Improving the OEE and UPH data quality by Automated Data Collection for the semiconductor assembly industry

Research highlights? We develop a real time communication system and MES into an ADC system for collecting useful data in the semiconductor assembly industry. ? The system has eliminated the unknown OEE losses from original 6% of manual recording system to near zero percent. ? The UPH data can be collected automatically once a production lot is finished. Most semiconductor companies follow the SEMI E10 and E79 guidelines to measure equipment availability and efficiency via OEE (Overall Equipment Effectiveness). However, several problems need to be conquered before implementing OEE. For example, the time intervals of OEE losses are critical to the improvement studies, but it is very hard to collect reliable and accurate data. Therefore, people will find the unknown differences between the recorded losses from their operational system and OEE losses from SEMI E79 definition once people implement OEE. Furthermore, how to obtain the theoretical unit throughput or standard UPH (unit-per-hour machine rates) to determine the average processing rates of an equipment is another issue to be conquered.In this study, we develop an IT integrated system to record the time intervals of OEE losses for the bottleneck equipment "wire bonder" in the semiconductor assembly industry. We integrate a communication system (that is developed by the wire bonder supplier) and MES (Manufacturing Execution System) into an ADC (Automated Data Collection) system for collecting useful data. The data quality is further guaranteed by the real time detection of equipment status from the communication system. The application of the ADC system has eliminated the unknown OEE losses from original 6% of manual recording system to near zero percent. Therefore, the UPH data can be collected automatically once a production lot (partition of a specific batch in an order) is finished. This resolves the accuracy and timeliness problem associated with traditional stopwatch sampling.

[1]  S. Divorski,et al.  Improving data quality for performance measures: results from a GAO study of verification and validation☆ , 2001 .

[2]  Graeme Simsion,et al.  Justifying Investment in Information Resource Management , 1995, Australas. J. Inf. Syst..

[3]  Terrence D. Perera,et al.  Automated data collection for simulation? , 2002, Simul. Pract. Theory.

[4]  Robert J. Kauffman,et al.  Reuse and Productivity in Integrated Computer-Aided Software Engineering: An Empirical Study , 1991, MIS Q..

[5]  Graeme C. Simsion.  Data modeling essentials - analysis, design, and innovation , 1993, VNR computer library.

[6]  D. G. Baker,et al.  An Automated System for monitoring the Quality and Quantity of Subsurface Drain Flow , 1999 .

[7]  Srinivasan Raghunathan,et al.  Impact of information quality and decision-maker quality on decision quality: a theoretical model and simulation analysis , 1999, Decis. Support Syst..

[8]  Ronie Navon,et al.  Automated project performance control of construction projects , 2005 .

[9]  Robert C. Leachman.  Closed-loop measurement of equipment efficiency and equipment capacity , 1995, Proceedings of SEMI Advanced Semiconductor Manufacturing Conference and Workshop.

[10]  Graeme Simsion,et al.  Data Modeling Essentials 2nd Edition: A Comprehensive Guide to Data Analysis, Design, and Innovation , 2000 .

[11]  Ronie Navon,et al.  Monitoring labor inputs: automated-data-collection model and enabling technologies , 2003 .

[12]  C. F. Bryan,et al.  Annual Report for 2004 , 2005 .

[13]  Ronie Navon,et al.  Automated inspection and restoration of steel bridges—a critical review of methods and enabling technologies , 2002 .

[14]  Albert W. L. Yao,et al.  Developing a PC-based automated monitoring and control platform for electric power systems , 2003 .