methodologies. The integration of QSS with the process loss index L_e marks a significant contribution to the field of quality control, enabling more nuanced evaluations of product quality and providing a groundbreaking framework for optimizing quality control processes while minimizing sample sizes, thereby enhancing efficiency and effectiveness.