Scheduling control method for wafer manufacturing with multiple processing modules involving multiple agents

Abstract

The multi-processing module wafer manufacturing scheduling control method with multi-agents includes the following steps: S1. A mathematical model with the maximum completion time, total workshop energy consumption and minimum wafer scrap rate as optimization objectives is constructed; S2. First, the parameters of wafer processing are initialized; S3. When the wafer to be processed arrives at the entrance and exit, the wafer handling area scheduling algorithm and the cooling / temporary storage area scheduling algorithm are executed; S4. For the completed wafer, the process processing area scheduling algorithm is executed; S5. The wafer handling scheduling algorithm and the cooling / temporary storage area scheduling algorithm are executed, and the wafer is placed in the cooling / temporary storage area for cooling. After cooling, it is a cooled wafer, and the wafer handling area scheduling algorithm is executed, and the atmospheric manipulator moves the cooled wafer to the entrance and exit. The scheme optimizes the maximum completion time, total workshop energy consumption, scrap rate and the control strategy of the manipulator at the same time.

Description

Technical Field

[0001] The invention relates to wafer production scheduling technology, in particular to a wafer manufacturing scheduling control method containing multiple processing modules and multiple intelligent agents. Background Art

[0002] In a wafer manufacturing workshop, there are a number of independent workpieces that need to be processed on multiple machines. Each workpiece only needs to go through one process, but this process will go through multiple different stages, each with different processing time and power requirements. The processed workpieces need to be cooled in the cooling zone, and the capacity of the cooling zone is limited. The workshop is equipped with two types of manipulators: an atmospheric manipulator and two vacuum manipulators. The atmospheric manipulator is responsible for moving the workpieces to be processed from the entrance to the vacuum manipulator, and the vacuum manipulator then moves the workpieces to the machine for processing. After the processing is completed, the same vacuum manipulator will move the workpiece to the cooling zone, and the atmospheric manipulator will finally move the cooled workpiece to the exit. It should be noted that the processing of the machine and the insulation during the unprocessed time will cause power loss; the capacity of the cooling zone is limited; at any time, the instantaneous total power of the workshop cannot exceed the maximum rated power; in addition, if the workpiece is not moved away by the vacuum manipulator within the specified time after the workpiece is processed on the machine, it will affect the quality of the workpiece and the workpiece will be considered as scrap. Therefore, it is necessary to find a suitable scheduling scheme to meet the solution of multiple objectives under many constraints.

[0003] During the wafer manufacturing process, the control of the robot also has a significant impact on the scheduling efficiency, total energy consumption of the workshop and wafer quality; considering the multiple goals of maximum completion time, total energy consumption of the workshop and scrap rate, and embedding the scheduling plan, it is also necessary to combine the interaction characteristics of the robot and the back-end area, and optimize the entire wafer manufacturing scheduling problem with the robot control strategy to further improve the effectiveness of the scheduling plan. Summary of the invention

[0004] In view of the above-mentioned defects, the purpose of the present invention is to propose a multi-processing module wafer manufacturing scheduling control method containing multiple intelligent agents, which simultaneously optimizes the maximum completion time, total energy consumption of the workshop, scrap rate and control strategy of the robot.

[0005] To achieve this object, the present invention adopts the following technical solutions:

[0006] A multi-agent multi-processing module wafer manufacturing scheduling control method includes the following steps:

[0007] S1. Aiming at the production scenario, production process and problem characteristics of multi-processing module wafer manufacturing scheduling problem, a mathematical model with the maximum completion time, the total energy consumption of the workshop and the minimum wafer scrap rate as the optimization objectives was constructed;

[0008] The physical modules of the workshop are the temporary storage / cooling area, wafer handling area and process processing area from left to right. The temporary storage / cooling area is mainly used to temporarily store wafers to be processed and cool completed wafers. It can accommodate 7 wafers at the same time; the wafer handling area is composed of 1 atmospheric manipulator and 2 vacuum manipulators, which are mainly responsible for handling wafers; the process processing area is composed of 6 processing stations for processing wafers;

[0009] S2. First, initialize the parameters of wafer processing, and then input the production and processing plan including wafer processing sorting and processing station production sequence; respectively propose corresponding algorithms to solve the three areas of the workshop: the robot in the wafer handling area, the robot in the process processing area, and the robot in the cooling / temporary storage area;

[0010] S3. When the wafer to be processed arrives at the entrance and exit, the wafer handling area scheduling algorithm and the cooling / temporary storage area scheduling algorithm are executed to determine whether there are idle processing stations and vacuum manipulators that interact with them. After predicting that the processing power of the wafer meets the maximum instantaneous power constraint, the wafer is transported to the corresponding processing station for processing according to the processing station production sequence;

[0011] S4. For the finished wafers, the process area scheduling algorithm is executed to sort the finished wafers in the process area, and it is determined that the atmospheric manipulator is in an idle state and the capacity of the cooling area is greater than 1, and the atmospheric manipulator moves the wafers to the cooling area for cooling;

[0012] S5. Execute the wafer handling scheduling algorithm and / or the cooling temporary storage area scheduling algorithm, place the wafer into the cooling / temporary storage area for cooling, and after cooling, the wafer is cooled. Execute the wafer handling area scheduling algorithm, and the atmospheric manipulator moves the cooled wafer to the entrance and exit.

[0013] In step S2, in the pull control of the atmospheric manipulator and the vacuum manipulator, a blocking list of the manipulator is first obtained, and the blocking list records the wafers to be transported and the input time; by designing the control strategy of the manipulator, the wafers in the blocking list are driven by the transport task to complete the wafer transport;

[0014] For the control strategy of the wafer handling area, under the premise of avoiding deadlock and the maximum instantaneous power constraint of the workshop, five robot control algorithms are proposed by comprehensively considering the handling efficiency of the robot, the production efficiency of the workshop and the energy consumption, which are:

[0015] The priority rule for the least remaining process, the priority rule for the earliest process that can be transported, the scheduled production plan sequence, the first-come-first-served rule, and the random transport rule.

[0016] In step S3, for the robot in the process area, considering the completion time and residence time constraints of the completed wafers, three robot control priority strategies are proposed to optimize the quality of the wafers;

[0017] Shortest remaining dwell time priority rule:

[0018] The shortest remaining residence time priority rule is to calculate the remaining residence time of all completed wafers in the process area and select the completed wafer with the shortest remaining residence time for transportation; the specific steps are as follows:

[0019] S31a, initializing the completed wafer set LC, adding all the processed wafers in the process area into the completed wafer set LC in the order of numbers;

[0020] S31b, if By calculating the remaining residence time (residence time minus waiting time) of each wafer in the set, a wafer remaining residence time set O is obtained, where the element numbers are 1, 2, ..., w.

[0021] S31c, find the wafer with the shortest remaining dwell time: o w =Min{o1,o2,...o w}, o w Remove from set O, the corresponding interacting vacuum manipulator selects and grabs wafer w, and removes wafer w from set L;

[0022] Earliest completion time priority rule:

[0023] The earliest completion time priority rule means that according to the order of completion time of the completed wafers in the process processing area, the completed wafers with the earliest completion time are selected for priority transportation. The specific steps are as follows:

[0024] S32a, initializing the completed wafer set LC, and adding all the processed wafers in the process area to the completed wafer set LC in the order of completion time;

[0025] S32b, if Then select the first element in the set, which represents the wafer, and remove it from the set L;

[0026] Random rules:

[0027] The random rule means randomly selecting a finished wafer in a process area for transport; the specific steps are as follows:

[0028] S33a, initializing the completed wafer set LC, adding all the processed wafers in the process area to the completed wafer set LC in the order of numbers, where the element numbers are represented by 1, 2, ...n;

[0029] S33b, if Then randomly select a completed wafer: w = Rand ({1, 2, ... n}), the random method Rand is used to randomly select an element in a specified set; select and grab wafer w, and remove wafer w from the completed wafer set LC.

[0030] In step S4, for the manipulator in the cooling / temporary storage area, considering that the number of wafers that can be accommodated in this area at the same time is limited, which causes problems such as resource preemption, two control rules are proposed;

[0031] Wafer temporary storage optimization algorithm refers to whether to let the robot give priority to moving the wafers to be processed to the cooling / temporary storage area for temporary storage based on the remaining capacity of the current cooling / temporary storage area; the specific steps are as follows:

[0032] S41a, obtaining the current remaining capacity of the cooling / temporary storage area in real time;

[0033] S41b, if it is detected that the remaining capacity is greater than or equal to a preset threshold, a temporary storage mechanism for the wafer to be processed is triggered, and the wafer is immediately placed in a cooling / temporary storage area to optimize the utilization rate of the robot;

[0034] S41c, after completing the temporary storage operation, the algorithm ends the current cycle and continues to execute the subsequent production process;

[0035] Wafer cooling priority algorithm: Wafer cooling priority algorithm means that according to the remaining capacity of the cooling / temporary storage area and the processing status of the wafers in the process processing area, the robot will prioritize the transportation of finished wafers to the cooling / temporary storage area for cooling; the specific steps are as follows:

[0036] S42a, obtaining the current remaining capacity of the cooling / temporary storage area in real time, and identifying the processing status of the wafers in the process area;

[0037] S42b, if there is remaining capacity and a finished wafer is detected, the cooling priority mechanism is activated to reserve space for the finished wafer and ensure that it enters the cooling / temporary storage area for cooling first, during which the entry of the wafer to be processed is suspended to ensure the effective use of cooling resources;

[0038] S42c. Once the finished wafer successfully enters the cooling state, the algorithm ends the current processing cycle and resumes the normal production process.

[0039] The five manipulator control algorithms are as follows:

[0040] S21. The least remaining process priority rule:

[0041] The rule of giving priority to the wafer with the least remaining process is to obtain the current process number of each wafer to be transported, calculate the remaining process number, and select the wafer with the least remaining process number for priority transport; the specific steps are as follows:

[0042] S21a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number;

[0043] S21b, if Calculate the number of remaining processes for each wafer in the set, the difference between the total number of processes and the current number of processes; combined with the previous description of wafers in different states, it can be seen that the number of remaining processes for cooled wafers is the least, followed by completed wafers, and the number of wafers to be processed is the most, and the set of wafer remaining process numbers O is obtained, and the element numbers are represented by 1, 2, ...w;

[0044] S21c, find the wafer with the least number of remaining steps: o w =Min{o1,o2,...o w}, if there are multiple wafers with the least number of remaining steps, randomly select one of the wafers with the least number of remaining steps and place o w Remove from the set O, select and grab wafer w, and remove wafer w from the set of wafers to be transported LM;

[0045] S21d, holding and waiting, judging whether relevant conditions are met, the wafer to be processed needs to meet the processable conditions, and the completed wafer needs to meet the cooling conditions. If they meet, they are moved to the subsequent corresponding equipment, otherwise they continue to hold and wait;

[0046] S22. The earliest possible transport priority rule:

[0047] The earliest available wafer priority rule refers to selecting the earliest available wafer for priority transport by judging and calculating the available transport time of each wafer to be transported according to the operating status of each device; the specific steps are as follows:

[0048] S22a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number;

[0049] S22b, if By judging and calculating the time when each wafer in the set can be transported, a time set O of the wafer can be transported is obtained, and the element numbers are represented by 1, 2, ...w;

[0050] S22c, find the earliest wafer that can be transported: o w =Min{o1,o2,...o w}, o w Remove from set O, select and grab wafer w, and remove wafer w from set LM;

[0051] S22d, transporting the wafer w to subsequent corresponding equipment;

[0052] S23, scheduled production schedule sequence:

[0053] The scheduled production schedule sequence refers to the wafer scheduling plan according to the scheduled production schedule sequence proposed above, and the wafers in the scheduling plan are transported to the corresponding processing modules for processing in an orderly manner; the specific steps are as follows:

[0054] S23a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number;

[0055] S23b, if Obtain the wafer processing sequence set O of the predetermined production schedule sequence and the processing wafer sequence set M of each processing station, arrange the set LM according to the sequence of the set O, select the first element in the set LM, the element represents the wafer, and remove it from the set LM;

[0056] S23c, holding and waiting, judging whether relevant conditions are met according to the wafer processing status, the wafer to be processed needs to meet the processable condition, and the completed wafer needs to meet the cooling condition. If they meet, they are transported to the subsequent corresponding equipment. For the wafer to be processed, it needs to be transported to the designated processing station for processing by the robot according to the set M, otherwise it continues to hold and wait until the corresponding conditions are met and then it is transported to the subsequent corresponding equipment;

[0057] S24. First-come, first-served rule:

[0058] The first-come-first-served rule is to select the wafer that first enters the robot blocking list for priority handling according to the time sequence of the wafers to be transported entering the blocking list; the specific steps are as follows:

[0059] S24a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in chronological order;

[0060] S24b, if Then select the first element in the set, which represents the wafer, and remove it from the set of wafers to be moved LM;

[0061] S24c, holding and waiting, judging whether relevant conditions are met according to the wafer type, the wafer to be processed needs to meet the processable condition, and the completed wafer needs to meet the cooling condition. If they meet, they are transported to the subsequent corresponding equipment, otherwise they continue to hold and wait until the corresponding conditions are met and then they are transported to the subsequent corresponding equipment;

[0062] S25. Random handling rules:

[0063] The random handling rule means randomly selecting a wafer to be handled from the robot blocking list for priority handling; the specific steps are as follows:

[0064] S25a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set to be transported in order of number, where the element numbers are represented by 1, 2, ...n;

[0065] S25b, if Then randomly select the wafer to be transported: w = Rand ({1, 2, ... n}), the random method Rand is used to randomly select an element in a specified set; select and grab wafer w, and remove wafer w from the set of wafers to be transported LM;

[0066] Waiting for the part to be held. Judging whether the relevant conditions are met based on the wafer processing status. The wafer to be processed must meet the processable conditions, and the completed wafer must meet the cooling conditions. If they meet the conditions, they will be moved to the subsequent corresponding equipment. Otherwise, they will continue to wait for the part to be held until the corresponding conditions are met and then they will be moved to the subsequent corresponding equipment.

[0067] One of the above technical solutions includes the following beneficial effects:. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is a schematic diagram of a workshop entity module of an embodiment of the present invention;

[0069] Figure 2 It is a multi-agent multi-processing module wafer manufacturing scheduling control algorithm framework of an embodiment of the present invention;

[0070] Figure 3 It is a multi-agent multi-processing module wafer manufacturing scheduling control algorithm flow of an embodiment of the present invention. DETAILED DESCRIPTION

[0071] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0072] A multi-agent multi-processing module wafer manufacturing scheduling control method includes the following steps:

[0073] S1. Aiming at the production scenario, production process and problem characteristics of multi-processing module wafer manufacturing scheduling problem, a mathematical model with the maximum completion time, the total energy consumption of the workshop and the minimum wafer scrap rate as the optimization objectives was constructed;

[0074] The physical modules of the workshop are the temporary storage / cooling area, wafer handling area and process processing area from left to right. The temporary storage / cooling area is mainly used to temporarily store wafers to be processed and cool completed wafers. It can accommodate 7 wafers at the same time; the wafer handling area is composed of 1 atmospheric manipulator and 2 vacuum manipulators, which are mainly responsible for handling wafers; the process processing area is composed of 6 processing stations for processing wafers;

[0075] S2. First, initialize the parameters of wafer processing, and then input the production and processing plan including wafer processing sorting and processing station production sequence; respectively propose corresponding algorithms to solve the three areas of the workshop: the robot in the wafer handling area, the robot in the process processing area, and the robot in the cooling / temporary storage area;

[0076] S3. When the wafer to be processed arrives at the entrance and exit, the wafer handling area scheduling algorithm and the cooling / temporary storage area scheduling algorithm are executed to determine whether there are idle processing stations and vacuum manipulators that interact with them. After predicting that the processing power of the wafer meets the maximum instantaneous power constraint, the wafer is transported to the corresponding processing station for processing according to the processing station production sequence;

[0077] S4. For the finished wafers, the process area scheduling algorithm is executed to sort the finished wafers in the process area, and it is determined that the atmospheric manipulator is in an idle state and the capacity of the cooling area is greater than 1, and the atmospheric manipulator moves the wafers to the cooling area for cooling;

[0078] S5. Execute the wafer handling scheduling algorithm and / or the cooling temporary storage area scheduling algorithm, place the wafer into the cooling / temporary storage area for cooling, and after cooling, the wafer is cooled. Execute the wafer handling area scheduling algorithm, and the atmospheric manipulator moves the cooled wafer to the entrance and exit.

[0079] Among them, in step S2, in the pull control of the atmospheric manipulator and the vacuum manipulator, a blocking list of the manipulator is first obtained, and the blocking list records the wafers to be transported and the input time; by designing the control strategy of the manipulator, the wafers in the blocking list are driven by the transport task to complete the wafer transport;

[0080] For the control strategy of the wafer handling area, under the premise of avoiding deadlock and the maximum instantaneous power constraint of the workshop, five robot control algorithms are proposed by comprehensively considering the handling efficiency of the robot, the production efficiency of the workshop and the energy consumption, which are:

[0081] The priority rule for the least remaining process, the priority rule for the earliest process that can be transported, the scheduled production plan sequence, the first-come-first-served rule, and the random transport rule.

[0082] Among them, in step S3, for the robot in the process area, considering the completion time and residence time constraints of the completed wafers, three robot control priority strategies are proposed to optimize the quality of the wafers;

[0083] Shortest remaining dwell time priority rule:

[0084] The shortest remaining residence time priority rule is to calculate the remaining residence time of all completed wafers in the process area and select the completed wafer with the shortest remaining residence time for transportation; the specific steps are as follows:

[0085] S31a, initializing the completed wafer set LC, adding all the processed wafers in the process area into the completed wafer set LC in the order of numbers;

[0086] S31b, if By calculating the remaining residence time (residence time minus waiting time) of each wafer in the set, a wafer remaining residence time set O is obtained, where the element numbers are 1, 2, ..., w;

[0087] S31c, find the wafer with the shortest remaining dwell time: o w =Min{o1,o2,...o w}, o w Remove from set O, the corresponding interacting vacuum manipulator selects and grabs wafer w, and removes wafer w from set L;

[0088] Earliest completion time priority rule:

[0089] The earliest completion time priority rule means that according to the order of completion time of the completed wafers in the process processing area, the completed wafers with the earliest completion time are selected for priority transportation. The specific steps are as follows:

[0090] S32a, initializing the completed wafer set LC, and adding all the processed wafers in the process area to the completed wafer set LC in the order of completion time;

[0091] S32b, if Then select the first element in the set, which represents the wafer, and remove it from the set L;

[0092] Random rules:

[0093] The random rule means randomly selecting a finished wafer in a process area for transport; the specific steps are as follows:

[0094] S33a, initializing the completed wafer set LC, adding all the processed wafers in the process area to the completed wafer set LC in the order of numbers, where the element numbers are represented by 1, 2, ...n;

[0095] S33b, if Then randomly select a completed wafer: w = Rand ({1, 2, ... n}), the random method Rand is used to randomly select an element in a specified set; select and grab wafer w, and remove wafer w from the completed wafer set LC.

[0096] Among them, in step S4, for the manipulator in the cooling / temporary storage area, considering that the number of wafers that can be accommodated in this area at the same time is limited, which causes problems such as resource preemption, two control rules are proposed;

[0097] Wafer temporary storage optimization algorithm refers to whether to let the robot give priority to moving the wafers to be processed to the cooling / temporary storage area for temporary storage based on the remaining capacity of the current cooling / temporary storage area; the specific steps are as follows:

[0098] S41a, obtaining the current remaining capacity of the cooling / temporary storage area in real time;

[0099] S41b, if it is detected that the remaining capacity is greater than or equal to a preset threshold, a temporary storage mechanism for the wafer to be processed is triggered, and the wafer is immediately placed in a cooling / temporary storage area to optimize the utilization rate of the robot;

[0100] S41c, after completing the temporary storage operation, the algorithm ends the current cycle and continues to execute the subsequent production process;

[0101] Wafer cooling priority algorithm: Wafer cooling priority algorithm means that according to the remaining capacity of the cooling / temporary storage area and the processing status of the wafers in the process processing area, the robot will prioritize the transportation of finished wafers to the cooling / temporary storage area for cooling; the specific steps are as follows:

[0102] S42a, obtaining the current remaining capacity of the cooling / temporary storage area in real time, and identifying the processing status of the wafers in the process area;

[0103] S42b, if there is remaining capacity and a finished wafer is detected, the cooling priority mechanism is activated to reserve space for the finished wafer and ensure that it enters the cooling / temporary storage area for cooling first, during which the entry of the wafer to be processed is suspended to ensure the effective use of cooling resources;

[0104] S42c. Once the finished wafer successfully enters the cooling state, the algorithm ends the current processing cycle and resumes the normal production process.

[0105] In addition, the five manipulator control algorithms are as follows:

[0106] S21. The least remaining process priority rule:

[0107] The rule of giving priority to the wafer with the least remaining process is to obtain the current process number of each wafer to be transported, calculate the remaining process number, and select the wafer with the least remaining process number for priority transport; the specific steps are as follows:

[0108] S21a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number;

[0109] S21b, if Calculate the number of remaining processes for each wafer in the set, the difference between the total number of processes and the current number of processes; combined with the previous description of wafers in different states, it can be seen that the number of remaining processes for cooled wafers is the least, followed by completed wafers, and the number of wafers to be processed is the most, and the set of wafer remaining process numbers O is obtained, and the element numbers are represented by 1, 2, ...w;

[0110] S21c, find the wafer with the least number of remaining steps: o w =Min{o1,o2,...o w}, if there are multiple wafers with the least number of remaining steps, randomly select one of the wafers with the least number of remaining steps and place o w Remove from the set O, select and grab wafer w, and remove wafer w from the set of wafers to be transported LM;

[0111] S21d, holding and waiting, judging whether relevant conditions are met, the wafer to be processed needs to meet the processable conditions, and the completed wafer needs to meet the cooling conditions. If they meet, they are moved to the subsequent corresponding equipment, otherwise they continue to hold and wait;

[0112] S22. The earliest possible transport priority rule:

[0113] The earliest available wafer priority rule refers to selecting the earliest available wafer for priority transport by judging and calculating the available transport time of each wafer to be transported according to the operating status of each device; the specific steps are as follows:

[0114] S22a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number;

[0115] S22b, if By judging and calculating the time when each wafer in the set can be transported, a time set O of the wafer can be transported is obtained, where the element numbers are represented by 1, 2, ..., w;

[0116] S22c, find the earliest wafer that can be transported: o w =Min{o1,o2,...o w}, o w Remove from set O, select and grab wafer w, and remove wafer w from set LM;

[0117] S22d, transporting the wafer w to subsequent corresponding equipment;

[0118] S23, scheduled production schedule sequence:

[0119] The scheduled production schedule sequence refers to the wafer scheduling plan according to the scheduled production schedule sequence proposed above, and the wafers in the scheduling plan are transported to the corresponding processing modules for processing in an orderly manner; the specific steps are as follows:

[0120] S23a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number;

[0121] S23b, if Obtain the wafer processing sequence set O of the predetermined production schedule sequence and the processing wafer sequence set M of each processing station, arrange the set LM according to the sequence of the set O, select the first element in the set LM, the element represents the wafer, and remove it from the set LM;

[0122] S23c, holding and waiting, judging whether relevant conditions are met according to the wafer processing status, the wafer to be processed needs to meet the processable condition, and the completed wafer needs to meet the cooling condition. If they meet, they are transported to the subsequent corresponding equipment. For the wafer to be processed, it needs to be transported to the designated processing station for processing by the robot according to the set M, otherwise it continues to hold and wait until the corresponding conditions are met and then it is transported to the subsequent corresponding equipment;

[0123] S24. First-come, first-served rule:

[0124] The first-come-first-served rule is to select the wafer that first enters the robot blocking list for priority handling according to the time sequence of the wafers to be transported entering the blocking list; the specific steps are as follows:

[0125] S24a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in chronological order;

[0126] S24b, if Then select the first element in the set, which represents the wafer, and remove it from the set of wafers to be moved LM;

[0127] S24c, holding and waiting, judging whether relevant conditions are met according to the wafer type, the wafer to be processed needs to meet the processable condition, and the completed wafer needs to meet the cooling condition. If they meet, they are transported to the subsequent corresponding equipment, otherwise they continue to hold and wait until the corresponding conditions are met and then they are transported to the subsequent corresponding equipment;

[0128] S25. Random handling rules:

[0129] The random handling rule means randomly selecting a wafer to be handled from the robot blocking list for priority handling; the specific steps are as follows:

[0130] S25a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set to be transported in order of number, where the element numbers are represented by 1, 2, ...n;

[0131] S25b, if Then randomly select the wafer to be transported: w = Rand ({1, 2, ... n}), the random method Rand is used to randomly select an element in a specified set; select and grab wafer w, and remove wafer w from the set of wafers to be transported LM;

[0132] Waiting for the part to be held. Judging whether the relevant conditions are met based on the wafer processing status. The wafer to be processed must meet the processable conditions, and the completed wafer must meet the cooling conditions. If they meet the conditions, they will be moved to the subsequent corresponding equipment. Otherwise, they will continue to wait for the part to be held until the corresponding conditions are met and then they will be moved to the subsequent corresponding equipment.

[0133] This paper comprehensively considers three key performance indicators in the wafer manufacturing process: maximum completion time, total energy consumption of the workshop, and wafer scrap rate. In order to select the best algorithm from the proposed composite rule algorithms, this paper uses the relative deviation index (RDI) as the evaluation criterion. Specifically, this paper compares the target values ​​obtained by each combination rule algorithm in each specific instance. The calculation formula of the relative deviation index is shown below.

[0134]

[0135] RDI in formula (4.1) PA It represents the relative deviation value of the combined rule algorithm A under the example P, FO IA represents the result of the combined rule algorithm A under the example P, Best P and Worst P They represent the optimal and worst results of all combination rule algorithms under example P respectively.

[0136] Experimental results analysis:

[0137] Based on the maximum completion time as the evaluation indicator:

[0138] After experimental verification, this paper found that the average relative deviation index (RDI) of the combined algorithm LSH-EF-WHOA was 0.1763 after 10 repeated experiments under 30 different combined algorithms. This value is the minimum value in all test cases. This result fully proves the superiority of the LSH-EF-WHOA combined rule algorithm in performance. It has the best comprehensive performance under the maximum completion time as the evaluation index; among them: Least Operations Remaining Handling (LORH), First Available Handling (FAH), LNSGA-Ⅱ Sequence Handling (LSH), First Come First Handling (FCFH), Random Handling (RH); Among them, LSH has no scheduled production schedule sequence, and the scheduled production schedule sequence is obtained by LNSGAⅡ algorithm, so LSH is used in the table to refer to the scheduled production schedule sequence.

[0139] Table 4-2 Average overall RDI based on maximum completion time

[0140]

[0141] As shown in Table 4-2, in order to analyze the performance of different combination algorithms under the maximum completion time indicator, it can be seen from Table 4-2 that: 1) Among the five wafer handling area control algorithms compared and analyzed, the LSH algorithm shows superior performance in the combination rules with other areas. The second is the FAH algorithm, which mainly controls the robot to handle wafers by judging the appropriate handling time, greatly reducing the waiting time. This rule effectively controls the wafer handling operation of the robot by judging the rationality of the handling time, thereby significantly reducing the waiting time and improving the overall operation efficiency. 2) The RDI average value of the FAH algorithm in the six process processing area and cooling / temporary storage area combination algorithms discussed in this article is relatively close. Compared with the LSH algorithm, its dependence on the process processing area and cooling / temporary storage area combination rule algorithm is not significant. 3) Among the three process processing area robot control algorithms, the EF rule algorithm performs best, and its optimization effect on the key performance indicator of maximum completion time is particularly significant.

[0142] Based on the total energy consumption of the workshop as the evaluation index:

[0143] Table 4-3 shows the comprehensive performance of each combination rule algorithm based on the evaluation index of total energy consumption of the workshop. Among the five wafer handling area robot control algorithms, the LSH rule also performs well under the indicator of total energy consumption of the workshop, mainly due to its combination with the LNSGA-Ⅱ intelligent algorithm, both of which have achieved significant optimization effects in terms of energy consumption. The second best performance is the FAH rule. After finding the earliest wafer that can be handled, this rule algorithm controls the robot to carry it, greatly reducing the idle time of the machine, thereby reducing the idle energy consumption of the machine and further improving the optimization level of the total energy consumption of the workshop.

[0144] Table 4-3 Average Overall RDI Based on Total Energy Consumption Of Workshop

[0145]

[0146] Evaluation indicators based on wafer scrap rate:

[0147] As shown in Table 4-4, in order to analyze the overall performance of different combination algorithms under the maximum completion time indicator, by observing the data in Table 4-4, it can be found that: 1) The five wafer handling area robot control rule algorithms do not show significant differences in their RDI average values ​​when combined with different process processing area and cooling / temporary storage area combination rule algorithms, which shows that under the evaluation indicator of wafer scrap rate, the performance of the wafer handling area robot control rule algorithm is not highly significant. 2) Among the three process processing area robot control rule algorithms, the SRDT rule sorts the remaining residence time of the completed wafers on the processing station and preferentially transports the wafers with the shortest remaining residence time out of the process processing area. This priority strategy significantly improves the quality of the wafer, thereby effectively reducing the wafer scrap rate. 3) Among the six process processing area and cooling / temporary storage area combination robot control rule algorithms, the SRDT-WCPA combination rule performs well. Especially when combined with the FAH wafer handling area robot control rule algorithm, the FAH-SRDT-WCPA combination algorithm performed outstandingly among 30 different combination algorithms, and its average RDI reached the minimum value of 0.3413. This result fully proves the superiority of the FAH-SRDT-WCPA combination rule algorithm in performance, especially when the wafer scrap rate is used as the evaluation indicator, its comprehensive performance is the best.

[0148] Table 4-4 Average Overall RDI Based on Wafer Scrap Rate

[0149]

[0150] The technical principle of the present invention is described above in conjunction with specific embodiments. These descriptions are only for explaining the principle of the present invention and cannot be interpreted as limiting the scope of protection of the present invention in any way. Based on the explanations herein, those skilled in the art can associate other specific implementations of the present invention without paying creative labor, and these methods will fall within the scope of protection of the present invention.

Claims

1. A multi-agent multi-processing module wafer manufacturing scheduling control method, characterized in that: The steps include: S1. Aiming at the production scenario, production process and problem characteristics of multi-processing module wafer manufacturing scheduling problem, a mathematical model with the maximum completion time, the total energy consumption of the workshop and the minimum wafer scrap rate as the optimization objectives was constructed; The physical modules of the workshop are the temporary storage / cooling area, wafer handling area and process processing area from left to right. The temporary storage / cooling area is used to temporarily store wafers to be processed and cool completed wafers, and can accommodate 7 wafers at the same time; the wafer handling area is composed of 1 atmospheric manipulator and 2 vacuum manipulators, which are responsible for handling wafers; the process processing area is composed of 6 processing stations, which are used to process wafers; S2. First, initialize the parameters of wafer processing, and then input the production and processing plan including wafer processing sorting and processing station production sequence; respectively propose corresponding algorithms to solve the three areas of the workshop: the robot in the wafer handling area, the robot in the process processing area, and the robot in the cooling / temporary storage area; S3. When the wafer to be processed arrives at the entrance and exit, the wafer handling area scheduling algorithm and the cooling / temporary storage area scheduling algorithm are executed to determine whether there are idle processing stations and vacuum manipulators that interact with them. After predicting that the processing power of the wafer meets the maximum instantaneous power constraint, the wafer is transported to the corresponding processing station for processing according to the processing station production sequence; S4. For the finished wafers, the process area scheduling algorithm is executed to sort the finished wafers in the process area, and it is determined that the atmospheric manipulator is in an idle state and the capacity of the cooling area is greater than 1, and the atmospheric manipulator moves the wafers to the cooling area for cooling; S5. Execute the wafer handling scheduling algorithm and / or the cooling temporary storage area scheduling algorithm, place the wafer into the cooling / temporary storage area for cooling, and after cooling, the wafer is cooled. Execute the wafer handling area scheduling algorithm, and the atmospheric manipulator moves the cooled wafer to the entrance and exit.

2. The multi-agent multi-processing module wafer manufacturing scheduling control method according to claim 1 is characterized in that: In step S2, in the pull control of the atmospheric manipulator and the vacuum manipulator, a blocking list of the manipulator is first obtained, and the blocking list records the wafers to be transported and the input time; by designing the control strategy of the manipulator, the wafers in the blocking list are driven by the transport task to complete the wafer transport; For the control strategy of the wafer handling area, under the premise of avoiding deadlock and the maximum instantaneous power constraint of the workshop, five robot control algorithms are proposed by comprehensively considering the handling efficiency of the robot, the production efficiency of the workshop and the energy consumption, which are: The priority rule for the least remaining process, the priority rule for the earliest process that can be transported, the scheduled production plan sequence, the first-come-first-served rule, and the random transport rule.

3. The multi-agent multi-processing module wafer manufacturing scheduling control method according to claim 1 is characterized in that: In step S3, for the robot in the process area, considering the completion time and residence time constraints of the completed wafers, three robot control priority strategies are proposed to optimize the quality of the wafers; Shortest remaining dwell time priority rule: The shortest remaining residence time priority rule is to calculate the remaining residence time of all completed wafers in the process area and select the completed wafer with the shortest remaining residence time for transportation; the specific steps are as follows: S31a, initializing the completed wafer set LC, adding all the processed wafers in the process area into the completed wafer set LC in the order of numbers; S31b, if By calculating the remaining residence time (residence time minus waiting time) of each wafer in the set, a wafer remaining residence time set O is obtained, where the element numbers are 1, 2, ..., w; S31c, find the wafer with the shortest remaining dwell time: o w =Min{o1,o2,...o w }, o w Remove from set O, the corresponding interacting vacuum manipulator selects and grabs wafer w, and removes wafer w from set L; Earliest completion time priority rule: The earliest completion time priority rule means that according to the order of completion time of the completed wafers in the process processing area, the completed wafers with the earliest completion time are selected for priority transportation. The specific steps are as follows: S32a, initializing the completed wafer set LC, and adding all the processed wafers in the process area to the completed wafer set LC in the order of completion time; S32b, if Then select the first element in the set, which represents the wafer, and remove it from the set L; Random rules: The random rule means randomly selecting a finished wafer in a process area for transport; the specific steps are as follows: S33a, initializing the completed wafer set LC, adding all the processed wafers in the process area to the completed wafer set LC in the order of numbers, where the element numbers are represented by 1, 2, ...n; S33b, if Then randomly select a completed wafer: w = Rand ({1, 2, ... n}), the random method Rand is used to randomly select an element in a specified set; select and grab wafer w, and remove wafer w from the completed wafer set LC.

4. The multi-agent multi-processing module wafer manufacturing scheduling control method according to claim 3 is characterized in that: In step S4, for the manipulator in the cooling / temporary storage area, considering that the number of wafers that can be accommodated in this area at the same time is limited, which causes the problem of resource preemption, two control rules are proposed; Wafer temporary storage optimization algorithm refers to whether to let the robot give priority to moving the wafers to be processed to the cooling / temporary storage area for temporary storage based on the remaining capacity of the current cooling / temporary storage area; the specific steps are as follows: S41a, obtaining the current remaining capacity of the cooling / temporary storage area in real time; S41b, if it is detected that the remaining capacity is greater than or equal to a preset threshold, the temporary storage mechanism of the wafer to be processed is triggered, and the wafer is immediately placed in a cooling / temporary storage area to optimize the utilization rate of the robot; S41c, after completing the temporary storage operation, the algorithm ends the current cycle and continues to execute the subsequent production process; Wafer cooling priority algorithm: Wafer cooling priority algorithm means that according to the remaining capacity of the cooling / temporary storage area and the processing status of the wafers in the process processing area, the robot will prioritize the transportation of finished wafers to the cooling / temporary storage area for cooling; the specific steps are as follows: S42a, obtaining the current remaining capacity of the cooling / temporary storage area in real time, and identifying the processing status of the wafers in the process processing area; S42b, if there is remaining capacity and a finished wafer is detected, the cooling priority mechanism is activated to reserve space for the finished wafer and ensure that it enters the cooling / temporary storage area for cooling first, during which the entry of the wafer to be processed is suspended to ensure the effective use of cooling resources; S42c, once the finished wafer successfully enters the cooling state, the algorithm ends the current processing cycle and resumes the normal production process.

5. The multi-agent multi-processing module wafer manufacturing scheduling control method according to claim 2 is characterized in that: The five manipulator control algorithms are as follows: S21. The least remaining process priority rule: The rule of giving priority to the wafer with the least remaining process is to obtain the current process number of each wafer to be transported, calculate the remaining process number, and select the wafer with the least remaining process number for priority transport; the specific steps are as follows: S21a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number; S21b, if Calculate the number of remaining processes for each wafer in the set, the difference between the total number of processes and the current number of processes; combined with the previous description of wafers in different states, it can be seen that the number of remaining processes for cooled wafers is the least, followed by completed wafers, and the number of wafers to be processed is the most, and the set of wafer remaining process numbers O is obtained, and the element numbers are represented by 1, 2, ...w; S21c, find the wafer with the least number of remaining steps: o w =Min{o1,o2,...o w }, if there are multiple wafers with the least number of remaining steps, randomly select one of the wafers with the least number of remaining steps and place o w Remove from the set O, select and grab wafer w, and remove wafer w from the set of wafers to be transported LM; S21d, holding and waiting, judging whether relevant conditions are met, the wafer to be processed needs to meet the processable conditions, and the completed wafer needs to meet the cooling conditions. If they meet, they are moved to the subsequent corresponding equipment, otherwise they continue to hold and wait; S22. The earliest possible transport priority rule: The earliest available wafer priority rule refers to selecting the earliest available wafer for priority transport by judging and calculating the available transport time of each wafer to be transported according to the operating status of each device; the specific steps are as follows: S22a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number; S22b, if By judging and calculating the time when each wafer in the set can be transported, a time set O of the wafer can be transported is obtained, where the element numbers are represented by 1, 2, ..., w; S22c, find the earliest wafer that can be transported: o w =Min{o1,o2,...o w }, o w Remove from set O, select and grab wafer w, and remove wafer w from set LM; S22d, transporting the wafer w to subsequent corresponding equipment; S23, scheduled production schedule sequence: The scheduled production schedule sequence refers to the wafer scheduling plan according to the scheduled production schedule sequence proposed above, and the wafers in the scheduling plan are transported to the corresponding processing modules for processing in an orderly manner; the specific steps are as follows: S23a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in order of number; S23b, if Obtain the wafer processing sequence set O of the predetermined production schedule sequence and the processing wafer sequence set M of each processing station, arrange the set LM according to the sequence of the set O, select the first element in the set LM, the element represents the wafer, and remove it from the set LM; S23c, holding and waiting, judging whether relevant conditions are met according to the wafer processing status, the wafer to be processed needs to meet the processable condition, and the completed wafer needs to meet the cooling condition. If they meet, they are transported to the subsequent corresponding equipment. For the wafer to be processed, it needs to be transported to the designated processing station for processing by the robot according to the set M, otherwise it continues to hold and wait until the corresponding conditions are met and then it is transported to the subsequent corresponding equipment; S24. First-come, first-served rule: The first-come-first-served rule is to select the wafer that first enters the robot blocking list for priority handling according to the time sequence of the wafers to be transported entering the blocking list; the specific steps are as follows: S24a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set LM to be transported in chronological order; S24b, if Then select the first element in the set, which represents the wafer, and remove it from the set of wafers to be moved LM; S24c, holding and waiting, judging whether relevant conditions are met according to the wafer type, the wafer to be processed needs to meet the processable condition, and the completed wafer needs to meet the cooling condition. If they meet, they are transported to the subsequent corresponding equipment, otherwise they continue to hold and wait until the corresponding conditions are met and then they are transported to the subsequent corresponding equipment; S25. Random handling rules: The random handling rule means randomly selecting a wafer to be handled from the robot blocking list for priority handling; the specific steps are as follows: S25a, initializing the atmospheric / vacuum robot wafer set LM to be transported, and adding all wafers waiting to be transported in the atmospheric / vacuum robot blocking list to the wafer set to be transported in order of number, where the element numbers are represented by 1, 2, ...n; S25b, if Then randomly select the wafer to be transported: w = Rand ({1, 2, ... n}), the random method Rand is used to randomly select an element in a specified set; select and grab wafer w, and remove wafer w from the set of wafers to be transported LM; Waiting for the part to be held. Judging whether the relevant conditions are met based on the wafer processing status. The wafer to be processed must meet the processable conditions, and the completed wafer must meet the cooling conditions. If they meet the conditions, they will be moved to the subsequent corresponding equipment. Otherwise, they will continue to wait for the part to be held until the corresponding conditions are met and then they will be moved to the subsequent corresponding equipment.

Images