Introducing pull methodologies in a semiconductor fab
Name
63183266-MIT.pdf
Description
Full printable version
Size
9.42 MB
Format
Adobe PDF
Checksum (MD5)
e801540d763a9633f0d6b75ef10b9689
Author(s)
Connally, Jason Walker
Advisor(s)
Stephen Graves and Daniel Whitney.
Date Issued
2005
Publisher
Massachusetts Institute of Technology
Abstract
Semiconductor manufacturing is a highly complex and re-entrant process. In a fabrication facility, hundreds of decisions are made during a production shift regarding how shared tool capacity will be prioritized. These decisions contribute to how balanced or unbalanced the manufacturing line will be. Characteristics of an unbalanced line are large WIP bubbles, long queue times, and expediting. A balanced line has less WIP bubbles, shorter queues, and WIP is positioned at all points throughout the line to be in position to meet the demand forecasted. This thesis focuses on work performed at Intel's Fab 23 to improve the process for assigning production priorities through the introduction of pull methodologies. Existing processes and tools were studied, and an improved methodology and decision-support tool was developed to aid operations managers in driving towards a balanced line. An experiment was also designed and executed in production to test the methods and tools developed. Target cycle time data was used along with throughput goals to construct a target inventory profile throughout the line. Actual inventory was then compared to the ideal "balanced" profile to determine where WIP deficits and surpluses existed.
(cont.) Using this information, the operations managers had objective metrics that could be used in determining which operations should receive priority. Significant externalities inhibited performance during the experiment, preventing measurable improvements in line balance and cycle time from being realized. However, these externalities were known prior to experimentation, and a decision was made to learn from the experiment.The tool proved helpful in promoting consistency across shifts in how the factory was run. There were many anecdotal examples of the decision-support tool driving more intelligent priority decisions than operations managers would have made without the tool.
Description
Thesis (M.B.A.)--Massachusetts Institute of Technology, Sloan School of Management; and, (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering; in conjunction with the Leaders for Manufacturing Program at MIT, 2005.
Includes bibliographical references (p. 85-86).
Subjects
Sloan School of Management.
Mechanical Engineering.
Leaders for Manufacturing Program.
MIT Department
Leaders for Manufacturing Program at MIT
Massachusetts Institute of Technology. Department of Mechanical Engineering
Sloan School of Management
Terms of Use
M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.
Persistent DSpace Link
