4.1 Evaluation criteria analysis
In the weight analysis on evaluation criteria (see Table 1), this study found that, for failure evaluation in lithography machine and lithography processing in semiconductor industry, the primary consideration is the broken wafer complexity (weight = 0.55). At the beginning of a broken wafer, fault interpretation is needed, which would affect the overall wafer production. Thus, the experience of maintenance work relies on in the overall semiconductor lithography processing. The analysis on criteria evaluation also supports this argument.
shows that the failure is due to the machine itself, the evaluation of ME repair and resume capacity should be done next.
Table 1 Weight of evaluation index for criteria
Evaluation index Weight
ME repair and resume capacity 0.34 Broken Wafer complexity 0.55 Possibility of ME scheduled work
resumption 0.09
Other related factors 0.02
The purpose is to ensure that, when the broken wafer complexity can be interpreted as failure in lithography machine, on-site maintenance engineers have enough capability to carry out troubleshooting.
Otherwise, the overall defect rate would increase fast, which would be a very significant loss to semiconductor manufacturing company. Moreover, it is a challenge to the repair and resume capacity of the on-site maintenance engineer on ME. If a timely shutdown and maintenance are not conducted, or fail to identify the failure cause and overcome it in time, resulting in stoppage of the production line, it would be a great loss to semiconductor manufacturing companies.
In the criteria evaluation of other detailed items, this study found that, for a failure evaluation in lithography machine, the impacts from the possibility of ME scheduled work resumption (weight = 0.09) and other related factors are very small (weight = 0.02). This also shows that, in the semiconductor lithography processing, if machine has failure, the primary work is to repair and then resume work, the conditions to be considered are time efficiency and efficiency. As to whether it is necessary to conduct evaluation on the possibility of ME scheduled work resumption, although it should be considered in troubleshooting, it is only a contingent condition. Both defect rate and shutdown would lead to great loss in semiconductor manufacturing companies. Shutdown is not allowed except for necessary reasons or the failure cause is the main critical part.
The weight of other relevant factors in criteria analysis is only 0.02, this also strongly supports that the time efficiency is needed in semiconductor manufacturing company advanced in this study. In other words, unless it is absolutely necessary to consider other factors, semiconductor manufacturing companies only require a fast fault interpretation and repair. This is also consistent with the analysis of evaluation indexes. Therefore, the maintenance engineers should emphasize on quality rather than quantity. However, this study suggested that it is not the best policy in the long run. In the competitive semiconductor processing era that involves high-tech and high costs, although zero defect rate is a desired goal for all companies, it is less attractive than time efficiency. Thus, it is obvious that desire for time efficiency is urgent in the maintenance of lithography machine.
5. Conclusion
5.1 Conclusions
To ensure the stability of lithography machine is the most important issue for current semiconductor processing. Because once the machine malfunctions, it may like to cause stock pile in the lithography processing area, thus unable to reach a balanced production line. If a set of diagnostic system is built, it could
troubleshooting in lithography machine.
Objectively, the enhancement of failure evaluation on lithography machine depends on a good machine system, as well as a good design and configuration of maintenance equipment. Also from the view of practical experience, over-dependency on the experience of existing staff is sufficient at this stage, but there may be restrictions on the company's future development, which may be the possible directions worth considering for breakthrough or improvement. In the weight analysis on evaluation index for criteria, this study found that, the primary consideration for semiconductor industry in lithography process is the broken wafer complexity. In other words, fault interpretation is needed once a broken wafer occurs, this process would affect the overall wafer production. Thus, the experience of maintenance work relies on in the overall semiconductor lithography process. If a timely shutdown and maintenance are not conducted, or fail to identify the failure cause and overcome it in time, resulting in stoppage of the production line, it would be a great loss to semiconductor manufacturing companies.
6. References
[1] Fan, C. M., Guo, R. S., Chen,A.,Hsu,K.C.and Wei,C.S.“DataMining and FaultDiagnosisBased on Wafer Acceptance Test Data and In-Line Manufacturing Data.”,2001 IEEE International Semiconductor Manufacturing Symposium, 171-174 (2001).
[2] Gupta,M.M.and Rao,D.H.“Case-Based Reasoning Applying PastExperienceto New Problems.”, Information Systems Management, 10(2), 77-80 (1994).
[3] ITRS, International Technology Roadmap for Semiconductor: 2001, Yield Enhancement (2002).
[4] Kramer,M.A.“Malfunction DiagnosisUsing Quantitative Models with Non-Boolean Reasoning in Expert Systems.”,AIChE Journal, 33(1), 130-140 (1987).
[5] Leang, S. and Spanos, C. J. “A General Equipment Diagnostic System and Its Application on PhotolithographicSequences.”,IEEE Transactions on Semiconductor Manufacturing, 10(3), 329-343 (1997).
[6] Petti,T.F.,Klein,J.and Dhurjati,P.S.“DiagnosticModelProcessor׃Using Deep Knowledge for Process FaultDiagnosis.”,AIChE Journal, 36(4), 565-575 (1992).
[7] Spanos, C. J., Guo, H. F., Miller, A. and Levine-Parrill,J.“Real-Time Statistical Process Control Using Tool Data.”,IEEE Transactions on Semiconductor Manufacturing, 5(4), 308-318 (1992).
[8] Surma,J.and Braunschweig,B.“Case-Based Retrieval in Process Engineering׃Supporting Design by Reusing Flowsheets.”,Engineering Applications of Artificial Intelligence, 9(4), 385-391 (1996).
A Study of Synchronous and Bucket Trading Behavior of Institutional Investors
P Sh K l,a Ch'h 2,b h 3,e
0- eng 0 Wen- I Lee C eng-Chung Wu lAssociate Professor, Department of Public
Finance and Taxation
National Kaohsiung University of Applied Sciences
2 Professor, Department of Institute Finance, Economics and Business Decision National Kaohsiung University of Applied Sciences
3 Doctoral student, Graduate School of Management National Kaohsiung First University of Science and Technology
a b c
psko@cc.kuas.edu.tw wenchih@cc.kuas.edu.tw wu _ 011 O@yahoo.com.tw
Abstract-The synchronous or contrarian trading behavior was made differently because the stock market's investment strategies for institutional investors who may have the same or different anticipation for the future. The purpose of this study is to explore whether the institutional investors were easy to make the synchronous or contrarian trading behavior for the specific periods. There are important empirical findings for the institutional investors. The results imply the institutional investors trading behavior is not consistent in the different period, especially after the crucial crisis happened, at the ending of the year and in the extreme down period of bull market.
Keywords: Institutional investors, synchronous trading behaviour, contrarian trading
I. INTRODUCTION
For long time, individual investors were the major force in Taiwan's stock market. However, the increase in the number of market players and trading volume during the recent decade prompted the authorities to address existing market deficiencies. A series of measures were introduced to realign the market composition and to create a stable and efficient market. Besides financial market reforms and liberation, restrictions on inbound investments in the local securities market were substantially relaxed to increase Taiwan's attractiveness to international investors. The revamped stock market now sees rising institutional dominance.
Most institutional investors employ a team of analysts to ensure that they have access to adequate market information.
Hence, different institutional investors may end up exhibiting a similar trading behaviour due to similar market expectations or having received similar market information.
Institutional investors may also exhibit herd behaviour, and induce their investors to do the same. In addition, one institutional investor might employ a positive feedback trading strategy, while another party might opt for a negative feedback strategy (contrarian trading). This results in opposing trading behaviours. Information asymmetry is another reason for the exhibition of antithetical trading behaviours. For example, according to Cao and Brennan (1997), local investors would have an advantage over foreign investors in collating local information. This leads to
978-1-61284-486-2/111$26.00 ©2011 IEEE
information asymmetry. As a counter-measure, foreign investors will usually employ positive feedback strategies to capture the private information that they lack. Here is another question: Are synchronous and contrarian trading most apparent in the stock market during specific periods?
Most studies have focused only on the causality of the interaction between institutional investors. Few have explored institutional investors' synchronous or contrarian trading activities as a result of the possible convergence (or divergence) of forecasts or information asymmetry. Hence, this paper aims to explore the synchronous or contrarian stock trading activities of the aforementioned three major categories of institutional investors during specific periods.