Towards smart semiconductor fabrication, the automated process recipe determination with AI is required under scarce data (NOT big data) conditions. This paper proposes precise recipe determination by ensemble learnin...
详细信息
ISBN:
(数字)9798350391633
ISBN:
(纸本)9798350391640
Towards smart semiconductor fabrication, the automated process recipe determination with AI is required under scarce data (NOT big data) conditions. This paper proposes precise recipe determination by ensemble learning combining backcasting & forecasting AIs. Proposed backcasting AI predicts etching recipe to optimize the equipment arm motion, traditionally set manually, to achieve the desired etching amount. Automatically estimated etching recipe is optimized in real time and both throughput and yield improve. By means of incorporating additional features such as the differential and variance of etching results and employing ensemble learning with multiple neural networks, RMSE (Root Mean Squared Error) of the objective variable representing the periodic motion of the arm reduces by more than 40%. In addition, forecasting AI that predicts etching results from a recipe made by backcasting AI automatically validates the recipe.
This paper proposes machine learning-based etching amount prediction from arm scanning movements for single wafer wet etching to optimize etching recipe. Compared with the conventional prediction method of fixed etchi...
This paper proposes machine learning-based etching amount prediction from arm scanning movements for single wafer wet etching to optimize etching recipe. Compared with the conventional prediction method of fixed etching rate, the proposed machine learning methods achieve 0.23 lower RMSE in predicting wet etching amount.
In this paper, we propose a technique that uses thermal measurement results for improved accuracy in thermal simulation of electronic apparatus. Because the modeling of the electronic components in such apparatus has ...
详细信息
作者：
Baker, CKrull, RSnyder, GLincoln, WMalone, TBClifford C. Baker
CIE CHFEP is a senior staff scientist at Carlow International Incorporated. He has applied most of his 24 years of experience in the application of human engineering technology to maritime systems. Mr. Baker has directed much of Carlow's efforts to reduce ship workload and to improve human performance and maritime safety through application of human factors methods and data. He is a Certified Industrial Ergonomist (CIE) as well as a Certified Human Factors Engineering Professional (CHFEP). Both certifications were granted by Oxford Research where Mr. Baker also serves as an Advisory Board member. Russell D. Krull
P.E. is a senior engineer with A&T/Proteus Engineering with more than 18 years of experi-ence in marine engineering naval architecture and program management including 16 years of active duty in the U.S. Coast Guard. Recent experience includes advanced ship design studies engineering software development technical support for the USMC Advanced Amphibious Assault Vehicle propulsion systems analyses ship structural engineering and cargo handling systems engineering. Mr. Krull has an M.S.E. in naval architecture and marine engineering and an M.S.E. in industrial and operations engineering from University of Michigan and a B.S. in ocean engineering from the U.S. Coast Guard Academy. Capt. Glenn L. Snyder
USCG. Regrettably since this paper was originally written Capt. Snyder has passed away. At the time of his death he was an operations specialist assigned to the Coast Guard's Deepwater Capabilities Replacement Project as Chief of Human Systems Integration. He served as commanding officer of the patrol boat Cape George (WPB-95306) the icebreaking tug Biscayne Bay (WTGB-104) and the cutter Legare (WMEC-911). A 1975 graduate of the U.S. Coast Guard Academy Capt. Snyder held an M.A. in national security and strategic studies from the U.S. Naval War College and an M.A. in international relations from Salve Regina College. In addition he was a 1998 fellow of the Foreign Service
The U.S. Coast Guard is in the concept exploration phase of its Integrated Deepwater System (IDS) acquisition project. This project will define the next generation of surface, air and command, control, communications,...
详细信息
The U.S. Coast Guard is in the concept exploration phase of its Integrated Deepwater System (IDS) acquisition project. This project will define the next generation of surface, air and command, control, communications, computers, intelligence, surveillance, and reconnaissance (C4ISR) assets used to perform the Coast Guard's missions in the IDS environment (>50 NM off the U. S. coastline). As part of early technology investigations, the needs exist to: (1) analyze the workload requirements of the IDS, (2) identify alternative means to perform ship's work, and (3) optimize ship manning consistent with ship workload, performance criteria, and the available tools and equipment aboard. To reduce shipboard work requires an understanding of the mission and support requirements placed on the vessel and crew, how these requirements are currently met, and how requirements might otherwise be met to reduce workload and crew size. This study examined currently implemented workload and manpower reducing approaches of commercial maritime fleets, U.S. and foreign navies, and foreign coastguards. These approaches were analyzed according to evaluation criteria approved by the IDS acquisition project team. From this, strategies for shipboard work reduction that may be considered for adoption by the IDS were identified and analyzed according to performance and costs factors. Ten workload-reducing strategies were identified: damage control, bridge, multiple crewing, engineering, risk acceptance, modularity, deck, enabling technologies, ship/personnel readiness, and operability and maintainability.
作者：
SUTO, YHAYASHI, YMedical Equipment Division
Toshiba Corporation Kawasaki 213 Japanwas born in Japan in July 193 He received M.S. and Ph.D. degrees in administration engineering from Keio University
Yokohama Japan in 197 and 1977
respectively. From 196 to 1978
he was with the Electronic Computer Division Toshiba Corporation Tokyo Japan engaged in the development of computer application software. He is currently with the Medical Equipment Division Toshiba Corporation and is concerned with nuclear medical data processing. Dr. Suto is a member of the Japan Society of Medical Electronics and Biological Engineering the Japan Society of Ergonomics and the Information Processing Society of Japan.YoshioFaculty of Engineering Keio University Yokohama 223 Japan was born in Tokyo
Japan in September 192 He received M.D. and Ph.D. degrees from Keio University
Tokyo Japan in 1960 and 1967 respectively. From 195 to 1965
he was associated with the school of Medicine Keio University. He is currently a Professor in the Department of Administration Engineering Keio University. He has been concerned with human engineering system engineering and bioengineering and is presently engaged in research in the areas of pattern recognition. Dr. Hayashi is a member of the Japan Society of Ergonomics the Operations Research Society of Japan and the Society of Instrument and Control Engineers of Japan.
A pattern recognition system is divided into two parts, a receptor and a categorizer. In this research, a creative receptor is designed, and then an effective pattern recognition system based on the receptor is consti...
详细信息
A pattern recognition system is divided into two parts, a receptor and a categorizer. In this research, a creative receptor is designed, and then an effective pattern recognition system based on the receptor is constituted. The receptor is composed of a set of unique matrix operators which allow it to perform efficient feature measurement extraction. The feature measurements are ranked in effective order, based on another function of the receptor. For the categorizer, a well-known optimization theory is applied. To evaluate the performance of this system, some evaluation functions are defined. Therefore, the receptor mechanism is optimized by the functions. This system is experimentally evaluated by computer simulations.
暂无评论