Effectiveness of the Design of Experiments (DoE) on variation reduction : empirical evidence for the automotive component-manufacturing sector in South Africa

Abstract

In today’s modern manufacturing environment, any process variability is an attack on quality and throughput. Variability reduction is paramount and necessary if higher levels of quality are to be obtained. Hence, this study evaluates the effectiveness of the Design of Experiments (DoE) as a strategic tool for reducing variation in the automotive component manufacturers in South Africa. As with the evolution of the manufacturing process, methods to identify and eliminate process variability are developed. Most of these methods are based on statistical DoE. A DoE is a test or series of tests that enables the experimenter optimise the yield of a process or minimise variability. Consequently, this study focuses on the effectiveness of DoE for variability reduction in the automotive sector in South Africa. Of the 193 individuals identified for participation, 164 completed the questionnaires. Middle-level Managers from four automotive component manufacturing companies in the eThekwini District Municipality participated in the study. The study investigated production and the related experiences of the automotive component manufacturing companies that have adopted a DoE strategy. Descriptive and correlation were used to analyse data. The results indicates that DoE reduces product variation in the automotive component manufacturers in South Africa. In order to maximise performance, a comprehensive variability reduction policy must be developed, which aligns DoE tools to business performance. DoE has the ability to screen a large number of variables to find important ones during product reformulation process in the Product and Development functional areas.