Abstract

Faced with a disappointing process yield of 86.5%, Asahi India Glass Co. applied the define, measure, analyze, improve and control (DMAIC) method to find and resolve problems in its automobile glass manufacturing process. Asahi found its furnace did not work effectively to bend and temper glass for multiple applications. The manufacturer engaged in defining and measuring the problem, revealing some parameters not in control and showing poor output during final inspection. An input-output (I-O) sheet revealed abnormalities and showed 23 potential causes of defects, which were validated in the analyze phase. To reduce bending and tempering defects, statistical software was used to optimize speed and temperature settings, resulting in a 9.8% in process yield improvement. Control charts demonstrated that the controls were effective and sustainable. By using the DMAIC method to optimize input factors, Asahi achieved and sustains process yield of 96.3% and realizes about $230,000 in annual recurring savings. The Six Sigma approach to process improvement through statistical tools was a new and winning strategy for Asahi and can be applied in other manufacturing situations needing to improve process yield.