Wafer scheduling sequence determination method for semiconductor process equipment and electronic equipment
By obtaining the target scheduling model status of the semiconductor process equipment, determining candidate changes and adding them to the chip scheduling sequence, the problem of rapid and effective determination of the wafer scheduling sequence in the multi-station rotation process chamber equipment is solved, and the generation efficiency and effectiveness are improved.
Patent Information
- Application Number
- CN202410993700.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-23
AI Technical Summary
How to quickly and effectively determine the wafer scheduling sequence of semiconductor process equipment mounted with multi-station rotary process chambers, solve the problem that traditional methods cannot quickly and effectively deal with this problem.
By obtaining the current state of the target scheduling model of the semiconductor process equipment, including multiple libraries and changes, characterizing the available resources, chip holding status and processing status of the chamber or robot, determining candidate changes and adding them to the chip scheduling sequence, realizing chip scheduling Automatic generation of sequences.
The generation efficiency and effectiveness of wafer scheduling sequences are improved, the rapid and effective determination of sequences is ensured, and the complex machining logic of multi-station rotation process chambers is adapted to.
Smart Images

Figure CN118943050A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of semiconductor process equipment, specifically, to equipment scheduling technology in the field of semiconductor process equipment, and more specifically, to a method for determining a wafer scheduling sequence for semiconductor process equipment and an electronic device. Background Art
[0002] With the continuous advancement of semiconductor technology, the production capacity requirements of semiconductor process equipment are getting higher and higher. The production capacity of semiconductor process equipment is related to the number of chambers it is mounted on. However, due to site limitations, semiconductor process equipment will also limit the equipment's footprint while pursuing production capacity. Semiconductor process equipment mounted with multi-station rotary process chambers has become a more efficient equipment structure due to its small footprint and higher production capacity per unit area.
[0003] However, the multi-station rotary process chamber has changed the simple processing logic architecture of the traditional single-station chamber, and has put forward higher requirements and challenges for the scheduling of semiconductor process equipment. How to quickly and effectively determine the wafer scheduling sequence of semiconductor process equipment equipped with a multi-station rotary process chamber has become a problem that needs to be solved urgently. Summary of the invention
[0004] To solve the above technical problems, the present application provides a wafer scheduling sequence determination method and electronic equipment for semiconductor process equipment to quickly and effectively determine the wafer scheduling sequence of semiconductor process equipment equipped with a multi-station rotary process chamber.
[0005] To achieve the above technical objectives, the present application provides the following technical solutions:
[0006] In a first aspect, an embodiment of the present specification provides a method for determining a wafer scheduling sequence for a semiconductor process device, wherein the semiconductor process device includes a plurality of chambers and a robot for transferring wafers between chambers, wherein the chambers include a process chamber and a non-process chamber, wherein the process chamber includes at least one multi-station rotary process chamber, and wherein the method includes:
[0007] Acquire the current state of the target scheduling model of the semiconductor process equipment; the target scheduling model includes a plurality of places and a plurality of transitions, each of the places respectively represents the available resources of the chamber or the robot, the wafer holding state of the robot and the processing state of the chamber, and the transition represents the action performed by the robot; the state of the target scheduling model includes the number of tokens of each place and the position vector of the multi-station rotary process chamber, and the position vector includes the number of wafers in each station of the multi-station rotary process chamber; the number of tokens of each place respectively represents the number of available resources of the chamber or the robot, the number of wafers held by the robot and the number of wafers in the chamber;
[0008] When the current state represents that the wafer scheduling sequence is not determined to be completed, the target data corresponding to each candidate transition is obtained, the target data including the subsequent state of the target scheduling model and the return value of the position vector after the candidate transition is executed in the current state; the candidate transition includes at least part of a plurality of the transitions, the return value is generated when the position vector changes, the return value includes the value of the target bit of the position vector before the change, and the target bit represents the station for taking wafers in the multi-station rotary process chamber;
[0009] A target transition at a current moment is determined based on each of the target data, and the target transition is added to the wafer scheduling sequence, wherein the target transition includes a preferred execution action of the robot at the current moment.
[0010] Optionally, obtaining target data corresponding to each candidate transition includes:
[0011] Based on the connection direction between the candidate transition and the target place, updating the number of tokens of the target place, the target place includes the place connected to the candidate transition; the connection direction between the candidate transition and the target place is used to determine the direction of change of the number of tokens of the target place when the candidate transition is executed;
[0012] When the target location includes a location corresponding to the multi-station rotary process chamber, based on the current state of the target scheduling model and the robot action corresponding to the candidate transition, the position vector of the multi-station rotary process chamber is updated and the return value is generated.
[0013] Optionally, updating the number of tokens of the target place based on the connection direction between the candidate transition and the target place includes:
[0014] When the connection direction between the candidate transition and the target library place is directed from the candidate transition to the target library place, the number of tokens of the target library place is increased by a first preset value; the direction from the candidate transition to the target library place indicates that when the robot action corresponding to the candidate transition is executed, the number of resources corresponding to the target library released is the first preset value, the number of wafers held by the robot is increased by the first preset value, or the number of wafers processed by the chamber is increased by the first preset value; or
[0015] When the connection direction between the candidate transition and the target library is from the target library to the candidate transition, the number of tokens of the target library is reduced by the first preset value; the target library pointing to the candidate transition indicates that when the robot action corresponding to the candidate transition is executed, the number of resources corresponding to the target library occupied is the first preset value, the number of wafers held by the robot is reduced by the first preset value, or the number of wafers processed by the chamber is reduced by the first preset value.
[0016] Optionally, updating the position vector of the multi-station rotary process chamber and generating the return value based on the current state of the target scheduling model and the robot action corresponding to the candidate transition includes:
[0017] Determining a current working state of the multi-station rotary process chamber based on a current state of the target scheduling model, wherein the working state of the multi-station rotary process chamber includes an entry state, a stable state, and an exit state;
[0018] Based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber, the position vector of the multi-station rotary process chamber is updated and the return value is generated.
[0019] Optionally, updating the position vector of the multi-station rotary process chamber and generating the return value based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber includes:
[0020] When the current working state of the multi-station rotary process chamber is an entering state or a stable state, taking the current value of the target position in the position vector as a return value;
[0021] Based on the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber, an updated value of the position vector is determined.
[0022] Optionally, determining the updated value of the position vector based on the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber includes:
[0023] If the robot action corresponding to the candidate transition includes placing a wafer into the multi-station rotary process chamber, the current value of the target position is deleted, and the number of wafers in each station other than the target position in the position vector is translated based on the direction of the target position, and the number of wafers in the station where the number of wafers in the position vector is empty is set to a second preset value;
[0024] If the robot action corresponding to the candidate transition includes taking a wafer from the multi-station rotary process chamber, the value of the target bit is updated to 0.
[0025] Optionally, updating the position vector of the multi-station rotary process chamber and generating the return value based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber includes:
[0026] When the current working state of the multi-station rotary process chamber is an exit state and the robot action corresponding to the candidate transition includes taking a sheet from the multi-station rotary process chamber, obtaining the current value of the target position;
[0027] If the current value of the target bit is greater than 0, the current value of the target bit is used as the return value, and the value of the target bit is updated to 0;
[0028] If the current value of the target position is 0, the shift operation is repeatedly performed until the current value of the target position is greater than 0; the shift operation includes deleting the current value of the target position, and translating the number of chips in each station other than the target position in the position vector based on the direction of the target position, and setting the number of chips in the station where the number of chips in the position vector is empty to 0.
[0029] Optionally, the obtaining the current state of the target scheduling model of the semiconductor process equipment includes:
[0030] When the current state is the initial state, the number of tokens in each of the libraries and the position vector of the multi-station rotary process chamber are determined based on the current operating state of the semiconductor process equipment and the configuration information of the chamber and the robot.
[0031] Optionally, determining the target transition at the current moment based on each target data includes:
[0032] If the number of tokens in each of the places and the position vector do not have a value less than 0, and the return value meets the target value corresponding to the candidate transition, the candidate transition is used as an enabling transition, and the enabling transition represents an executable action of the manipulator at the current moment;
[0033] When there are multiple enabling transitions, the target transition is determined from the enabling transitions based on the current states of the chambers and / or manipulators corresponding to the libraries connected to the enabling transitions.
[0034] In a second aspect, an embodiment of the present specification provides an electronic device comprising at least one processor and at least one memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, a method for determining a wafer scheduling sequence for semiconductor process equipment as described in any one of the above items is implemented.
[0035] In a third aspect, an embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for determining a wafer scheduling sequence for semiconductor process equipment as described in any one of the above items is implemented.
[0036] In a fourth aspect, an embodiment of the present specification provides a computer program product or a computer program, wherein the computer program product includes a computer program, and the computer program is stored in a computer-readable storage medium; the processor of the computer device reads the computer program from the computer-readable storage medium, and when the processor executes the computer program, it implements the wafer scheduling sequence determination method for semiconductor process equipment as described in any one of the above items.
[0037] It can be seen from the above technical solution that the embodiment of the present application provides a method for determining a wafer scheduling sequence for semiconductor process equipment and an electronic device. The semiconductor process equipment includes multiple chambers and a robot for transferring wafers between chambers. The chambers include process chambers and non-process chambers. The process chambers include at least one multi-station rotary process chamber. The method for determining a wafer scheduling sequence for semiconductor process equipment obtains the current state of a target scheduling model of the semiconductor process equipment. The target scheduling model includes multiple libraries and multiple transitions. Each library represents the available resources of the chamber or robot, the wafer holding state of the robot, and the processing state of the chamber. The transition represents the action performed by the robot. The state of the target scheduling model includes the number of tokens of each library and the position vector of the multi-station rotary process chamber. The position vector includes the number of wafers in each station of the multi-station rotary process chamber. The number of tokens of each library represents the number of available resources of the chamber or robot, the number of wafers held by the robot, and the number of wafers in the chamber. Therefore, the state of the target scheduling model can fully describe the mounting of the multi-station rotary process chamber. The invention relates to a method for determining a target transition of a semiconductor process equipment in a process chamber and a processing state thereof; when the current state of the target scheduling model indicates that the wafer scheduling sequence has not been determined to be completed, obtaining target data corresponding to each candidate transition, wherein the candidate transition includes at least part of multiple transitions, and the target data includes the subsequent state of the target scheduling model after executing the candidate transition in the current state and the return value of the position vector, and the return value is generated when the position vector changes, and the return value includes the value of the target bit of the position vector before the change, and the target bit represents the station for taking wafers in the multi-station rotary process chamber, and determining the target transition at the current moment based on each target data, wherein the target transition includes the preferred execution action of the robot at the current moment, and adding the target transition to the wafer scheduling sequence, realizing the automatic generation of the wafer scheduling sequence, thereby ensuring the generation efficiency of the wafer scheduling sequence, and at the same time, in the process of determining the wafer scheduling sequence of the semiconductor process equipment, it is possible to determine whether the candidate transition meets the execution condition according to the value of the target bit of the position vector and the state change of the target scheduling model after executing the candidate transition, thereby ensuring the validity of the wafer scheduling sequence. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0039] Figure 1 A schematic diagram of the structure of a semiconductor process equipment provided for one embodiment of the present specification;
[0040] Figure 2 A schematic flow chart of a method for determining a wafer scheduling sequence for semiconductor process equipment provided as an embodiment of the present specification;
[0041] Figure 3 A schematic diagram of wafer distribution at each station in a multi-station rotary process chamber when the working state of the multi-station rotary process chamber is an entry state provided for one embodiment of the present specification;
[0042] Figure 4 A schematic diagram of wafer distribution at each station in a multi-station rotary process chamber provided in one embodiment of the present specification when the working state of the multi-station rotary process chamber is a stable state;
[0043] Figure 5 A schematic diagram of wafer distribution at each station in a multi-station rotary process chamber provided in one embodiment of the present specification when the working state of the multi-station rotary process chamber is an exit state;
[0044] Figure 6 A schematic diagram of a process path provided for one embodiment of this specification;
[0045] Figure 7 A schematic diagram of another process path provided for one embodiment of the present specification;
[0046] Figure 8 A schematic diagram of the connection between a transition and movement library and a processing library provided for one embodiment of the present specification;
[0047] Fig. 9 A schematic diagram of the structure of a target scheduling model provided for one implementation of this specification. DETAILED DESCRIPTION
[0048] Unless otherwise defined, the technical terms or scientific terms used in the embodiments of this specification shall have the common meanings understood by persons with ordinary skills in the field to which this specification belongs. The words "first", "second" and similar words used in the embodiments of this specification do not indicate any order, quantity or importance, but are only used to avoid confusion of constituent elements.
[0049] Unless the context requires otherwise, throughout the specification, "plurality" means "at least two", and "including" is interpreted as an open, inclusive meaning, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "examples", "specific examples" or "some examples" are intended to indicate that a particular feature, structure, material or characteristic associated with the embodiment or example is included in at least one embodiment or example of the specification. The schematic representation of the above terms does not necessarily refer to the same embodiment or example.
[0050] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.
[0051] Overview
[0052] As described in the background technology, with the continuous advancement of semiconductor technology, the production capacity requirements of semiconductor process equipment are becoming higher and higher. The production capacity of semiconductor process equipment is related to the number of chambers mounted thereon. However, due to site limitations, semiconductor process equipment will also limit the equipment's footprint while pursuing production capacity.
[0053] The semiconductor process equipment with a multi-station rotary process chamber can realize multi-station simultaneous processing, has a small footprint, and has a high production capacity per unit area. Through the rotation of the turntable inside the chamber, the wafer undergoes a process at each processing station, which can ensure the consistency of the process and process results of different wafers, thereby becoming a more efficient equipment structure. Taking a rotary process chamber containing four stations as an example, a wafer is placed in the chamber of the first station, and after completing a process, it rotates clockwise to the second station. At the same time, the second wafer can be placed in the chamber of the first station. After completing a process, the first wafer rotates to the third station, and the second wafer rotates to the second station. At this time, the third wafer can be placed in the chamber of the first station. After completing a process, the first wafer rotates to the fourth station, the second wafer rotates to the third station, and the third wafer rotates to the second station. At this time, the fourth wafer can be placed in the chamber of the first station. After completing a process, the first wafer is taken out and rotated once, and the fifth wafer can be placed in the chamber of the first station.
[0054] At the same time, semiconductor processing and manufacturing is one of the more complex manufacturing industries. In order to adapt to the complexity and precision requirements of semiconductor processing, semiconductor process equipment is often composed of a variety of modules. For example, commonly used semiconductor process equipment is called cluster equipment, such as Figure 1 As shown, it usually includes a processing chamber (101, 102), a vacuum transfer robotic arm (VTR) 103, a vacuum lock (LL, Loadlock) (104, 105) and an equipment front end module (EFEM, Equipment Front End Module) 106 and other modules. The equipment front end module 106 can be further disassembled into an atmospheric transfer robotic arm (ATR) 1061, a load port (LP, LoadPort) 1062 and an aligner 1063. The process chamber may include a single-station chamber 101 and a rotary process chamber 102, and the vacuum lock may include a first vacuum lock 104 and a second vacuum lock 105. Through the scheduling of semiconductor process equipment, the processing sequence of the wafers in each module of the semiconductor process equipment can be reasonably arranged, so that a certain number of wafers can be completed in a shorter time or the equipment capacity can be higher under the premise of meeting the process requirements.
[0055] However, the multi-station rotary process chamber has changed the simple processing logic architecture of the traditional single-station chamber, and has put forward higher requirements and challenges for the scheduling of semiconductor process equipment. How to quickly and effectively determine the wafer scheduling sequence of semiconductor process equipment equipped with a multi-station rotary process chamber has become a problem that needs to be solved urgently.
[0056] In order to solve the problem that the traditional method cannot quickly and effectively determine the wafer scheduling sequence of the semiconductor process equipment mounted with a multi-station rotary process chamber, in the technical solution of the present application, the semiconductor process equipment includes multiple chambers and a robot for transferring wafers between chambers, the chambers include process chambers and non-process chambers, and the process chambers include at least one multi-station rotary process chamber. The wafer scheduling sequence determination method for the semiconductor process equipment obtains the current state of the target scheduling model of the semiconductor process equipment. The target scheduling model includes multiple libraries and multiple transitions. Each library represents the available resources of the chamber or the robot, the wafer holding state of the robot, and the processing state of the chamber. The transition represents the action performed by the robot. The state of the target scheduling model includes the number of tokens of each library and the position vector of the multi-station rotary process chamber. The position vector includes the number of wafers in each station of the multi-station rotary process chamber. The number of tokens of each library represents the number of available resources of the chamber or the robot, the number of wafers held by the robot, and the number of wafers in the chamber. Therefore, the state of the target scheduling model can fully describe the multi-station mounted with the multi-station. The invention relates to a semiconductor process equipment of a rotating process chamber, and changes in the processing procedures and processing states; when the current state of the target scheduling model represents that the wafer scheduling sequence has not been determined to be completed, the target data corresponding to each candidate transition is obtained, the candidate transition includes at least part of the multiple transitions, the target data includes the subsequent state of the target scheduling model after executing the candidate transition in the current state and the return value of the position vector, the return value is generated when the position vector changes, the return value includes the value of the target bit of the position vector before the change, the target bit represents the station for taking the wafer in the multi-station rotary process chamber, and the target transition at the current moment is determined based on each target data, the target transition includes the preferred execution action of the robot at the current moment, and the target transition is added to the wafer scheduling sequence, the automatic generation of the wafer scheduling sequence is realized, thereby ensuring the generation efficiency of the wafer scheduling sequence, and at the same time, in the process of determining the wafer scheduling sequence of the semiconductor process equipment, it is possible to determine whether the candidate transition meets the execution condition according to the value of the target bit of the position vector and the state change of the target scheduling model after executing the candidate transition, thereby ensuring the validity of the wafer scheduling sequence.
[0057] Based on the above-mentioned inventive concept, the wafer scheduling sequence determination method for semiconductor process equipment provided in the embodiment of this specification is exemplarily described below.
[0058] Exemplary Methods
[0059] refer to Figure 2The present specification provides a method for determining a wafer scheduling sequence for a semiconductor process device, wherein the semiconductor process device comprises a plurality of chambers and a robot for transferring wafers between chambers, wherein the chambers comprise a process chamber and a non-process chamber, wherein the process chamber comprises at least one multi-station rotary process chamber, and wherein the method comprises:
[0060] S201. Obtain the current state of the target scheduling model of the semiconductor process equipment; the target scheduling model includes multiple places and multiple transitions, each of the places represents the available resources of the chamber or the robot, the wafer holding state of the robot and the processing state of the chamber, and the transition represents the action performed by the robot; the state of the target scheduling model includes the number of tokens of each place and the position vector of the multi-station rotary process chamber, and the position vector includes the number of wafers in each station of the multi-station rotary process chamber; the number of tokens of each place represents the number of available resources of the chamber or the robot, the number of wafers held by the robot and the number of wafers in the chamber.
[0061] Specifically, the semiconductor process equipment may include multiple chambers and at least one robot for transferring wafers between chambers. The chambers may include process chambers and non-process chambers, wherein the process chambers may include at least one multi-station rotary process chamber and may also include single-station chambers. The non-process chambers may include vacuum locks and calibration positions, etc. The robots may include vacuum robots, atmospheric robots, etc.
[0062] The target scheduling model of semiconductor process equipment can be constructed based on the target process path of the semiconductor process equipment. The target scheduling model can be a Petri net model, including a place set P and a transition set T. The place set P can include multiple places, and the transition set T can include multiple transitions. Transitions and places can be connected alternately. Places and transitions can be connected through directed edges, and each directed edge can constitute a directed edge set F, that is, the target scheduling model N can be expressed as N = (P, T, F).
[0063] For any library place, the library place can represent one of the available resources of the chamber or robot, the film holding state of the robot, and the processing state of the chamber; for any transition, the transition can represent the action performed by the robot, and the action performed by the robot can include taking and placing the film.
[0064] The place set P can include the resource place set S, the mobile place set A and the processing place set H, that is, P = A∪H∪S, and The resource library set S may include multiple resource libraries, which are used to characterize the available resources of the chamber or robot, the mobile library set A may include multiple mobile libraries, which are used to characterize the wafer holding state of the robot, and the processing library set H may include multiple processing libraries, which are used to characterize the processing state of the chamber. For example, the processing library corresponding to the process chamber is used to characterize the process processing state of the process chamber, the processing library corresponding to the vacuum lock is used to characterize the switching state between the vacuum environment and the atmospheric environment of the vacuum lock, and the processing library corresponding to the calibration position is used to characterize the wafer calibration state of the calibration position.
[0065] During implementation, a resource library and a processing library for each chamber can be constructed according to the target process path of the semiconductor process equipment, and a resource library and a mobile library for each robot can be constructed.
[0066] The state of the target scheduling model is the state of the target scheduling model at the current moment, and the state of the target scheduling model may include the number of tokens of each library. Among them, the number of tokens of the resource library can represent the number of available resources of the chamber or the robot, that is, the number of wafers that can be accommodated in the remaining space, the number of tokens of the mobile library can represent the number of wafers held by the robot, that is, the occupied space, and the number of tokens of the processing library can represent the number of wafers in the chamber, that is, the occupied space. It can be understood that the total number of tokens of all libraries remains unchanged and only circulates with the flow of wafer processing status.
[0067] The state of the target scheduling model may also include the position vector of each multi-station rotary process chamber. The position vector of the multi-station rotary process chamber may include multiple data bits, and the multiple data bits respectively represent multiple stations of the multi-station rotary process chamber. For any data bit in the position vector, the value of the data bit represents the number of wafers in the corresponding station in the multi-station rotary process chamber. It can be understood that a chamber of a station can hold one wafer or multiple wafers. In addition, the number of tokens of the resource library corresponding to the multi-station rotary process chamber may include the sum of the number of wafers that can be accommodated in each station, and the number of tokens of the processing library corresponding to the multi-station rotary process chamber may include the sum of the number of wafers in each station.
[0068] Therefore, the state of the target scheduling model can completely describe the processing steps and changes in the processing state of the semiconductor process equipment equipped with a multi-station rotary process chamber.
[0069] S202. When the current state represents that the wafer scheduling sequence has not been completed, the target data corresponding to each candidate transition is obtained, the target data including the successor state of the target scheduling model and the return value of the position vector after the candidate transition is executed in the current state; the candidate transition includes at least part of the multiple transitions, the return value is generated when the position vector changes, and the return value includes the value of the target bit of the position vector before the change, and the target bit represents the station for taking wafers in the multi-station rotary process chamber.
[0070] Specifically, the chip scheduling sequence of the semiconductor process equipment may include a transition sequence, that is, the transitions may be performed in a sequential order to realize the scheduling of the semiconductor process equipment. In the process of determining the chip scheduling sequence of the semiconductor process equipment, before determining a new transition, it may be first detected whether the chip scheduling sequence is determined to be completed. For example, the termination state of the target scheduling model may be set, and when the current state of the target scheduling model is the termination state, it indicates that the chip scheduling sequence is determined to be completed. The termination state may include that the number of tokens of each resource library is the maximum number of tokens of the resource library, the number of tokens of each mobile library and the number of tokens of each processing library are all 0, and at the same time, there is no new chip to be placed in the loading port. It is understandable that the library corresponding to the loading port may not be set.
[0071] When the wafer scheduling sequence characterized by the current state is determined to be complete, the current wafer scheduling sequence may be used as the final wafer scheduling sequence and output.
[0072] When the current state characterizes that the wafer scheduling sequence is not determined to be complete, a new transition can be determined based on the current state of the target scheduling model. In the process of determining the new transition, the target data corresponding to each candidate transition can be obtained. The candidate transitions may include all transitions in the transition set T, and may also include some transitions in the transition set T. For example, the candidate transitions may be determined based on the current state of the target scheduling model and the historical scheduling sequence, and the candidate transitions may also be randomly selected from the transition set T.
[0073] For any candidate transition, the target data of the candidate transition may include the subsequent state of the target scheduling model after the candidate transition is executed in the current state, and the subsequent state is the state of the target scheduling model after the candidate transition is executed in the current state. In implementation, the number of tokens of each target library place can be updated according to the robot action corresponding to the candidate transition, or according to the connection direction between the candidate transition and each target library place, and the target library place is the library place connected to the candidate transition. At the same time, when the target library place includes the library place corresponding to the multi-station rotary process chamber, that is, the robot hand corresponding to the candidate transition is used to put or take a film into the multi-station rotary process chamber, the position vector of the multi-station rotary process chamber can also be updated according to the robot action corresponding to the candidate transition, and at the same time, the return value of the position vector of the multi-station rotary process chamber is obtained, and the return value may include the value of the target position before the position vector of the multi-station rotary process chamber is updated. The return value can be used to determine whether the robot action corresponding to the candidate transition can be executed. For example, the target position can represent the station used for taking a film among the stations of the multi-station rotary process chamber.
[0074] S203, determining a target transition at a current moment based on each of the target data, and adding the target transition to the wafer scheduling sequence, wherein the target transition includes a preferred execution action of the robot at the current moment.
[0075] Specifically, when there are multiple candidate transitions, it is possible to determine whether each candidate transition meets the execution condition based on the target data corresponding to each candidate transition, and determine one candidate transition from each candidate transition that meets the execution condition as the target transition. When there is only one candidate transition, it is possible to determine whether the candidate transition meets the execution condition based on the target data corresponding to the candidate transition, and when the candidate transition meets the execution condition, the candidate transition is used as the target transition, that is, the target transition includes the preferred execution action of the manipulator at the current moment.
[0076] For any candidate transition, when determining whether the candidate transition meets the execution condition based on the target data corresponding to the candidate transition, when there is no value less than 0 in the number of tokens and the position vector of each library, and the return value of the position vector meets the target value corresponding to the candidate transition, it is determined that the candidate transition meets the execution condition; when there is a value less than 0 in the number of tokens or the position vector of each library, or the return value of the position vector does not meet the target value corresponding to the candidate transition, it is determined that the candidate transition does not meet the execution condition. By comprehensively considering the number of tokens in each library, the number of chips in each workstation in the position vector and the return value of the position vector, the reliability of the determination result of whether the candidate transition meets the execution condition can be effectively guaranteed, thereby ensuring the effectiveness of the chip scheduling sequence.
[0077] It is understandable that if there is no candidate transition that meets the execution condition, steps S202 and S203 may be repeatedly executed until the target transition is obtained.
[0078] After obtaining the target transition, the target transition can be added to the wafer scheduling sequence of the semiconductor process equipment, and step S201 is repeated until the wafer scheduling sequence is determined to be completed, thereby realizing automatic generation of the wafer scheduling sequence and ensuring the generation efficiency of the wafer scheduling sequence.
[0079] In order to quickly and effectively obtain the target data corresponding to each candidate transition, in one embodiment of the present specification, obtaining the target data corresponding to each candidate transition includes:
[0080] Based on the connection direction between the candidate transition and the target place, updating the number of tokens of the target place, the target place includes the place connected to the candidate transition; the connection direction between the candidate transition and the target place is used to determine the direction of change of the number of tokens of the target place when the candidate transition is executed;
[0081] When the target location includes a location corresponding to the multi-station rotary process chamber, based on the current state of the target scheduling model and the robot action corresponding to the candidate transition, the position vector of the multi-station rotary process chamber is updated and the return value is generated.
[0082] Specifically, in the process of constructing the target scheduling model, after determining multiple places and multiple transitions, the connection direction between the place and the transition can be further determined, that is, the direction of the directed edge between the place and the transition. For any transition, the target place corresponding to the transition may include the place connected to the transition. When the transition is executed, the number of tokens of the corresponding target place will change. The corresponding relationship between the transition, the target place, the connection direction and the direction of change of the number of tokens can be set. In implementation, the direction of change of the number of tokens of the corresponding target place when the transition is executed can be determined based on the connection direction between the transition and the corresponding target place and the corresponding relationship. The direction of change of the number of tokens can be increase or decrease.
[0083] At the same time, the number of films taken or placed when the transition is executed can be determined based on the configuration information of the manipulator corresponding to the transition, and the number of films taken or placed can be used as the change in the number of tokens of each target library. Thus, the number of tokens of each target library can be updated quickly and accurately based on the change direction and change amount of the number of tokens of the target library.
[0084] It can be understood that, for each place other than the target place in the place set P, the number of tokens thereof may remain unchanged when the candidate transition is executed.
[0085] At the same time, it can be determined whether the target library corresponding to the candidate transition includes the library corresponding to the multi-station rotary process chamber. If not, the position vector of the multi-station rotary process chamber may not be updated, and the return value of the position vector is empty; if included, the position vector of the multi-station rotary process chamber may be further updated based on the current state of the target scheduling model and the action of the robot corresponding to the candidate transition, and the return value of the position vector is generated. For example, the current working state of the multi-station rotary process chamber may be determined based on the current state of the target scheduling model, and the position vector of the multi-station rotary process chamber may be updated and the return value of the position vector may be generated according to the current working state of the multi-station rotary process chamber, the action of the robot corresponding to the candidate transition, and the current value of the position vector of the multi-station rotary process chamber.
[0086] Among them, the working state of the multi-station rotary process chamber may include an entry state, a stable state and an exit state. The entry state may include a state when there are wafers entering the multi-station rotary process chamber and no wafers are taken out after processing, that is, a state before the multi-station rotary process chamber is fully loaded; the stable state may include a state when there are wafers entering the multi-station rotary process chamber and wafers are taken out after processing, that is, a state when the multi-station rotary process chamber is fully loaded; the exit state may include a state when there are no wafers entering the multi-station rotary process chamber and there are wafers being processed, that is, a state when the multi-station rotary process chamber is about to end its work. Taking a multi-station rotary process chamber including four stations as an example, when the working states of the multi-station rotary process chamber are an entry state, a stable state and an exit state, the wafer distribution of each station in the multi-station rotary process chamber may be as follows: Figures 3 to 5 As shown, Figure 3 In the embodiment, wafer a and wafer b have entered the multi-station rotary process chamber and are being processed, and no wafer has been taken out after the processing is completed. At this time, the working state of the multi-station rotary process chamber is the entering state; Figure 4 In the process, wafer a, wafer b, wafer c and wafer d are processed 4 / 3 / 2 / 1 times respectively, that is, wafer a can be taken out after processing, and wafer e can be put in after wafer a is taken out. At this time, the working state of the multi-station rotary process chamber is stable; Figure 5 In the process, wafer d and wafer e are in the processing state, and no new wafer enters. At this time, the working state of the multi-station rotary process chamber is the exit state.
[0087] In a feasible implementation manner, updating the number of tokens of the target place based on the connection direction between the candidate transition and the target place includes:
[0088] When the connection direction between the candidate transition and the target library place is directed from the candidate transition to the target library place, the number of tokens of the target library place is increased by a first preset value; the direction from the candidate transition to the target library place indicates that when the robot action corresponding to the candidate transition is executed, the number of resources corresponding to the target library released is the first preset value, the number of wafers held by the robot is increased by the first preset value, or the number of wafers processed by the chamber is increased by the first preset value; or
[0089] When the connection direction between the candidate transition and the target library is from the target library to the candidate transition, the number of tokens of the target library is reduced by the first preset value; the target library pointing to the candidate transition indicates that when the robot action corresponding to the candidate transition is executed, the number of resources corresponding to the target library occupied is the first preset value, the number of wafers held by the robot is reduced by the first preset value, or the number of wafers processed by the chamber is reduced by the first preset value.
[0090] Specifically, when constructing a target scheduling model, a transition pointing to a location may indicate that the robot action corresponding to the transition releases the resources corresponding to the target location, increases the number of wafers held by the robot, or increases the number of wafers processed by the chamber when it is executed, and a location pointing to a transition may indicate that the robot action corresponding to the transition occupies the resources corresponding to the target location, reduces the number of wafers held by the robot, or reduces the number of wafers processed by the chamber when it is executed.
[0091] In implementation, when constructing a target scheduling model, after determining multiple warehouses and multiple transitions, the transitions, mobile warehouses, and processing warehouses can be connected based on the target process path of the semiconductor process equipment. For example, the transitions can be divided into wafer pick-up transitions and wafer placement transitions according to the robot actions corresponding to the transitions, and the transitions, mobile warehouses, and processing warehouses can be connected in sequence in the order and direction of "wafer pick-up transition → mobile warehouse → wafer placement transition → processing warehouse → wafer pick-up transition", that is, the wafer pick-up transition points to the mobile warehouse, the wafer placement transition points to the processing warehouse, and the wafer pick-up transition can increase the mechanical arm movement. The number of wafers held by the hand and the wafer placing transition can increase the number of wafers processed by the chamber. Therefore, for any transition, the robot action corresponding to the transition can increase the number of tokens pointing to the mobile library or processing library when it is executed; at the same time, the mobile library points to the wafer placing transition, and the processing library points to the wafer taking transition. Since the wafer placing transition can reduce the number of wafers held by the robot and the wafer taking transition can reduce the number of wafers processed by the chamber, therefore, for any transition, the robot action corresponding to the transition can reduce the number of tokens pointing to the mobile library or processing library when it is executed.
[0092] In addition, for any transition, the resource library connected to the transition can be obtained, and the resource library connected to the transition can include the resource library corresponding to the chamber and the resource library corresponding to the manipulator; at the same time, the connection direction of the transition and the resource library connected to it can be determined according to the manipulator action corresponding to the transition. When the manipulator corresponding to the transition acts as a film taker, the connection direction is from the transition to the resource library corresponding to the chamber, and from the resource library corresponding to the manipulator to the transition; when the manipulator corresponding to the transition acts as a film putter, the connection direction is from the transition to the resource library corresponding to the manipulator, and from the resource library corresponding to the chamber to the transition; that is, the film taker transition points to the resource library corresponding to the chamber, and the film putter transition points to the resource library corresponding to the manipulator. Since the film taker transition releases the resources corresponding to the chamber, and the film putter transition releases the resources corresponding to the manipulator, therefore, for any A transition, the robot action corresponding to the transition can increase the number of tokens of the resource library pointed to when it is executed; at the same time, the resource library corresponding to the robot points to the film-taking transition, and the resource library corresponding to the chamber points to the film-putting transition. Since the film-taking transition occupies the resources corresponding to the robot, and the film-putting transition occupies the resources corresponding to the chamber, therefore, for any transition, the robot action corresponding to the transition can reduce the number of tokens of the resource library pointed to by the transition when it is executed, thereby improving the updating efficiency of the state of the target state model while realizing the automatic construction of the target scheduling model.
[0093] In the process of determining the wafer scheduling sequence of semiconductor process equipment, for any candidate transition, the number of tokens of each target library pointed to by the candidate transition can be increased by a first preset value, and at the same time, the number of tokens of each target library pointed to by the candidate transition can be reduced by a first preset value. The first preset value can be determined based on the configuration information of the robot corresponding to the candidate transition. For example, the first preset value can be the number of wafers taken or placed each time by the robot corresponding to the candidate transition, so that the number of tokens of each target library can be updated quickly and efficiently.
[0094] In a feasible implementation, the updating of the position vector of the multi-station rotary process chamber and the generation of the return value based on the current state of the target scheduling model and the robot action corresponding to the candidate transition include:
[0095] Determining a current working state of the multi-station rotary process chamber based on a current state of the target scheduling model, wherein the working state of the multi-station rotary process chamber includes an entry state, a stable state, and an exit state;
[0096] Based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber, the position vector of the multi-station rotary process chamber is updated and the return value is generated.
[0097] Specifically, based on the number of tokens of each library in the current state of the target scheduling model, it can be determined whether there are wafers being processed before the multi-station rotary process chamber. For example, if the number of tokens of the processing library and / or the mobile library before the library corresponding to the multi-station rotary process chamber is greater than 0, it means that there are wafers being processed before the multi-station rotary process chamber. If the number of tokens of the processing library and the mobile library before the library corresponding to the multi-station rotary process chamber are both 0, it means that there are no wafers being processed before the multi-station rotary process chamber.
[0098] If there are wafers being processed in the multi-station rotary process chamber before, and the number of wafers in each station in the position vector of the multi-station rotary process chamber has a value greater than 0, it means that the current working state of the multi-station rotary process chamber is the entering state or the stable state.
[0099] If the multi-station rotary process chamber has not processed any wafers before, and the number of wafers in each station in the position vector of the multi-station rotary process chamber has a value greater than 0, it means that the current working state of the multi-station rotary process chamber is the exit state.
[0100] During implementation, the position vector of the multi-station rotary process chamber can be updated and the return value of the position vector can be generated based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber, thereby effectively ensuring that the position vector update result and the return value are consistent with the actual working state of the multi-station rotary process chamber, thereby improving the effectiveness of the position vector update result and the return value.
[0101] In a feasible implementation manner, updating the position vector of the multi-station rotary process chamber and generating the return value based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition, and the current value of the position vector of the multi-station rotary process chamber includes:
[0102] When the current working state of the multi-station rotary process chamber is an entering state or a stable state, taking the current value of the target position in the position vector as a return value;
[0103] Based on the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber, an updated value of the position vector is determined.
[0104] Specifically, when the current working state of the multi-station rotary process chamber is an entry state or a stable state, the target position can be determined based on the target rotation direction of the multi-station rotary process chamber, and the target position represents the position for taking out the wafer in the multi-station rotary process chamber, that is, the position for performing the last process on the wafer in the multi-station rotary process chamber. Among them, the current value of the target position in the position vector of the multi-station rotary process chamber can be used as the return value, thereby, according to the current value of the target position, it can be effectively determined whether the multi-station rotary process chamber meets the wafer placement or taking out conditions, and then determine whether the execution conditions of the candidate transition are met.
[0105] In addition, the updated value of the position vector can be determined based on the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber. For example, the change state of the wafer in the multi-station rotary process chamber when the candidate transition is executed can be determined based on the robot action corresponding to the candidate transition, and the updated value of the position vector can be determined based on the change state of the wafer in the multi-station rotary process chamber when the candidate transition is executed and the current value of the position vector of the multi-station rotary process chamber, thereby effectively ensuring the reliability of the determination result of the updated value of the position vector.
[0106] In a feasible implementation manner, determining the updated value of the position vector based on the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber includes:
[0107] If the robot action corresponding to the candidate transition includes placing a wafer into the multi-station rotary process chamber, the current value of the target position is deleted, and the number of wafers in each station other than the target position in the position vector is translated based on the direction of the target position, and the number of wafers in the station where the number of wafers in the position vector is empty is set to a second preset value;
[0108] If the robot action corresponding to the candidate transition includes taking a wafer from the multi-station rotary process chamber, the value of the target bit is updated to 0.
[0109] Specifically, if the robot action corresponding to the candidate transition includes placing a wafer into a multi-station rotary process chamber, it is necessary to trigger the multi-station rotary process chamber to rotate once. Through the rotation, the chambers of each station except the station for taking the wafer rotate to the next station. At the same time, the chamber of the station for taking the wafer rotates to the station for placing the wafer. At this time, a wafer that has not been processed by the multi-station rotary process chamber can be placed in the chamber of the station for placing the wafer. Among them, the station for taking the wafer can be physically adjacent to the station for placing the wafer. For example, taking a multi-station rotary process chamber containing four stations as an example, the first station can be a station for placing the wafer, and the fourth station can be a station for taking the wafer. The first station to the fourth station are arranged in sequence, and the fourth station is adjacent to the first station. In the process of updating the position vector, the current value of the target position in the position vector can be deleted, and the number of chips in each station other than the target position in the position vector can be translated one station toward the target position. At this time, the number of chips in the station for placing chips is empty, and the number of chips in the station with an empty number of chips is set to a second preset value. The second preset value can be the number of chips that can be placed by the robot corresponding to the candidate transition at each time, so as to simulate the state change of chips in the multi-station rotary process chamber when the robot places chips into the multi-station rotary process chamber.
[0110] If the robot action corresponding to the candidate transition includes taking wafers from a multi-station rotary process chamber, the value of the target bit can be directly updated to 0, and the number of wafers in the other stations remains unchanged to simulate the state change of the wafers in the multi-station rotary process chamber when the robot takes wafers from the multi-station rotary process chamber in the entering state or stable state.
[0111] Therefore, the position vector of the multi-station rotary process chamber can be updated according to the state change of the wafer in the multi-station rotary process chamber when placing or taking wafers into or out of the multi-station rotary process chamber, thereby ensuring that the update result of the position vector conforms to the actual working situation of the multi-station rotary process chamber, and further ensuring the effectiveness of the update result of the position vector, providing technical support for improving the effectiveness of the determination result of the wafer scheduling sequence.
[0112] In a feasible implementation manner, updating the position vector of the multi-station rotary process chamber and generating the return value based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition, and the current value of the position vector of the multi-station rotary process chamber includes:
[0113] When the current working state of the multi-station rotary process chamber is an exit state and the robot action corresponding to the candidate transition includes taking a sheet from the multi-station rotary process chamber, obtaining the current value of the target position;
[0114] If the current value of the target bit is greater than 0, the current value of the target bit is used as the return value, and the value of the target bit is updated to 0;
[0115] If the current value of the target position is 0, the shift operation is repeatedly performed until the current value of the target position is greater than 0; the shift operation includes deleting the current value of the target position, and translating the number of chips in each station other than the target position in the position vector based on the direction of the target position, and setting the number of chips in the station where the number of chips in the position vector is empty to 0.
[0116] Specifically, when the current working state of the multi-station rotary process chamber is the exit state, it indicates that the multi-station rotary process chamber has not processed any wafers before. At this time, if the robot action corresponding to the candidate transition includes placing a wafer into the multi-station rotary process chamber, the updated value of the position vector can be determined according to the current working state of the multi-station rotary process chamber being the non-exit state (i.e., the entry state or the stable state), and the robot action corresponding to the candidate transition includes the position vector update method when placing a wafer into the multi-station rotary process chamber. It can be understood that since the multi-station rotary process chamber has not processed any wafers before, i.e., there are no wafers that can be placed in the multi-station rotary process chamber, the candidate transition cannot meet the execution conditions, and when determining the candidate transition, the transition can be excluded as a candidate transition.
[0117] If the robot action corresponding to the candidate transition includes taking a wafer from a multi-station rotary process chamber, the current value of the target position in the position vector of the multi-station rotary process chamber can be obtained. If the current value of the target position is greater than 0, it indicates that there is a wafer at the station for taking a wafer in the multi-station rotary process chamber, and the wafer can be taken. At this time, the current value of the target position can be directly used as the return value, and the value of the target position is updated to 0.
[0118] If the current value of the target position is 0, it indicates that there is no wafer at the station used for taking wafers in the multi-station rotary process chamber, and wafer taking cannot be performed. At this time, a shift operation can be performed. The shift operation can include deleting the current value of the target position in the position vector, and translating the number of wafers in each station other than the target position in the position vector in the direction of the target position by one station, so as to rotate the wafers in the multi-station rotary process chamber to the next station for process processing. At this time, the number of wafers in the station used for placing wafers is empty. Since no new wafers enter the multi-station rotary process chamber, the number of wafers in the station with an empty number of wafers can be set to 0 to simulate the state change of the wafers in the multi-station rotary process chamber when the robot takes wafers from the multi-station rotary process chamber in the exit state.
[0119] After executing a shift operation, the current value of the target bit can be further obtained. If the current value of the target bit is greater than 0, the current value of the target bit can be directly used as the return value, and the value of the target bit can be updated to 0. If the current value of the target bit is 0, the shift operation can be continued until the value of the target bit is greater than 0. In this way, the position vector of the multi-station rotary process chamber can be updated and the return value of the position vector can be determined according to the state change of the chip in the multi-station rotary process chamber when the chip is taken to the multi-station rotary process chamber in the exit state, thereby ensuring that the return value and the updated value of the position vector are in line with the actual working of the multi-station rotary process chamber, thereby ensuring the effectiveness of the return value and the updated result of the position vector, and providing technical support for improving the effectiveness of the determination result of the chip scheduling sequence.
[0120] In order to further improve the effectiveness of the determination result of the wafer scheduling sequence, in one embodiment of the present specification, the obtaining of the current state of the target scheduling model of the semiconductor process equipment includes:
[0121] When the current state is the initial state, the number of tokens in each of the libraries and the position vector of the multi-station rotary process chamber are determined based on the current operating state of the semiconductor process equipment and the configuration information of the chamber and the robot.
[0122] Specifically, the initial state may be the state of the target scheduling model when the wafer scheduling sequence is initially determined. When the current state of the target scheduling model is the initial state, it indicates that the wafer scheduling sequence is empty. At this time, the current operating state of the semiconductor process equipment may be obtained. The current operating state may include a process processing state and an idle state. The process processing state indicates that there are wafers being processed in the semiconductor process equipment, and the idle state indicates that there are no wafers being processed in the semiconductor process equipment.
[0123] Among them, when the current operating state of the semiconductor process equipment is idle, the number of tokens of each library and the position vector of the multi-station rotary process chamber can be determined based on the configuration information of the chamber and the robot. For example, the number of tokens of the resource library can be the capacity of the chamber or robot corresponding to the resource library, that is, the maximum number of wafers that can be placed or held. The number of tokens of the mobile library and the processing library can both be 0. At the same time, the number of wafers in each station in the position vector of the multi-station rotary process chamber is 0.
[0124] When the current operating state of the semiconductor process equipment is the process processing state, the number of tokens of each library and the position vector of the multi-station rotary process chamber can be determined based on the current position of the chip in the semiconductor process equipment and the configuration information of the chamber and the robot. For example, the number of tokens of the resource library can be the difference between the capacity of the chamber or robot corresponding to the resource library and the number of wafers in the chamber or robot corresponding to the resource library, the number of tokens of the mobile library can be the number of wafers in the robot corresponding to the mobile library, and the number of tokens of the processing library can be the number of wafers in the chamber corresponding to the processing library; at the same time, the number of wafers of each station in the position vector of the multi-station rotary process chamber can be the number of wafers of each station in the multi-station rotary process chamber.
[0125] It can be understood that when the current state of the target scheduling model is a non-initial state, that is, when the current state is the state after the target transition is executed in the previous state of the target scheduling model, the current state may include a subsequent state after the target transition is executed in the previous state of the target scheduling model.
[0126] Therefore, the accuracy of the determination result of the current state of the target scheduling model can be effectively guaranteed, and in the process of determining the chip scheduling sequence according to the current state of the target scheduling model, the effectiveness of the determination result of the chip scheduling sequence can be further improved.
[0127] In order to further improve the effectiveness of the determination result of the wafer scheduling sequence, in one embodiment of the present specification, the determining of the target transition at the current moment based on each of the target data includes:
[0128] If the number of tokens in each of the places and the position vector do not have a value less than 0, and the return value meets the target value corresponding to the candidate transition, the candidate transition is used as an enabling transition, and the enabling transition represents an executable action of the manipulator at the current moment;
[0129] When there are multiple enabling transitions, the target transition is determined from the enabling transitions based on the current states of the chambers and / or manipulators corresponding to the libraries connected to the enabling transitions.
[0130] Specifically, for any candidate transition, when the number of tokens of each place in the successor state corresponding to the candidate transition is not less than 0, and there is no value less than 0 in the position vector, it can be further determined whether there is a place corresponding to the multi-station rotary process chamber in the place connected to the candidate transition. If so, it can be further determined whether the return value of the position vector of the multi-station rotary process chamber meets the target value corresponding to the candidate transition. For example, when the return value of the position vector of the multi-station rotary process chamber is the same as the target value corresponding to the candidate transition, it can be determined that the return value of the position vector of the multi-station rotary process chamber meets the target value corresponding to the candidate transition. Otherwise, it is determined that the return value of the position vector of the multi-station rotary process chamber does not meet the target value corresponding to the candidate transition. If it meets, it indicates that the candidate transition meets the execution condition and can be used as an enabling transition. If it does not meet, it indicates that the candidate transition does not meet the execution condition and cannot be used as an enabling transition. The enabling transition represents the executable action of the robot at the current moment, that is, the enabling transition includes the transition that meets the execution condition in the current state of the target scheduling model.
[0131] Among them, when the robot hand corresponding to the candidate transition is used to place wafers into the multi-station rotary process chamber, the target value corresponding to the candidate transition can be 0, indicating that there is no wafer at the station for taking wafers in the multi-station rotary process chamber. At this time, the chamber in the station for taking wafers can be rotated to the station for placing wafers and a new wafer can be placed; when the robot hand corresponding to the candidate transition is used to take wafers into the multi-station rotary process chamber, the target value corresponding to the candidate transition can be a specified value, and the specified value can be the number of wafers taken by the robot each time, indicating that there are wafers at the station for taking wafers in the multi-station rotary process chamber and the wafer taking action can be performed.
[0132] It can be understood that if the number of tokens in each library in the successor state corresponding to the candidate transition has a value less than 0, there is a value less than 0 in the position vector, and the return value of the position vector of the multi-station rotary process chamber does not meet any of the conditions of the target value corresponding to the candidate transition, it indicates that the candidate transition does not meet the execution conditions and cannot be used as an enabling transition.
[0133] In implementation, when the number of enabling transitions is one, the candidate transition can be directly used as the target transition and updated to the chip scheduling sequence. When the number of enabling transitions is multiple, the target transition can be determined from each enabling transition based on the current state of the chamber and / or robot corresponding to the library connected to each enabling transition. The current state of the chamber may include the operating state of the chamber at the current moment, such as idle state, processing state, etc.; the current state of the robot may include the operating state of the robot at the current moment, such as wafer taking state, wafer placing state, wafer transfer state, etc. In implementation, for any enabling transition, the delay corresponding to the enabling transition can be determined based on the current state of the chamber and / or robot corresponding to the library connected to the enabling transition, that is, the length of time required to wait when the enabling transition is executed, and the target transition can be determined from each enabling transition based on the delay corresponding to each enabling transition, so as to effectively ensure the validity of the determination result of the target transition.
[0134] The following uses two different process paths as examples to illustrate the method of determining the wafer scheduling sequence:
[0135] Path L1 may be a process path that does not include a multi-station rotary process chamber, such as Figure 6 As shown, it includes: loading port 1062 → atmospheric manipulator 1061 → calibration position 1063 → atmospheric manipulator 1061 → first vacuum lock 104 → vacuum manipulator 103 → single-station chamber 101 → vacuum manipulator 103 → second vacuum lock 105 → atmospheric manipulator 1061 → loading port 1062 .
[0136] Path L2 may be a process path including a multi-station rotary process chamber, such as Figure 7 As shown, it includes: loading port 1062 → atmospheric manipulator 1061 → calibration position 1063 → atmospheric manipulator 1061 → first vacuum lock 104 → vacuum manipulator 103 → multi-station rotary process chamber 102 → vacuum manipulator 103 → second vacuum lock 105 → atmospheric manipulator 1061 → loading port 1062 .
[0137] That is, the path L1 and the path L2 are different only in the process chamber, and the same target scheduling model can be constructed for the two paths. The process of constructing the target scheduling model may include:
[0138] Construct a transition set T = {t1, t2, t3, t4, t5, t6, t7, t8, t9, t 10}, wherein t1 represents taking a wafer from the loading port 1062 to the atmospheric robot 1061, t2 represents placing a wafer from the atmospheric robot 1061 to the calibration position 1063, t3 represents taking a wafer from the calibration position 1063 to the atmospheric robot 1061, t4 represents placing a wafer from the atmospheric robot 1061 to the first vacuum lock 104, t5 represents taking a wafer from the first vacuum lock 104 to the vacuum robot 103, t6 represents placing a wafer from the vacuum robot 103 to the process chamber, t7 represents taking a wafer from the process chamber to the vacuum robot 103, t8 represents placing a wafer from the vacuum robot 103 to the second vacuum lock 105, t9 represents taking a wafer from the second vacuum lock 105 to the atmospheric robot 1061, t 10 It indicates that the atmospheric manipulator 1061 places the wafer into the loading port 1062 .
[0139] Construct a mobile library set A = {m1, m3, m5, m7, m9}, where m1 represents the wafer holding state from the loading port 1062 to the calibration position 1063, m3 represents the wafer holding state from the calibration position 1063 to the first vacuum lock 104, m5 represents the wafer holding state from the first vacuum lock 104 to the process chamber, m7 represents the wafer holding state from the process chamber to the second vacuum lock 105, and m9 represents the wafer holding state from the second vacuum lock 105 to the loading port 1062.
[0140] Construct a processing library set H = {p2, p4, p6, p8}, where p2 represents the calibration processing of the calibration position 1063 on the wafer, p4 represents the processing of the first vacuum lock 104 switching from the atmospheric state to the vacuum state, p6 represents the process processing state of the process chamber, and p8 represents the processing of the second vacuum lock 105 switching from the vacuum state to the atmospheric state.
[0141] Construct a resource library set S = {rp0, rp1, rp2, rp3, rob0, rob1}, where rp0 represents the available resources of the calibration position 1063, rp1 represents the available resources of the first vacuum lock 104, rp2 represents the available resources of the process chamber, rp3 represents the available resources of the second vacuum lock 105, rob0 represents the available resources of the atmospheric manipulator 1061, and rob1 represents the available resources of the vacuum manipulator 103, wherein the library corresponding to the loading port 1062 may not be set.
[0142] Therefore, the place set P can be expressed as:
[0143] P=A∪H∪S={m1,p2,m3,p4,m5,p6,m7,p8,m9,rp0,rp1,rp2,rp3,rob0,rob1}.
[0144] Construct a directed edge set F = (F1, F2), where F1 includes directed edges between transition and move places and processing places, that is, F1 = ((H∪A)×T)∪(T×(H∪A)). According to the paths L1 and L2, the connection diagram of the transition and move places and the processing places can be as follows: Figure 8 As shown. F2 includes directed edges between transitions and resource libraries, that is, F2 = (S×T)∪(T×S), where the resource library corresponding to the robot points to the fetch transition, the resource library corresponding to the chamber points to the put-away transition, the fetch transition points to the resource library corresponding to the chamber, and the put-away transition points to the resource library corresponding to the robot. The chamber includes the calibration position 1063, the first vacuum lock 104, the process chamber and the second vacuum lock 105. Thus, Figure 8 The connection between each transition in and the resource library can be obtained as follows Fig. 9 The target scheduling model shown is N=(P, T, F).
[0145] Construct a mapping L between the mobile library, the processing library and the resource library to find the available resources of the corresponding resource library according to the current status of the chip, L = {m1:rob0,p2:rp0,m3:rob0,p4:rp1,m5:rob1,p6:rp2,m7:rob1,p8:rp3,m9:rob0}.
[0146] Before generating a wafer scheduling sequence, the state of the target scheduling model is initialized according to the current task state of the semiconductor process equipment.
[0147] For path L1, when no wafer enters the semiconductor process equipment, the number of tokens of each resource library M(S)={rp0:1, rp1:1, rp2:1, rp3:1, rob0:1, rob1:2}, that is, the maximum capacity of the calibration position 1063, the first vacuum lock 104, the single-station chamber 101, the second vacuum lock 105 and the atmospheric manipulator 1061 is 1, and the vacuum manipulator 103 is a double-arm manipulator with a maximum capacity of 2. The number of tokens of each mobile library M(A)={m1:0, m3:0, m5:0, m7:0, m9:0}, indicating that each mobile library has no token. The number of tokens of each processing library M(H)={p2:0, p4:0, p6:0, p8:0}, indicating that each processing library has no token, and the total number of tokens of each library is |M(S)|=7.
[0148] When a wafer enters the semiconductor process equipment, the number of tokens of the corresponding place is updated according to the position of the wafer in the semiconductor process equipment. The wafer state g is any place in the mobile place set A or the processing place set H, that is, g∈(H∪A). If the wafer state g is a place in the mobile place set A, the number of tokens M(S) of the resource place set S and the number of tokens M(A) of the mobile place set A need to be updated. The update rule is that the number of tokens of the resource place M(L(g)) is reduced by 1, and the number of tokens of the mobile place M(g) is increased by 1. For example, if the chip state is m5 and L(m5) is rob1, the number of tokens of the updated resource library rob1 is M(rob1)=2-1=1, and the number of tokens of the updated mobile library m5 is M(m5)=0+1=1; if the chip state is p6 and L(p6) is rp2, the number of tokens of the updated resource library rp2 is M(rp2)=1-1=0, and the number of tokens of the updated processing library p6 is M(p6)=0+1=1.
[0149] For path L2, when no wafer enters the semiconductor process equipment, the number of tokens of each resource library M(S)={rp0:1, rp1:1, rp2:4, rp3:1, rob0:1, rob1:2}, that is, the maximum capacity of the calibration position 1063, the first vacuum lock 104, the second vacuum lock 105 and the atmospheric manipulator 1061 is 1, the vacuum manipulator 103 is a dual-arm manipulator with a maximum capacity of 2, and the multi-station rotary process chamber 102 includes four stations, each of which can accommodate a wafer, so the maximum capacity of the multi-station rotary process chamber 102 is 4. The number of tokens of each mobile library M(A)={m1:0, m3:0, m5:0, m7:0, m9:0}, indicating that each mobile library has no token. The number of tokens of each processing location M(H) = {p2:0, p4:0, p6:0, p8:0}, indicating that each processing location has no token, and the total number of tokens of each location is |M(S)| = 10. At the same time, the position vector W of the multi-station rotary process chamber 102 = [0, 0, 0, 0], indicating that each station of the multi-station rotary process chamber 102 has no wafer.
[0150] When a wafer enters the semiconductor process equipment, the number of tokens of the corresponding place is updated according to the position of the wafer in the semiconductor process equipment. The wafer state g is any place in the mobile place set A or the processing place set H, that is, g∈(H∪A). If the wafer state g is a place in the mobile place set A, the number of tokens M(S) of the resource place set S and the number of tokens M(A) of the mobile place set A need to be updated. The update rule is that the number of tokens of the resource place M(L(g)) is reduced by 1, and the number of tokens of the mobile place M(g) is increased by 1. For example, if the wafer state is m5 and L(m5) is rob1, the number of tokens of the resource library rob1 is updated to M(rob1)=2-1=1, and the number of tokens of the mobile library m5 is updated to M(m5)=0+1=1; if the wafer state is p6 and L(p6) is rp2, the number of tokens of the resource library rp2 is updated to M(rp2)=4-1=3, and the number of tokens of the processing library p6 is updated to M(p6)=0+1=1. In addition, when there is a wafer in the multi-station rotary process chamber 102, the position vector of the multi-station rotary process chamber 102 needs to be updated, that is, the number of wafers in the station where the wafer is located in the position vector is updated to 1. For example, if there are two wafers in the multi-station rotary process chamber 102, and are located at the first station and the second station respectively, then M(rp2)=4-2=2, M(p6)=0+2=2, and W=[1,1,0,0] are updated.
[0151] In the process of generating a chip scheduling sequence, the target data corresponding to the candidate transition is obtained. The target data includes the number of tokens in each library in the target scheduling model after the candidate transition is executed, the position vector of the multi-station rotary process chamber 102, and the return value of the position vector, and whether the candidate transition is an enabling transition is determined based on the target data.
[0152] In the process of determining the number of tokens of each place in the target scheduling model after the candidate transition is executed, the number of tokens of each place pointing to the candidate transition is reduced by 1, which can be specifically shown as formula (1):
[0153] M(P∩(P×t i ))=M(P∩(P×t i ))-1,i=1,2,…,I (1)
[0154] In the formula, P×t i Represents all candidate transitions t in the target scheduling model i Directed edges, P∩(P×t i ) represents all the candidate transitions t in the target scheduling model i The starting vertex of the directed edge of iFor example, when the candidate transition is t3, the number of tokens in places p2 and rob0 pointing to candidate transition t3 is reduced by 1.
[0155] At the same time, the number of tokens of each place pointed to by the candidate transition is increased by 1, which can be specifically shown as formula (2):
[0156] M(P∩(t i ×P))=M(P∩(t i ×P))+1,i=1,2,…,I (2)
[0157] Where, t i ×P represents all candidate transitions t in the target scheduling model i The outgoing directed edge, P∩(t i ×P) represents all candidate transitions t in the target scheduling model i The end vertex of the outgoing directed edge, that is, the candidate transition t i The library location pointed to.
[0158] For path L2, it is necessary to further determine whether the library connected by the candidate transition includes the library corresponding to the multi-station rotary process chamber 102, and update the position vector of the multi-station rotary process chamber 102 if it includes it. For example, it can be determined or Is it true, R is the resource library corresponding to the multi-station rotary process chamber 102. If it is true, it means that the candidate transition t i The connected libraries include resource libraries corresponding to the multi-station rotary process chamber 102 .
[0159] At this time, when the working state of the multi-station rotary process chamber 102 is the entering state or the stable state, if the library R points to the candidate transition t i , that is, when a wafer is placed into the multi-station rotary process chamber 102, the number of wafers in the last station in the position vector of the multi-station rotary process chamber 102 is used as the return value f1, that is, f1 = W(R∩(R×t i ))[-1],R×t i The candidate transition t pointed to by the resource library corresponding to all multi-station rotary process chambers 102 i Directed edges, R∩(R×t i ) represents the candidate transitions pointed to by the resource libraries corresponding to all multi-station rotary process chambers 102 i The starting vertex of the directed edge of , that is, the resource library corresponding to all multi-station rotary process chambers 102, pointing to the candidate transition t iIn addition, the number of wafers in the last station in the position vector of the multi-station rotary process chamber 102 can be deleted, and the number of wafers in the first three stations can be shifted backward by one position, and the number of wafers in the first station can be set to 1, that is, W(R∩(R×t i ))=[1]∪W(R∩(R×t i ))[1:-1], [1:-1] indicates the number of wafers in the previous station from the first station to the last station.
[0160] If the candidate transition t i Pointing to the library R, that is, taking a wafer from the multi-station rotary process chamber 102, the number of wafers in the last station in the position vector of the multi-station rotary process chamber 102 is used as the return value f2, that is, f2 = W (R ∩ (t i ×R))[-1],t i ×R represents candidate transition t i The directed edge pointing to the resource library corresponding to each multi-station rotary process chamber 102, R∩(t i ×R) represents the candidate transition t i The end vertex of the directed edge pointing to the resource library corresponding to each multi-station rotary process chamber 102, that is, among all the resource libraries corresponding to the multi-station rotary process chamber 102, the candidate transition t i In addition, the number of wafers in the last station in the position vector of the multi-station rotary process chamber 102 can be updated to 0, that is, W(R∩(t i ×R))[-1]=0.
[0161] In addition, when the working state of the multi-station rotary process chamber 102 is exit, if the candidate transition t i Pointing to the library R, that is, taking a wafer from the multi-station rotary process chamber 102, and there is no wafer to be processed later, in the process of updating the position vector, if the number of wafers in the last station is 0, the number of wafers in the last station in the position vector is deleted, and the number of wafers in the first three stations is shifted backward by one position in turn, and the number of wafers in the first station is set to 0, until the number of wafers in the last station is 1, that is, W(R∩(t i ×R))=[0]∪W(R∩(t i ×R))[1:-1]. When the number of wafers in the last station is 1, the number of wafers in the last station in the position vector can be used as the return value f2, and the number of wafers in the last station in the position vector of the multi-station rotary process chamber 102 can be updated to 0.
[0162] For example, when the working state of the multi-station rotary process chamber 102 is the entry state, and the wafer is placed into the multi-station rotary process chamber 102, if W=[1,0,0,0], the number of wafers in the last station 0 can be used as the return value, first shifted and then supplemented by 1 to obtain the updated position vector W=[1,1,0,0]. When the working state of the multi-station rotary process chamber 102 is the stable state, and the wafer is taken from the multi-station rotary process chamber 102, if W=[1,1,1,1], the number of wafers in the last station 1 can be used as the return value, and the number of wafers in the last station is set to 0 to obtain the updated position vector W=[1,1,1,0]. At this time, if the wafer is placed into the multi-station rotary process chamber 102, the number of wafers in the last station 0 can be used as the return value, first shifted and then supplemented by 1 to obtain the updated position vector W=[1,1,1,1]. When the working state of the multi-station rotary process chamber 102 is the exit state, no subsequent wafers enter the multi-station rotary process chamber 102, so only wafers can be taken out. If W=[1,1,0,0], the position vector can be changed to W=[0,1,1,0] and W=[0,0,1,1] in sequence. At this time, the number of wafers in the last station is 1, and the wafers can be taken out. The number of wafers in the last station, 1, is used as the return value, and the number of wafers in the last station is set to 0.
[0163] For any candidate transition, when determining whether the candidate transition is an enabling transition, it is possible to first determine whether the number of tokens in each library in the target scheduling model after the execution of the candidate transition and whether there is a value less than 0 in the position vector of the multi-station rotary process chamber 102. If so, it indicates that the candidate transition does not meet the execution conditions. If not and the library connected to the candidate transition includes the library corresponding to the multi-station rotary process chamber 102, then continue to judge the return value of the position vector of the multi-station rotary process chamber 102. If the candidate transition is a film placement transition and the return value is greater than or equal to 1, or the candidate transition is a film take-out transition and the return value is 0, it indicates that the candidate transition does not meet the execution conditions; if the candidate transition is a film placement transition and the return value is 0, or the candidate transition is a film take-out transition and the return value is greater than or equal to 1, it indicates that the candidate transition meets the execution conditions, that is, the candidate transition is an enabling transition.
[0164] For example, for path L1, the number of tokens in each place is M(P) = {rp0:1, rp1:1, rp2:1, rp3:1, rob0:1, rob1:2, m1:0, m3:0, m5:0, m7:0, m9:0, p2:0, p4:0, p6:0, p8:0}. After executing candidate transition t1, the number of tokens in rob0 decreases by 1, and the number of tokens in m1 increases by 1, that is, the number of tokens in each place changes to M(P)' = {rp0:1, rp1:1, rp2:1, rp3:1, rob0:0, rob1:2, m1:1, m3:0, m5:0, m7:0, m9:0, p2:0, p4:0, p6:0, p8:0}. 0,m7:0,m9:0,p2:0,p4:0,p6:0,p8:0}, there is no place with a token number less than 0, that is, candidate transition t1 is an enabling transition; if candidate transition t1 is executed again on this basis, the token number of each place is changed to M(P)"={rp0:1,rp1:1,rp2:1,rp3:1,rob0:-1,rob1:2,m1:2,m3:0,m5:0,m7:0,m9:0,p2:0,p4:0,p6:0,p8:0}, there is a place with a token number less than 0, that is, candidate transition t1 does not meet the execution conditions.
[0165] For path L2, the number of tokens in each library M(P) = {rp0:1, rp1:1, rp2:4, rp3:1, rob0:1, rob1:1, m1:0, m3:0, m5:1, m7:0, m9:0, p2:0, p4:0, p6:0, p8:0}, and the position vector W of the multi-station rotary process chamber 102 = [0,0,0,0], that is, the vacuum robot 103 holds a wafer, and the next step The multi-station rotary process chamber 102 is placed. At this time, after executing candidate transition t6, the number of tokens of rob1 and p6 is increased by 1, and the number of tokens of m5 and rp2 is reduced by 1, that is, the number of tokens of each library is changed to M(P)'={rp0:1,rp1:1,rp2:3,rp3:1,rob0:1,rob1:2,m1:0,m3:0,m5:0,m7:0,m9:0,p2:0,p4:0,p6: 1, p8:0}, the position vector of the multi-station rotary process chamber 102 is updated to W'=[1,0,0,0], the return value of the position vector of the multi-station rotary process chamber 102 is 0, the number of tokens in each library and the position vector of the multi-station rotary process chamber 102 do not have a value less than 0, and the return value of the position vector is 0, and the candidate transition t6 is an enabling transition; if the candidate transition t6 is executed again on this basis, the number of tokens in each library is changed to M(P)”={rp0:1,rp1:1,rp2:2,rp3:1,rob0:1,rob1:3,m1:0,m3:0,m5:-1,m7:0,m9:0,p2:0,p4:0,p6:2,p8:0}, the position vector of the multi-station rotary process chamber 102 is updated to W”=[1,1,0,0], there is a library with a token number less than 0, that is, the candidate transition t6 does not meet the execution conditions.
[0166] Exemplary Devices
[0167] An embodiment of the present specification also provides an electronic device, including at least one processor and at least one memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the method for determining a wafer scheduling sequence for semiconductor process equipment as described in any of the above embodiments is implemented.
[0168] Exemplary computer program products and storage media
[0169] In addition to the above-mentioned methods and devices, the chip scheduling sequence determination method for semiconductor process equipment provided in the embodiments of this specification can also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the chip scheduling sequence determination method for semiconductor process equipment according to various embodiments of this specification described in the above "Exemplary Method" section of this specification.
[0170] The computer program product may be written in any combination of one or more programming languages to write program code for performing the operations of the embodiments of the present specification, including object-oriented programming languages such as Java, C++, etc., and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.
[0171] In addition, an embodiment of the present specification also provides a computer-readable storage medium on which a computer program is stored, and the computer program is executed by a processor to execute the steps of the wafer scheduling sequence determination method for semiconductor process equipment according to various embodiments of the present specification described in the above "Exemplary Method" section of the present specification.
[0172] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this specification can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0173] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0174] The above-mentioned embodiments only express several implementation methods of this specification, and the descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the solutions provided by the embodiments of this specification. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of this specification, which all belong to the protection scope of this specification. Therefore, the protection scope of the patent of this specification shall be based on the attached claims.
Claims
1. A method for determining a wafer scheduling sequence for a semiconductor process equipment, wherein the semiconductor process equipment comprises a plurality of chambers and a robot for transferring wafers between chambers, wherein the chambers comprise a process chamber and a non-process chamber, wherein the process chamber comprises at least one multi-station rotary process chamber, wherein: include: Acquire a current state of a target scheduling model of the semiconductor process equipment; The target scheduling model includes a plurality of places and a plurality of transitions, each of the places respectively represents the available resources of the chamber or the robot, the wafer holding state of the robot and the processing state of the chamber, and the transition represents the action performed by the robot; the state of the target scheduling model includes the number of tokens of each place and the position vector of the multi-station rotary process chamber, and the position vector includes the number of wafers in each station of the multi-station rotary process chamber; the number of tokens of each place respectively represents the number of available resources of the chamber or the robot, the number of wafers held by the robot and the number of wafers in the chamber; When the current state represents that the wafer scheduling sequence is not determined to be completed, the target data corresponding to each candidate transition is obtained, the target data including the subsequent state of the target scheduling model and the return value of the position vector after the candidate transition is executed in the current state; the candidate transition includes at least part of a plurality of the transitions, the return value is generated when the position vector changes, the return value includes the value of the target bit of the position vector before the change, and the target bit represents the station for taking wafers in the multi-station rotary process chamber; A target transition at a current moment is determined based on each of the target data, and the target transition is added to the wafer scheduling sequence, wherein the target transition includes a preferred execution action of the robot at the current moment.
2. The method according to claim 1, characterized in that The step of obtaining target data corresponding to each candidate transition includes: Based on the connection direction between the candidate transition and the target place, updating the number of tokens of the target place, the target place includes the place connected to the candidate transition; the connection direction between the candidate transition and the target place is used to determine the direction of change of the number of tokens of the target place when the candidate transition is executed; When the target location includes a location corresponding to the multi-station rotary process chamber, based on the current state of the target scheduling model and the robot action corresponding to the candidate transition, the position vector of the multi-station rotary process chamber is updated and the return value is generated.
3. The method according to claim 2, characterized in that The updating of the number of tokens of the target place based on the connection direction between the candidate transition and the target place includes: When the connection direction between the candidate transition and the target library place is directed from the candidate transition to the target library place, the number of tokens of the target library place is increased by a first preset value; the direction from the candidate transition to the target library place indicates that when the robot action corresponding to the candidate transition is executed, the number of resources corresponding to the target library released is the first preset value, the number of wafers held by the robot is increased by the first preset value, or the number of wafers processed by the chamber is increased by the first preset value; or When the connection direction between the candidate transition and the target library is from the target library to the candidate transition, the number of tokens of the target library is reduced by the first preset value; the target library pointing to the candidate transition indicates that when the robot action corresponding to the candidate transition is executed, the number of resources corresponding to the target library occupied is the first preset value, the number of wafers held by the robot is reduced by the first preset value, or the number of wafers processed by the chamber is reduced by the first preset value.
4. The method according to claim 2, characterized in that: The updating of the position vector of the multi-station rotary process chamber and the generation of the return value based on the current state of the target scheduling model and the robot action corresponding to the candidate transition include: Determining a current working state of the multi-station rotary process chamber based on a current state of the target scheduling model, wherein the working state of the multi-station rotary process chamber includes an entry state, a stable state, and an exit state; Based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber, the position vector of the multi-station rotary process chamber is updated and the return value is generated.
5. The method according to claim 4, characterized in that The updating of the position vector of the multi-station rotary process chamber and the generation of the return value based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber comprises: When the current working state of the multi-station rotary process chamber is an entering state or a stable state, taking the current value of the target position in the position vector as a return value; Based on the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber, an updated value of the position vector is determined.
6. The method according to claim 5, characterized in that The determining the updated value of the position vector based on the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber includes: If the robot action corresponding to the candidate transition includes placing a wafer into the multi-station rotary process chamber, the current value of the target position is deleted, and the number of wafers in each station other than the target position in the position vector is translated based on the direction of the target position, and the number of wafers in the station where the number of wafers in the position vector is empty is set to a second preset value; If the robot action corresponding to the candidate transition includes taking a wafer from the multi-station rotary process chamber, the value of the target bit is updated to 0.
7. The method according to claim 4, characterized in that The updating of the position vector of the multi-station rotary process chamber and the generation of the return value based on the current working state of the multi-station rotary process chamber, the robot action corresponding to the candidate transition and the current value of the position vector of the multi-station rotary process chamber comprises: When the current working state of the multi-station rotary process chamber is an exit state and the robot action corresponding to the candidate transition includes taking a sheet from the multi-station rotary process chamber, obtaining the current value of the target position; If the current value of the target bit is greater than 0, the current value of the target bit is used as the return value, and the value of the target bit is updated to 0; If the current value of the target position is 0, the shift operation is repeatedly performed until the current value of the target position is greater than 0; the shift operation includes deleting the current value of the target position, and translating the number of chips in each station other than the target position in the position vector based on the direction of the target position, and setting the number of chips in the station where the number of chips in the position vector is empty to 0.
8. The method according to claim 1, characterized in that The obtaining of the current state of the target scheduling model of the semiconductor process equipment includes: When the current state is the initial state, the number of tokens in each of the libraries and the position vector of the multi-station rotary process chamber are determined based on the current operating state of the semiconductor process equipment and the configuration information of the chamber and the robot.
9. The method according to any one of claims 1 to 8, characterized in that: The determining the target change at the current moment based on each of the target data includes: If the number of tokens in each of the places and the position vector do not have a value less than 0, and the return value meets the target value corresponding to the candidate transition, the candidate transition is used as an enabling transition, and the enabling transition represents an executable action of the manipulator at the current moment; When there are multiple enabling transitions, the target transition is determined from the enabling transitions based on the current states of the chambers and / or manipulators corresponding to the libraries connected to the enabling transitions.
10. An electronic device, characterized in that: The method comprises at least one processor and at least one memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method for determining a wafer scheduling sequence for semiconductor process equipment according to any one of claims 1 to 9 is implemented.
Citation Information
Patent Citations
Production scheduling method and device, semiconductor processing equipment and storage medium
CN110895733A
Wafer scheduling method and device, semiconductor processing equipment and storage medium
CN111952211A
Wafer scheduling method and electronic equipment
CN115794506A
A method and device for optimized dispatch of wafer
WO2009089715A1
Wafer dispatching method and electronic device
WO2024088294A1
