Site News  |   Monitor Keywords  |   Monitor Archive  |   Organizer  |   Account Info  |

Methods and systems for controlling a semiconductor fabrication process

Methods and systems for controlling a semiconductor fabrication process description/claims

This claims the benefit of U.S. App. No. 60/746,163 filed on May 1, 2006 and U.S. App. No. 60/807,189 filed on Jul. 12, 2006. This application is a continuation-in-part of U.S. application Ser. No. 10/985,834, filed on Nov. 10, 2004, which claims the benefit of U.S. App. No. 60/518,823 filed on Nov. 10, 2003 and U.S. App. No. 60/607,649 filed on Sep. 7, 2004. This application is also a continuation-in-part of U.S. application Ser. No. 11/123,966 filed on May 6, 2005, and a continuation-in-part of U.S. application Ser. No. 11/302,563 filed on Dec. 13, 2005.

Each of the foregoing commonly-owned applications is incorporated by reference herein in its entirety.

1. Field

This invention relates to, inter alia, methods of utilizing a wafer-centric database to improve system throughput.

2. Related Art

The handling of workpieces such as wafers within a semiconductor manufacturing environment can present significant computing challenges. Hardware such as process modules, handlers, valves, robots, and other equipment are commonly assembled from a variety of different manufacturers each of which may provide proprietary or pre-compiled software unsuitable for a newly conceived process. In addition, fabrication-wide software typically compiles relevant data as simple, chronological logs of output from sensors, process modules, controllers, and the like, so that finding information for handler or wafer-specific processing requires an initial search of all of the potentially relevant log files for data, followed by processing the search results into a form suitable for process control such as scheduling decisions.

There remains a need for improved software suitable for real-time control of semiconductor manufacturing processes.

Software for controlling processes in a heterogeneous semiconductor manufacturing environment may include a wafer-centric database, a real-time scheduler using a neural network, and a graphical user interface displaying simulated operation of the system. These features may be employed alone or in combination to offer improved usability and computational efficiency for real time control and monitoring of a semiconductor manufacturing process. More generally, these techniques may be usefully employed in a variety of real time control systems, particularly systems requiring complex scheduling decisions or heterogeneous systems constructed of hardware from numerous independent vendors.

In one aspect, a user interface disclosed herein includes a display of a three-dimensional simulation of a semiconductor workpiece handling system that includes a hardware item; and a link to information related to the hardware item, wherein the link may be substantially contained within an area of the display where the hardware item resides, and wherein the information may include at least a status of the hardware item and technical information for the hardware item.

The three-dimensional simulation may be a real-time simulation based upon operation of a physical semiconductor workpiece handling system. The semiconductor workpiece handling system may include a plurality of hardware items, each one of the plurality of hardware items may have a link to information associated therewith. The technical information may include a list of replacement parts for the hardware item. The technical information may include a manual for the hardware item. The technical information may include a maintenance log for the hardware item. The link may be activated by a mouse over of the link. The link may be activated by a mouse click of the link. The information may be displayed in a new window upon activation of the link. The information may be displayed in a pop-up window upon activation of the link. The status information may include sensor data received from the hardware item. The status information may include diagnostic information for the hardware item. The diagnostic information may include one or more of a performance evaluation, an expected time to failure, a maintenance alert, and an operating condition alert. The hardware item may include one or more of a robotic arm, an end effector, an isolation valve, a heating station, a cooling station, a load lock, a vacuum pump, a robot drive, a metrology device, a sensor, a process module, and a device within a process module. The hardware item may include a workpiece. The workpiece may include a semiconductor wafer. The status information may include a particle map. The status information may include an estimated temperature. The status information may include a wafer center location. The status information may include substantially real time data for the hardware item. The display may include a tool for user selection of a perspective for viewing the three-dimensional simulation.

In one aspect, a system disclosed herein includes a state machine that controls operation of a semiconductor manufacturing system that may schedule processing of one or more workpieces, the state machine may include a plurality of states associated by a plurality of transitions, each one of the plurality of transitions may have a weight assigned thereto, wherein when the state machine is operating within one of the plurality of states, a selection of a transition from the one of the plurality of states to another one of the plurality of states may be determined by evaluating the weight assigned to each one of a number of possible transitions from the one of the plurality of states; and a neural network that may receive as inputs data from the semiconductor manufacturing system and may provide as outputs the weights for one or more of the plurality of transitions.

At least one of the states may represent a state of an item of hardware within the semiconductor manufacturing system. At least one of the states may represent a position of a workpiece within the semiconductor manufacturing system. At least one of the states may represent a position of an isolation valve within the system. The neural network may be updated in substantially real time. The neural network may be updated every 20 milliseconds. The inputs to the neural network may include one or more of sensor data, temperature data, a detected workpiece position, an estimated workpiece temperature, an actual workpiece temperature, a valve state, an isolation valve state, robotic drive encoder data, robotic arm position data, end effector height data, a process time, a process status, a pick time, a place time, and a control signal. The inputs to the neural network may include at least one process time for a workpiece within the semiconductor manufacturing system. The at least one process time may include one or more of a target duration, a start time, an end time, and an estimated end time. The inputs may include a transition time. The transition time may include one or more of a pump down to vacuum time and a vent to atmosphere time. At least one of the states may include a transition to itself. The state machine may be updated in substantially real time. The state machine may be updated every 20 milliseconds. The system may also further include a plurality of state machines, each one of the plurality of state machines may control a portion of the semiconductor manufacturing system according to one of a plurality of neural networks.

In one aspect, a computer program product disclosed herein includes computer executable code embodied in a computer readable medium that, when executing on one or more computing devices, performs the steps of: controlling operation of a semiconductor manufacturing system with a state machine to schedule processing of one or more workpieces, the state machine may include a plurality of states associated by a plurality of transitions, each one of the plurality of transitions may have a weight assigned thereto; receiving data from the semiconductor manufacturing system; calculating the weight assigned to each one of a number of possible transitions from a current state of the plurality of states by applying the data as inputs to a neural network; and selecting a transition from the current state of the plurality of states by evaluating the weight assigned to each one of the number of possible transitions from the current state.

At least one of the plurality of states may represent a state of an item of hardware within the semiconductor manufacturing system. At least one of the states may represent a position of a workpiece within the semiconductor manufacturing system. At least one of the states may represent a position of an isolation valve within the system. The computer executable code may further perform the step of updating the neural network in substantially real time. The computer executable code may further perform the step of updating the neural network every 20 milliseconds. The inputs to the neural network may include one or more of sensor data, temperature data, a detected workpiece position, an estimated workpiece temperature, an actual workpiece temperature, a valve state, an isolation valve state, robotic drive encoder data, robotic arm position data, end effector height data, a process time, a process status, a pick time, a place time, and a control signal. The inputs to the neural network may include at least one process time for a workpiece within the semiconductor manufacturing system. The at least one process time may include one or more of a target duration, a start time, an end time, and an estimated end time. The inputs may include a transition time. The transition time may include one or more of a pump down to vacuum time and a vent to atmosphere time. At least one of the states includes a transition to itself. The computer executable code may further perform the step of updating the state machine in substantially real time. The computer executable code may further perform the step of updating the state machine every 20 milliseconds. The computer executable code may further perform the step of controlling operation of a semiconductor manufacturing system with a plurality of state machines, each one of the plurality of state machines controlling a portion of the semiconductor manufacturing system according to one of a plurality of neural networks.

In one aspect, a method disclosed herein includes controlling operation of a semiconductor manufacturing system with a state machine to schedule processing of one or more workpieces, the state machine may include a plurality of states associated by a plurality of transitions, each one of the plurality of transitions may have a weight assigned thereto; receiving data from the semiconductor manufacturing system; calculating the weight assigned to each one of a number of possible transitions from a current state of the plurality of states by applying the data as inputs to a neural network; and selecting a transition from the current state of the plurality of states by evaluating the weight assigned to each one of the number of possible transitions from the current state.

At least one of the plurality of states may represent a state of an item of hardware within the semiconductor manufacturing system. At least one of the states may represent a position of a workpiece within the semiconductor manufacturing system. At least one of the states may represent a position of an isolation valve within the system. The computer executable code may further perform the step of updating the neural network in substantially real time. The computer executable code may further perform the step of updating the neural network every 20 milliseconds. The inputs to the neural network include one or more of sensor data, temperature data, a detected workpiece position, an estimated workpiece temperature, an actual workpiece temperature, a valve state, an isolation valve state, robotic drive encoder data, robotic arm position data, end effector height data, a process time, a process status, a pick time, a place time, an encoder position, a time remaining for a workpiece in a process module, a time remaining for a workpiece in a load lock, and a control signal. The inputs to the neural network may include at least one process time for a workpiece within the semiconductor manufacturing system. The at least one process time may include one or more of a target duration, a start time, an end time, and an estimated end time. The inputs may include a transition time. The transition time may include one or more of a pump down to vacuum time and a vent to atmosphere time. At least one of the states may include a transition to itself. The method may further include the step of updating the state machine in substantially real time. The method may further include the step of updating the state machine every 20 milliseconds. The method may further include the step of controlling operation of a semiconductor manufacturing system with a plurality of state machines, each one of the plurality of state machines may control a portion of the semiconductor manufacturing system according to one of a plurality of neural networks. The method may further include training the neural network to calculate weights for a desired workpiece processing schedule. The state machine may schedule concurrent processing of a plurality of workpieces.

In one aspect, a method disclosed herein includes connecting a plurality of nodes into a neural network, each one of the nodes represented by a programming object; defining a condition for converting one of the plurality of nodes into a second plurality of nodes; and when the condition is met, converting the one of the plurality of nodes into two or more nodes. The condition may include a processing constraint.

In one aspect, a system disclosed herein includes a semiconductor manufacturing system; a controller that controls processing of workpieces within the semiconductor manufacturing system, wherein the controller permits a selection of one or more of a plurality of scheduling techniques to control processing.

• Provisional Patent
• Utility Patent

• What Is a Patent?
• What Is a Trademark or Servicemark?
• What Is a Copyright?
• Patent Laws