Semiconductor Demand Data Mining and Knowledge Discovery for Optimization of Capacity Allocatio
Date Issued
2005
Date
2005
Author(s)
DOI
932213E002005
Abstract
The demand signal is the most unreliable source of
information that plagues the operation effectiveness in a
demand-supply network. Moreover, the demand uncertainty
is not only propagated but also magnified over the network
and causes a chain effect on the planning quality of the entire
supply chain. Semiconductor manufacturing network is one
of the most complicated demand-supply networks and thus
suffers greatly from the untrustworthy demand information.
In the second year of this project, the effect on the overall
equipment effectiveness (OEE) has been explored. The
effects of demand grouping for equipment capacity allocation
are then modeled mathematically. The model is aimed to help
practitioners comprehend how demand plans work together
with capacity allocation to affect the OEE. The third year of
this project will focus on optimization of capacity allocation.
The goal is to find optimum capacity allocation for demand
groups to minimize the required equipment capacity or
equivalently maximize the OEE subject to uncertain demand
signals. Various combinatorial optimization algorithms, such
as Greedy Algorithm, Genetic Algorithms, etc., will be
investigated. Effective optimization methodologies are then
suggested and tested using actual semiconductor demand and
manufacturing data.
Subjects
Demand Planning
Capacity Allocation
Overall
Equipment Effectiveness (OEE)
Equipment Effectiveness (OEE)
Publisher
臺北市：國立臺灣大學工業工程學研究所
Type
report
File(s)![Thumbnail Image]()
Loading...
Name
932213E002005.pdf
Size
419.41 KB
Format
Adobe PDF
Checksum
(MD5):a87d77489da58fb52ad74f13af4cff1b