18153573. SIMULATION SYSTEM FOR SEMICONDUCTOR PROCESS AND SIMULATION METHOD THEREOF simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
SIMULATION SYSTEM FOR SEMICONDUCTOR PROCESS AND SIMULATION METHOD THEREOF
Organization Name
Inventor(s)
Sanghoon Myung of Goyang-si (KR)
Hyunjae Jang of Hwaseong-si (KR)
Hyeon Kyun Noh of Gwangmyeong-si (KR)
Min-Chul Park of Hwaseong-si (KR)
Changwook Jeong of Hwaseong-si (KR)
SIMULATION SYSTEM FOR SEMICONDUCTOR PROCESS AND SIMULATION METHOD THEREOF - A simplified explanation of the abstract
This abstract first appeared for US patent application 18153573 titled 'SIMULATION SYSTEM FOR SEMICONDUCTOR PROCESS AND SIMULATION METHOD THEREOF
Simplified Explanation
The abstract describes a simulation method using a recurrent neural network (RNN) in a process simulator for semiconductor manufacturing. The RNN consists of process emulation cells that train and predict the profiles of each process step based on a final target profile.
- The simulation method uses a recurrent neural network (RNN) in a process simulator for semiconductor manufacturing.
- The RNN includes process emulation cells arranged in a time series.
- The RNN is trained to predict the profiles of each process step based on a final target profile.
- The method involves receiving a previous output profile, a target profile, and process condition information at a process emulation cell.
- The process emulation cell generates a current output profile for the current process step based on the received information and prior knowledge information.
- The prior knowledge information defines the causal relationship between the previous process step and the current process step.
Potential Applications
- Semiconductor manufacturing process optimization
- Predictive maintenance in semiconductor manufacturing
- Quality control and defect detection in semiconductor manufacturing
Problems Solved
- Lack of accurate and efficient simulation methods for semiconductor manufacturing processes
- Difficulty in predicting process profiles and optimizing process steps
- Inability to effectively utilize prior knowledge and causal relationships in process simulations
Benefits
- Improved accuracy and efficiency in simulating semiconductor manufacturing processes
- Enhanced prediction and optimization of process profiles
- Utilization of prior knowledge and causal relationships for better process simulations
Original Abstract Submitted
Provided is a simulation method performed by a process simulator, implemented with a recurrent neural network (RNN) including a plurality of process emulation cells, which are arranged in time series and configured to train and predict, based on a final target profile, a profile of each process step included in a semiconductor manufacturing process. The simulation method includes: receiving, at a first process emulation cell, a previous output profile provided at a previous process step, a target profile and process condition information of a current process step; and generating, at the first process emulation cell, a current output profile corresponding to the current process step, based on the target profile, the process condition information, and prior knowledge information, the prior knowledge information defining a time series causal relationship between the previous process step and the current process step.
