TY - JOUR
T1 - The multi-source fabless-foundry partnership selection model
AU - Shuai, Jia Jane
AU - Tzeng, Gwo Hshiung
AU - Li, Han-Lin
PY - 2004/6/11
Y1 - 2004/6/11
N2 - Foundry selection is an important decision to a semiconductor fabless (FABricationLESS) design firm. This partnership is a long-term commitment and the success of a fabless is directly affected by the decision. This paper presents a multiple criteria evaluation model for international multiple foundry partnership selection. In a multiple source foundry environment, the fabless maintains partnership with several different foundries to diversify the risk of capacity shortage and natural disaster. We outline the potential planning problems for foundry partnership selection and analyse issues related to cost, yield and time to market. A hierarchical multiple criteria decision-making method for selecting the foundry partners is proposed. Based on a hierarchical structure, non-additive and additive fuzzy measurement is used to show the relative importance of criteria. To select a group of foundry partners, management culture, technology level and future growth are heavily interrelated, the fuzzy integral is, therefore, used to evaluate the synthetic performance of the best partners. To show the practicability and usefulness of this method, a Taiwan consumer integrated circuit (IC) fabless case was presented for evaluation. Results in this study not only confirm the practical nature of the proposed model, but also demonstrate the high flexibility in evaluation of a multiple criteria decision making (MCDM) problem.
AB - Foundry selection is an important decision to a semiconductor fabless (FABricationLESS) design firm. This partnership is a long-term commitment and the success of a fabless is directly affected by the decision. This paper presents a multiple criteria evaluation model for international multiple foundry partnership selection. In a multiple source foundry environment, the fabless maintains partnership with several different foundries to diversify the risk of capacity shortage and natural disaster. We outline the potential planning problems for foundry partnership selection and analyse issues related to cost, yield and time to market. A hierarchical multiple criteria decision-making method for selecting the foundry partners is proposed. Based on a hierarchical structure, non-additive and additive fuzzy measurement is used to show the relative importance of criteria. To select a group of foundry partners, management culture, technology level and future growth are heavily interrelated, the fuzzy integral is, therefore, used to evaluate the synthetic performance of the best partners. To show the practicability and usefulness of this method, a Taiwan consumer integrated circuit (IC) fabless case was presented for evaluation. Results in this study not only confirm the practical nature of the proposed model, but also demonstrate the high flexibility in evaluation of a multiple criteria decision making (MCDM) problem.
KW - Analytic hierarchy process
KW - Fabless-foundry partnership
KW - Fuzzy integral
KW - Fuzzy measure
UR - http://www.scopus.com/inward/record.url?scp=2542534578&partnerID=8YFLogxK
U2 - 10.1504/IJMTM.2004.004512
DO - 10.1504/IJMTM.2004.004512
M3 - Review article
AN - SCOPUS:2542534578
SN - 1368-2148
VL - 6
SP - 137
EP - 154
JO - International Journal of Manufacturing Technology and Management
JF - International Journal of Manufacturing Technology and Management
IS - 1-2
ER -