µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
ADAPÀÇ ¹ÝµµÃ¼ Á¦Á¶°øÁ¤ ¹°Áú ¸ð´ÏÅ͸µ ¹× ¿¹Ãø Àû¿ë »ç·Ê |
¿µ¹®Á¦¸ñ(English Title) |
An Application of AI & Big Data-based Predictive Maintenance Solution IDAP/ADAP for Material Monitoring and Prediction in Semiconductor Manufacturing Process |
ÀúÀÚ(Author) |
Á¶Çü¹Î
½ÉÀçö
±èÀ缺
ÀÌ°æÈñ
¹Ú¿¬¼ø
HyungMin Cho
Peter Shim
JaeSung Kim
Anthony Park
KyungHee Lee
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 37 NO. 01 PP. 0046 ~ 0057 (2021. 04) |
Çѱ۳»¿ë (Korean Abstract) |
¹ÝµµÃ¼ Á¦Á¶ °øÁ¤¿¡´Â ´Ù¾çÇÑ ÈÇй°ÁúÀÌ »ç¿ëµÇ¸ç, °øÁ¤ ³»¿¡¼ ¹°ÁúÀÇ Á¤¹ÐÇÑ Á¦¾î¸¦ ÅëÇØ Ç¥¸é ó¸®ÀÇ ±ÕÁú¼º°ú Ç°Áú°ü¸®°¡ ÀÌ·ç¾îÁø´Ù. ¹ÝµµÃ¼ Á¦Á¶¿¡¼ °¡Àå Å« °ü½É»ç´Â °¢ °øÁ¤ÀÇ ¹Ýº¹¼º°ú ÀçÇö¼ºÀ̸ç, Á¤ÇØÁø »ç¾ç¿¡¼ ¾à°£ÀÇ ÆíÂ÷¿¡µµ °ª ºñ½Ñ Àåºñ ¿À¿°°ú ¿þÀÌÆÛ ½ºÅ©·¦ÀÌ ¹ß»ýÇÒ ¼ö ÀÖ´Ù. º» ¿¬±¸¿¡¼´Â ½Ç½Ã°£ ºòµ¥ÀÌÅÍ ¿¹Ãø À¯Áöº¸¼ö ¼Ö·ç¼ÇÀÎ IDAP/ADAP(IoT Data Analytics Platform /AI Bigdata Platform)À» ¹ÝµµÃ¼ Á¦Á¶°øÀåÀÇ ¿©·¯ ÁöÁ¡¿¡¼ ÃøÁ¤µÈ ¹°ÁúÀÇ »óŸ¦ ÅëÇÕ °ü¸®Çϴµ¥ Àû¿ëÇÑ »ç·Ê´Ù. »óÅ µ¥ÀÌÅʹ Ư¼º¿¡ µû¶ó ¹èÄ¡ ȤÀº ½Ç½Ã°£À¸·Î ¼öÁýÇϸç, ¼³Á¤µÈ »óÇÑ ¹× ÇÏÇÑ°ª ÃÊ°ú½Ã °ü¸®ÀÚ¿¡°Ô ¾Ë¶÷ ¸Þ½ÃÁö¸¦ Àü´ÞÇÑ´Ù. ¶ÇÇÑ, ºòµ¥ÀÌÅÍ¿Í ÀΰøÁö´É ±â¼úÀ» È°¿ëÇÏ¿© ÃàÀûµÈ »óÅ µ¥ÀÌÅ͸¦ ºÐ¼®ÇÏ°í, ¿¹Ãø¸ðµ¨À» ¸¸µé¾î »óŸ¦ ¹Ì¸® ¿¹ÃøÇÏ¿© ¿¹ÁöÁ¤ºñ°¡ °¡´ÉÇϵµ·Ï ÇÑ´Ù. AI ¿¹Ãø ¸ðµ¨À» »ý¼ºÇÒ ¶§ °¢ ÁöÁ¡ÀÇ °íÀ¯ÇÑ Æ¯¼ºÀ» ¹Ý¿µÇØ¾ß ÇϹǷΠ½Ã°£°ú ³ë·ÂÀÌ ¸¹ÀÌ ¼Ò¿äµÇ´Â ÀÛ¾÷ÀÌ´Ù. º» ³í¹®¿¡¼´Â ÃàÀûµÈ µ¥ÀÌÅͷκÎÅÍ Á¤È®µµ°¡ ³ôÀº »óÀ§ N°³ ÁöÁ¡À» ¼±Á¤ÇÏ¿© ¸ÕÀú ¸ð´ÏÅ͸µÇÏ°í ¿¹ÃøÇÏ°í, Á¤È®µµ°¡ ³·Àº ³ª¸ÓÁö ÁöÁ¡¿¡ ´ëÇؼ´Â Ãß°¡ µ¥ÀÌÅ͸¦ È®º¸ÇÏ¿© Á¡ÁøÀûÀ¸·Î Á¤È®µµ¸¦ ³ôÀÌ´Â ¹æ½ÄÀ» Á¦¾ÈÇÑ´Ù. º» ¿¬±¸´Â °³º°ÁöÁ¡ÀÇ ¿¹Ãø Á¤È®µµ¸¦ ³ôÀÌ´Â °Íº¸´Ù Àü»çÀû ¼öÁØ¿¡¼ ¹°Áú»óŸ¦ ¸ð´ÏÅ͸µÇÏ°í ¿¹ÃøÇÏ´Â ¹æ¹ý·Ð°ú Ç÷§ÆûÀ» Á¦½ÃÇϴµ¥ ¸ñÇ¥°¡ ÀÖ´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Various chemical substances are used in the semiconductor manufacturing, and homogeneity of surface treatment and quality control are achieved through precise control of substances within the process. One of the biggest concerns in semiconductor manufacturing is the repeatability and reproducibility of each process, and even a slight deviation from a given specification can lead to expensive equipment contamination or wafer scrap. In this study, we introduce a case of applying IDAP/ADAP (IoT Data Analytics Platform/AI Bigdata Platform), a real-time big data predictive maintenance solution, to integrated management of state information of substances measured at various points in a semiconductor factory. Status data is collected in batches or in real time according to characteristics, and an alarm message is delivered to the manager when the set upper and lower limits are exceeded. In addition, the accumulated state data is analyzed using big data and artificial intelligence technology. When creating an AI predictive model, it is a time-consuming and labor-intensive task as it must reflect the unique characteristics of each point. This study aims to present a methodology and platform for monitoring and predicting material conditions at the enterprise level rather than improving the prediction accuracy of individual points. |
Å°¿öµå(Keyword) |
ºòµ¥ÀÌÅͱâ¹Ý ¿¹Ãø
AI
½º¸¶Æ®Á¦Á¶
¹ÝµµÃ¼ Á¦Á¶°øÁ¤
Bigdata-based prediction
AI
Smart factory
Semi-conduct manufacturing
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|