August_EDFA_Digital

edfas.org 55 ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 20 NO. 3 AI-ASSISTED HOT SPOT IDENTIFICATION Identifying layout hotspots during the design stage is an important yet challenging design-for-manufacture (DFM) process. Due to its success in image-based clas- sification applications, machine learning based on deep neural networks [3] promises an alternative that is faster and more accurate than rigorous model-based simula- tion. [4] Moreover, revealing the correlations between scan diagnostic results and DFM is an emerging field, driven by the increasedusageof diagnosis combinedwith theneces- sity of identifying systematic critical features. In a recent Advanced Semiconductor Manufacturing Conference paper, [5] a novel flow to correlate diagnostic results and DFM ruleswas proposed. PFA engineers canuse these new technologies not only to identify systematic defects, but alsodiƒerentiate if a systematic defect is causedby design issues or manufacturing problems. SUMMARY In summary, PFA engineers should keep upwith recent advancements in diagnosis and yield learning EDA tools, especially those based on AI and machine leaning tech- nologies. These advancements can help them interpret diagnosis reports, such that picking the right die for PFA leads to a higher success rate; prioritize die for PFA based on their likelihood of revealing systematic defects; and diƒerentiate the root cause between a design or a manu- facturing issue to ensure an eƒicient and eƒective process to solve the yield problem. REFERENCES 1. Y. Huang, B. Benware, R. Klingenberg, H. Tang, J. Dsouza, and W.-T. Cheng: “Scan Chain Diagnosis Based on Unsupervised Machine Learning,” Proc. Asian Test Symposium (ATS), 2017, pp. 220-225. 2. Y. Huang, W. Yang, andW. Cheng: “Advancements inDiagnosis Driven YieldAnalysis (DDYA): ASurvey of State-of-the-Art ScanDiagnosis and Yield Analysis Technologies,” Proc. IEEE European Test Symposium, 2015, pp. 1-10. 3. I. Goodfellow, Y. Bengio, and A. Courville: Deep Learning, The MIT Press, 2016, ISBN: 0262035618. 4. T. Matsunawa, S. Nojima, and T. Kotani: “Automatic Layout Feature Extraction for Lithography Hotspot Detection Based on DeepNeural Network,” SPIE Advanced Lithography, 2016, 9781. 5. C. Schuermyer, S. Malik, and T. Herrmann: “Identifying Systematic Critical Features Using Silicon Diagnosis Data,” Advanced Semi- conductor Manufacturing Conference (ASMC), 2012, pp. 1-6. ABOUT THE AUTHOR Yu Huang is a principal engineer in the Silicon Test Systems Division of Mentor, a Siemens Business. His research interests include VLSI SoC testing, ATPG, compression, and diagnosis. He has a B.S in electronic science and an M.S. in semiconduc- tor devices and technology, both from Nankai University, China, and a Ph.D. in electrical and computer engineering from the University of Iowa. He holds 27 U.S. patents with an additional nine pending. Huang has published more than 100 papers in leading IEEE journals, conferences, andworkshops. He is a senior member of IEEE and served as a technical program com- mittee member for DAC, ITC, ATS, ETS, ASPDAC, NATW, as well as other testing conferences and workshops. Whether networking at events or accessing information through EDFA , ISTFA proceedings, or journals, our members have the edge. Now it’s time to introduce EDFAS to others in the industrywhowould like to take advantage of these career- enhancing benefits. Help us help the industry by expanding our membership and oƒering others the same exceptional access to information and networking that sets EDFAS apart. To reacquaint yourself with and introduce others to the EDFAS member benefits, visit asminternational.org/web/edfas/membership. EDFAS MEMBERSHIP

RkJQdWJsaXNoZXIy MjA4MTAy