Back to results
US 2026/0153850 A1Published application
Monitoring Stability of Semiconductor Manufacturing Process Systems Using Digital Twins and Neural Networks
AssigneeInspiring Atoms Pte Ltd
InventorYang Pan
Life of this patent
Estimating the term…
Abstract
Disclosed herein is a method for monitoring the stability of process systems in semiconductor manufacturing using digital twins and neural networks. Digital twins of both individual and group process systems are created to identify sources of subsystem parameter changes over time and assess their precise impact on process performance. This method enhances operational efficiency, uniformity, and precision in complex semiconductor manufacturing processes.
Claims20
A method for monitoring stability of a process system, comprising:
- selecting a process system at a predetermined time by a group controller from a group of process systems;
- selecting one or more subsystems from the selected process system for evaluating the stability;
- conducting a measurement routine by a system controller for selected subsystems;
- determining, by the system controller, subsystem-specific parameters for the selected subsystems based on a predetermined algorithm;
- evaluating determined subsystem-specific parameters against their respective trend charts to identify parameters deviating from predefined control limits;
- evaluating the parameters that deviate from the control limits against statistical distributions of the parameters in the group;
- and/or
- generating by the system controller a recipe based on the determined subsystem-specific parameters.
5 more elements
A group of process systems, comprising:
- a group controller for managing operations of the process systems in the group;
- a system controller for controlling a process system in the group, wherein the process system is represented by a process system-specific digital twin, which further includes subsystem-specific digital twins, a reactor digital twin, and a process digital twin that uses outputs from the reactor digital twin as inputs;
- wherein the system controller determines subsystem-specific parameters at predetermined times by running measurement routines;
- and
- wherein the group controller evaluates the subsystem specific parameters against statistical distributions of the parameters within the group and determines if the process system is operating outside of control limits based on a predetermined algorithm.
2 more elements
Dependencies are inferred from the claim text.