By Topic

Historical data modeling determines future collection criteria

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Gwin, J. ; Sony Semiconductor of America, USA ; Rodriguez, R. ; Colbert, C. ; Olvera, J.

Sony Semiconductor of America's Bipolar Division is achieving increased capacity, reduced cycle time, improved yields and reduced costs by systematically combining straightforward statistical tools and specialized expert knowledge with typical Statistical Process Control (SPC) methods. This project completed its prototype phase in 1996 and is now being applied during 1997. Preliminary results show some processes/areas will realize up to 87% reduction in data collection on critical metrology tools and a 33% reduction in monitor materials. The Bipolar Division's manufacturing line, known as Fab 2A, makes multiple products for Sony and several foundry customers. Inconsistent application and interpretation of statistical process control (both across and within functional areas) developed over time. Recent conversion from manual paper SPC charts to the PROMIS electronic data collection system provides the opportunity to analyze and improve SPC methodology. The need to implement a more robust process control philosophy, consisting of appropriate data collection, chart types and response mechanisms gave birth to the SPC improvement effort. This endeavor's foundation is statistical modeling of historical process data to determine pertinent factors and factor interactions. Subject-matter experts examine current data to determine whether the necessary data is being collected; results indicate if more, less or different data is required to adequately model the process. Once a good model is attained, similar data for kindred processes is examined to determine future collection criteria for that “realm”

Published in:

Advanced Semiconductor Manufacturing Conference and Workshop, 1997. IEEE/SEMI

Date of Conference:

10-12 Sep 1997