High-throughput periodic scheduling methods

By constructing a process flow automata model and a discrete event system monitoring and control simulation platform, the utilization of equipment and processing sequence were optimized, solving the high-throughput production scheduling problem under limited hardware resources. This maximized equipment utilization and shortened the production cycle, ensuring a highly efficient and smooth production process.

CN120996440BActive Publication Date: 2026-11-17SICHUAN HIGH ENTROPY TECHNOLOGY CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202511085829.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-08-04
Publication Date
2026-11-17
Estimated Expiration
2045-08-04

AI Technical Summary

Technical Problem

In industrial production processes such as semiconductor manufacturing, biomedical research and development, and lithium battery manufacturing, how can we achieve efficient high-throughput production scheduling with limited hardware execution equipment resources to ensure efficient, smooth, and dynamically balanced production processes?

Method used

A high-throughput-oriented cycle scheduling method is adopted. By constructing a process flow automaton model, candidate scheduling paths that meet process requirements and are not deadlocked are generated. An intelligent scheduling system is used to optimize equipment utilization and processing sequence. Combined with a discrete event system supervision and control simulation platform, the time interval is dynamically adjusted to achieve the shortest production cycle.

Benefits of technology

Maximize equipment utilization, shorten production cycles, ensure smooth transitions between different process stages, improve production efficiency and resource utilization, and achieve efficient production scheduling.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120996440B_ABST
    Figure CN120996440B_ABST
Patent Text Reader

Abstract

This invention discloses a high-throughput cycle scheduling method, belonging to the field of industrial production cycle scheduling technology. The invention includes: inputting production data, including: work area division information, equipment information, and process flow information; constructing a process flow automaton model for each process flow based on a discrete event system supervisory control simulation platform, and generating candidate scheduling paths that meet process requirements and are not deadlocked, using work areas, process steps, and equipment as scheduling path nodes; selecting the candidate scheduling path with the shortest total processing time for each production cycle as the optimal scheduling path; and then distributing scheduling information to the corresponding work equipment within the work area based on the optimal scheduling path, executing the processing operations of the processing objects based on the received scheduling information. This invention can maximize equipment utilization, shorten the production cycle, and ensure smooth connection between each process stage; it maximizes resource utilization while ensuring production efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of industrial production cycle scheduling technology, and in particular to a cycle scheduling method for high throughput. Background Technology

[0002] Industrial production cycle scheduling refers to the process of inspecting, evaluating, and adjusting the entire production system according to a certain time cycle. This includes multiple aspects such as the allocation of production tasks, the determination of production sequence, the allocation of production resources, and the monitoring of production progress. For industries such as semiconductor manufacturing, biopharmaceutical research and development, and lithium battery manufacturing, the production processes are complex, with close and intricate collaboration between various links, high production continuity, and rapid changes in conditions. If a failure occurs in one link, or if a planned measure is not implemented on schedule, it can have a ripple effect, affecting the stable operation of the entire production system. Therefore, strengthening production scheduling is crucial. Ultimately, the problem of production cycle scheduling is a dynamic equilibrium problem. It requires finding a stable rhythm between complexity, uncertainty, and dynamism. This is not only a technical issue but also a systemic one.

[0003] High-throughput (HT) refers to the ability to process or analyze large numbers of samples, data, or tasks per unit of time, typically achieved through significant efficiency improvements via automation, parallelization, or intelligent technologies. In fields such as semiconductors, materials science, and biomedicine, the core objectives of high-throughput methods are to accelerate research and development, optimize production, and reduce costs.

[0004] How to achieve high-quality, high-throughput production scheduling in industrial production processes such as semiconductor manufacturing, biomedical research and development, and lithium battery manufacturing is an urgent problem to be solved, so as to achieve efficient production cycle scheduling of process processing under the condition of limited hardware execution equipment resources. Summary of the Invention

[0005] This invention provides a high-throughput cycle scheduling method, which aims to achieve high-throughput cycle scheduling with the shortest total processing cycle scheduling time under the condition of limited hardware execution device resources, so as to ensure the efficiency and smoothness of industrial production processes.

[0006] The technical solution adopted in this invention is as follows:

[0007] A high-throughput periodic scheduling method includes the following steps:

[0008] Input production data, including: work area division information for process processing, equipment information, and process flow information;

[0009] The equipment information includes processing equipment, auxiliary equipment, and loading devices for loading the processing objects (workpieces);

[0010] The process flow information includes: process flow information of the processed object and process flow information of the loading device based on the description of the work area and process steps; wherein, the process flow information includes: the working equipment, the number of working equipment, the logical relationship between the working equipment and the process steps, and the duration of each action in each process step in each work area.

[0011] Based on the discrete event system supervisory control simulation platform, a process flow automaton model is constructed for each process flow. The work area, process step and equipment are used as scheduling path nodes to generate candidate scheduling paths that meet the process requirements and are not deadlocked. The candidate scheduling path with the shortest total processing time for each production cycle is selected as the optimal scheduling path. The process automaton model represents each process step according to a unified naming rule to represent the name (which can be described as an event name), action, work equipment, location (i.e., the work area to which the process flow belongs) and state of each process step.

[0012] Based on the optimal scheduling path, the corresponding scheduling information is distributed to each working device in the work area, so as to execute the processing operation of the processing object based on the received scheduling information.

[0013] In this invention, the scheduling information can be set as: the current operation information of the working equipment (including acquired production data), and the preceding and following nodes on the optimal scheduling path. This invention's method is applicable to intelligent scheduling systems including a host computer and robotic arms (i.e., auxiliary equipment), wherein the host computer generates the optimal scheduling path for the processing object and sends the optimal scheduling path to the corresponding robotic arm, and distributes its relevant scheduling information to each working equipment on the optimal scheduling path; the robotic arm moves the loading device or the processing object based on the optimal scheduling path, and each working equipment executes the optimal scheduling path based on the received scheduling information.

[0014] The high-throughput cycle scheduling method proposed in this invention determines the process parameters (such as processing time and other key process parameters) of the workpiece based on its technological requirements before production scheduling. These parameters are then sent to a host computer, which can also be set as the main control unit of an intelligent scheduling system. Upon receiving the process parameter information, the host computer uses the intelligent scheduling algorithm proposed in this invention to comprehensively consider factors such as equipment performance, processing time, and area capacity. It then selects a high-throughput scheduling path from numerous possible candidate paths: the candidate scheduling path with the shortest total processing time for each production cycle. This high-throughput scheduling path aims to maximize equipment utilization, shorten the production cycle, and ensure smooth transitions between different process stages. In other words, the host computer can adjust the transport sequence and frequency of loading devices in each work area in real time according to production needs to ensure efficient and smooth production processes. After generating the high-throughput scheduling path, the main control unit accurately calls each workpiece based on the corresponding scheduling path information, enabling each machine to execute tasks according to a predetermined sequence and actions, thereby realizing the execution of the entire scheduling task.

[0015] Furthermore, constructing the process flow automata model specifically includes:

[0016] For each process step in the process flow, generate an automated processing machine, and divide the generated automated processing machines into several groups based on the number of available processing equipment; perform a synchronous product operation on each group to obtain the process step model for each group.

[0017] Generate a first loading machine entering the work area, a second loading machine leaving the current work area and entering the next work area, and a third loading machine leaving the work area for the loading device; perform synchronous product on the loading machines involved in the current process flow to form a loading entry and exit model of the process flow;

[0018] Construct auxiliary equipment operation constraint automata and processing equipment access constraint automata, and generate process flow automaton model based on the loading and unloading model of the current process flow, the process step model involved in the current process flow, and the synchronous product of constraint automata.

[0019] Synchronization product is an operation that combines multiple system models into a single joint model. It represents their behavior when executed synchronously on a specified event (such as a process step) and independently on other events. This is one of the fundamental operations in discrete-event system modeling and control. In short: if two systems are modeled for a common event, they must execute that event simultaneously; if only one system can execute an event, then that system executes it, while the other system remains unchanged; ultimately, a new model is generated describing all possible behaviors of the multiple systems running together. Synchronization product is commonly used for: combining multiple subsystems into a complete system; synthesizing systems with control specifications; and checking whether system behavior satisfies user-specified constraints (such as deadlock, reachability, etc.).

[0020] Furthermore, the generation of several scheduling paths also includes:

[0021] Define the time interval between the start of processing of the first processing object in the adjacent layer of the loading device as T0;

[0022] Within the adjustable range of the set time interval T0, the time interval T0 is iteratively and elastically adjusted to obtain the shortest time interval T0 under the condition that the duration of each action in the process flow is satisfied and there is at least one candidate scheduling path without deadlock.

[0023] Based on the shortest time interval T0, the optimal scheduling path is selected from the candidate scheduling paths based on the shortest total processing time.

[0024] Furthermore, when executing the process steps in the designated work area, the same layer of the loading device is divided into multiple parts, and each part is executed synchronously by multiple working devices.

[0025] Furthermore, when performing the process steps in a designated work area, the start interval between adjacent processing objects on the same work equipment is defined as T1, and the start interval between adjacent processing objects between different work equipment is defined as T2.

[0026] Within the boundary range of T1 and T2, the time intervals T1 and T2 are flexibly adjusted to obtain the shortest time intervals T1 and T2 under the condition of at least one candidate scheduling path that meets the process requirements and is not deadlocked.

[0027] Based on the shortest time intervals T1 and T2, the optimal scheduling path is selected from the candidate scheduling paths based on the shortest total processing time.

[0028] Furthermore, when executing the process steps in a designated work area, two processing devices can be configured for the process steps in the current designated work area. The processing objects on the same layer of the loading device are divided into two parts according to odd and even numbers. The time interval between the start of processing of two adjacent processing objects on the same layer is defined as T1, and the time interval between the start of the two processing devices is defined as T2. The time intervals T1 and T2 are flexibly adjusted to obtain the shortest time intervals T1 and T2 under the condition of at least one candidate scheduling path that satisfies the process parameters and does not deadlock. Based on the shortest time intervals T1 and T2, the optimal scheduling path is selected from the candidate scheduling paths based on the shortest total processing time.

[0029] The scheduling strategies used to generate candidate scheduling paths include at least one of the following:

[0030] Single-threaded scheduling strategy based on for loop:

[0031] Set up inner and outer for loops, where the outer for loop is used to serially traverse the different values ​​of the time interval T1 starting from the lower limit of the boundary range, and the inner for loop is used to serially traverse the different values ​​of the time interval T2 starting from the lower limit of the boundary range.

[0032] Each iteration of the outer for loop must wait for the inner for loop to complete before continuing to iterate over the next value of T1.

[0033] For each iteration of the inner for loop, the next value of T2 can only be iterated after the supervisory controller has verified it. The supervisory control check is used to determine whether the current time intervals T1 and T2 can generate a candidate scheduling path that meets the process requirements and is not deadlocked.

[0034] The first pair of time intervals (T1, T2) verified by the supervisory controller is found and used as the shortest time intervals T1 and T2. The optimal scheduling path is then obtained based on the candidate scheduling paths corresponding to the shortest T1 and T2.

[0035] Multi-threaded parallel scheduling strategy:

[0036] Based on the boundary range of T1 and T2 and the step size of value traversal, generate all possible value combinations (T1, T2) between intervals T1 and T2;

[0037] For all value combinations (T1,T2), candidate scheduling paths that satisfy the duration of each action in the process flow and do not deadlock are generated, i.e., candidate scheduling paths that meet the process requirements and do not deadlock are generated.

[0038] The optimal scheduling path is obtained by selecting the candidate scheduling path corresponding to the smallest (T1,T2).

[0039] In the scheduling method of this invention, a two-layer cyclic scheduling can be implemented based on set time intervals T1 and T2. The cyclic scheduling with respect to time interval T1 is considered the outer loop, while the cyclic scheduling with respect to time interval T2 is considered the inner loop. The outer loop is primarily responsible for handling the processing sequence constraints of different processing objects on the same processing equipment and ensuring that deadlocks do not occur during resource allocation. Given multiple processing equipment and evenly distributed tasks, the outer loop focuses on how to rationally arrange the start times of different processing objects on the same processing equipment to maximize overall production efficiency. Specifically, after a processing object on the current loading device completes its processing task, it is necessary to determine when the next processing object should begin scheduling. To achieve this goal, the outer loop dynamically adjusts the time interval T1. During each iteration, the system determines whether a valid and feasible scheduling path has been successfully found. By continuously adjusting T1 and performing feasibility checks, the system continuously tries and optimizes the scheduling scheme under limited resource conditions, aiming to find the shortest cycle T1 that can achieve the production target. The shortest cycle time T1 not only signifies increased production efficiency but also maximizes resource utilization, ensuring each processing device performs at its full potential and minimizing idle time. The inner loop focuses on resolving the sequential constraints of different processing devices while ensuring no deadlock occurs during operation across multiple devices. Following the principle of even task allocation, processing objects on each device are rationally distributed according to the sequence of processing steps. The time interval T2 is defined here as the cycle time from the start of operation on one processing device to processing on another. The inner loop elastically adjusts T2 to seek the optimal scheduling path while ensuring smooth workflow across different processing devices and preventing deadlocks. By continuously optimizing T2, the system aims to find the most efficient and shortest scheduling path while minimizing both T1 and T2. This path fully utilizes the processing capacity of each device, reducing waiting time between processing objects and thus improving the overall efficiency of the production process. In summary, the two-layer loop scheduling method of this invention employs a two-layer loop architecture. The outer loop represents the upper and lower bounds of the scheduling time, while the inner loop solves for the path. If the inner loop fails to find a valid scheduling path that meets preset conditions during the current solution process, the outer loop will automatically trigger the next iteration. Through this iterative mechanism, scheduling parameters are continuously adjusted and optimized, and feasible scheduling schemes are continuously explored until the optimal solution that meets production needs and achieves efficient scheduling is found. Under the condition of multiple processing equipment and evenly distributed tasks, the time intervals T1 and T2 are continuously optimized to maximize production efficiency and optimize resource utilization, ultimately finding an efficient and short-time scheduling path.

[0040] It should be noted that the lower limit of the inner and outer circulation is not necessarily 0, and can be determined by specific process parameters.

[0041] Furthermore, the process flow information also includes the step priority of each process step. When selecting the optimal scheduling path, candidate scheduling paths that do not meet the step priority are cut off.

[0042] Furthermore, it also includes path optimization of the generated candidate scheduling paths. When adjacent path nodes in the scheduling path are loading device moving steps and auxiliary equipment operation steps (such as robotic arm grasping workpieces), the priority of the loading device moving steps is set higher than the priority of the auxiliary equipment operation steps.

[0043] Furthermore, the work area includes a preparation area, a transfer area, a core processing area, a secondary processing area, an analysis area, and a final processing area;

[0044] The preparation area is used to transfer the workpiece to the loading device;

[0045] The transfer area is used to allocate the loading device for loading the processed objects to the corresponding work area for processing, according to the cycle of the processed objects;

[0046] The core processing area is used to perform designated core process processing on the objects being processed on the loading devices located in this work area;

[0047] The secondary processing zone is used for secondary processing of the workpieces on the loading devices located in this work area, and all workpieces on the loading devices located in the secondary processing zone have completed the core processing.

[0048] The analysis area is used to analyze and inspect the processing quality of the workpieces on the loading device located in this work area, and all workpieces on the loading device located in the analysis area have completed the next process.

[0049] The analysis area is used to analyze and inspect the workpieces on the loading device. After the workpieces on the loading device have completed processing in the core processing area and secondary processing area, they are moved to the analysis area. Using appropriate testing equipment and analysis methods, the processing quality of the workpieces is comprehensively inspected. Through the inspection in the analysis area, problems existing in the processing process can be identified in a timely manner, providing a basis for process optimization and quality improvement, and ensuring that the quality of the workpieces meets the requirements.

[0050] The end area serves as a storage area, used to store loading devices for processed objects that have passed quality inspection.

[0051] Furthermore, the operational scenarios corresponding to the generated data include, but are not limited to: semiconductor manufacturing, biomedical research and development, and lithium battery manufacturing.

[0052] The technical solution provided by this invention brings at least the following beneficial effects:

[0053] When processing materials for a target object using the optimal scheduling path generated by this invention, equipment utilization can be maximized, production cycle shortened, and smooth connection between each process stage can be ensured; while ensuring production efficiency, resource utilization is maximized. Attached Figure Description

[0054] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0055] Figure 1 This is a schematic diagram illustrating the processing steps of the high-throughput periodic scheduling method provided in an embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram of the processing of perovskite semiconductor material in an embodiment of the present invention;

[0057] Figure 3 This is a schematic diagram of a pharmaceutical material processing procedure in an embodiment of the present invention;

[0058] Figure 4 This is a schematic diagram of the state of the automatic spin coater model in an embodiment of the present invention;

[0059] Figure 5 This is a schematic diagram of the state of the automated pipetting machine model in an embodiment of the present invention;

[0060] Figure 6 This is a schematic diagram of the state of the TIP taking / taking model in an embodiment of the present invention;

[0061] Figure 7 This is a schematic diagram of the state of the annealing automata model in an embodiment of the present invention;

[0062] Figure 8 This is a schematic diagram of the state of the carrier model for the area to be coated in this embodiment of the invention;

[0063] Figure 9 This is a schematic diagram of the state of the carrier model of the area to be coated in an embodiment of the present invention;

[0064] Figure 10 This is a schematic diagram of the state of the photographing model in an embodiment of the present invention. Detailed Implementation

[0065] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions of the embodiments of the present invention will be described in detail and completely below with reference to the accompanying drawings. Obviously, the embodiments described with reference to the accompanying drawings are exemplary and intended to explain the present invention, but should not be construed as limiting the present invention.

[0066] In one embodiment, the high-throughput cycle scheduling method provided by this invention is applicable to intelligent scheduling systems including a main control unit and a robotic arm, and can be used in operational scenarios such as semiconductor fabrication, biomedical research and development, and lithium battery fabrication. The main control unit is the host computer of the intelligent scheduling system, used to generate the optimal scheduling path for controlling the robotic arm. The robotic arm moves the loading device (such as a carrier) or the processing object (such as a glass slide) based on the optimal scheduling path. Figure 1 As shown, the method for generating the optimal scheduling path includes the following steps:

[0067] Step S1: Input production data, including: work area division information, equipment information, and process flow information; wherein, the equipment information includes processing equipment, auxiliary equipment, and loading devices for loading the processed objects; the process flow information includes: process flow information of the processed objects and the loading device, describing each process step based on the work area and process flow; and the process flow information includes: the working equipment, the number of working equipment, the logical relationship between the working equipment and the process steps, and the duration of each action in each process step in each work area;

[0068] Step S2: Based on the Discrete Event System (DIS) Supervisory Control (SDC) simulation platform, a process flow automaton model is constructed for each process flow. Using the work area, process step, and equipment as scheduling path nodes, candidate scheduling paths that meet process requirements and avoid deadlocks are generated based on the process flow. The candidate scheduling path with the shortest total processing time for each production cycle is selected as the optimal scheduling path. The process flow automaton model represents each process step of each process flow according to a unified naming convention, representing the event name, action, working equipment, location, and state of the corresponding process step (i.e., event).

[0069] Step S3: Distribute the corresponding scheduling information to each work device in the work area based on the optimal scheduling path, so as to execute the processing operation of the processing object based on the received scheduling information.

[0070] In one embodiment, the processing area can be subdivided into a preparation area, a transfer area, a core processing area, a secondary processing area, an analysis area, and a final processing area; each area is specifically as follows:

[0071] (1) The preparation area is used to pre-process the objects to be processed, so as to make full preparations for subsequent process operations. Before the scheduling begins, the objects to be processed are loaded onto the carrier, completing the loading step. By performing pre-processing in the preparation area, it can be ensured that the carrier has the correct initial state when it enters the subsequent process area, which facilitates the orderly development of subsequent process operations and improves the overall processing efficiency.

[0072] (2) The transfer area implements a refined zoning management strategy for processing objects with different cycle periods. Based on the cycle period of the processing objects, they are assigned to corresponding sub-areas. Through reasonable scheduling and management of the carriers in each sub-area, the core processes are ensured to proceed in an orderly manner according to the predetermined sequence and rhythm. This refined zoning management method can effectively avoid mutual interference between carriers with different cycle periods, improving the execution efficiency and stability of the core processes.

[0073] (3) The core processing area is the main area where core processing techniques are implemented on the processed objects. Within this area, the processed objects on the carrier are precisely processed according to predetermined core process parameters and operating procedures to achieve the key performance requirements of the processed objects. The core processing area is the core link of the entire carrier processing system, and the quality and efficiency of its process execution directly affect the final processing result.

[0074] (4) The secondary processing area is used to perform secondary processing on the object to be processed. After the core process is completed, the object to be processed is transferred to the secondary processing area for further processing according to the requirements of the secondary process. The secondary process can improve the performance of the object to be processed or meet specific additional requirements.

[0075] (5) The analysis area is used for analyzing and inspecting the processed carrier. After the carrier completes processing in the core processing area and secondary processing area, it is moved to the analysis area, where appropriate testing equipment and analysis methods are used to comprehensively inspect the processing quality of the processed object. Through the inspection in the analysis area, problems existing in the processing can be identified in a timely manner, providing a basis for process optimization and quality improvement, and ensuring that the quality of the processed object meets the requirements.

[0076] (6) The end area serves as a storage area for racks containing processed objects that have passed quality inspection. After the quality of the processed objects on the racks is confirmed to be qualified in the analysis area, the racks are moved to the end area for storage. The end area enables centralized management of qualified racks, facilitating subsequent outbound, transportation, or other operations.

[0077] In one embodiment, when performing a process step in a designated work area (such as a core processing area), the processing objects on the same layer of the carrier are divided into two parts according to odd and even numbers, and the process step is executed synchronously on each part by two core processing equipment. Of course, embodiments of the present invention also support the synchronous execution of multiple core processing equipment. In this case, the processing objects on the same layer of the carrier need to be divided into several parts, and each part is configured with one processing equipment to execute the corresponding process step.

[0078] Under normal circumstances, when scheduling is carried out on a two-level rack cycle (level=2), each designated work area requires the use of Zone 1 and Zone 2, i.e., two corresponding processing machines to perform the corresponding processing operations. When scheduling is carried out on a four-level rack cycle (level=4), the transfer area requires the use of up to four zones to improve scheduling efficiency.

[0079] In one embodiment, the scheduling process for the aforementioned work area includes:

[0080] 1) Before production scheduling begins, key process parameters such as the core process processing time and secondary process processing time are determined based on the process requirements of the objects to be processed. These process parameter information is then accurately sent to the main control unit. Upon receiving the process parameter information, the main control unit uses the optimal scheduling path generation method (i.e., intelligent scheduling algorithm) of this embodiment of the invention to comprehensively consider factors such as equipment performance, processing time, and area capacity, selecting a high-throughput scheduling path from among many possible paths. This high-throughput scheduling path aims to maximize equipment utilization, shorten the production cycle, and ensure smooth connection between each process stage. Subsequently, based on the planned scheduling path information, the main control unit precisely calls upon processing machines such as robotic arms, enabling each machine to execute tasks according to a predetermined sequence and actions, thereby completing the entire scheduling task.

[0081] 2) In the preparation area, based on the production plan and equipment capacity, a pre-set number of racks with a predetermined number of layers are prepared to ensure that the quantity and condition of the racks meet subsequent production needs. Through the precise operation of a robotic arm, the racks are moved from the preparation area to the transfer area. In the transfer area, the racks are in a standby state, awaiting entry into the core process processing stage. This step ensures the orderly flow and preparation of the processed objects before entering the core process, laying the foundation for the efficient execution of subsequent processes.

[0082] 3) Using a robotic arm, the objects to be processed on the racks in the transfer area are removed one by one and moved to the core processing area. To improve the processing efficiency and equipment utilization rate in the core processing area, the objects on the racks are numbered. According to the numbering rules, odd-numbered objects are moved to core processing area one, and even-numbered objects are moved to core processing area two. This zoned processing method can make full use of the equipment resources in the core processing area, avoid equipment idleness and conflicts, and improve the processing efficiency and quality of core processes.

[0083] 4) After the core processing of an object is completed in the core processing area, a robotic arm moves it from the core processing area to the secondary processing area for secondary processing. After completing the transport of the current object, the robotic arm can immediately return to handle the next object without waiting for the secondary processing to finish. This parallel processing method effectively shortens the waiting time between different process stages and improves overall production efficiency.

[0084] 5) After processing all objects on a layer of the rack, the rack is moved to the analysis area by a robotic arm. In the analysis area, the processed objects on the rack are analyzed using detection equipment and analytical methods. After the analysis, if the processed object is of acceptable quality, the rack is moved to the end area for storage; if quality problems exist, they are handled according to the preset exception handling procedure. The scheduling paths for processed objects differ depending on the processing cycle. The use of different areas also varies.

[0085] 6) After processing all objects within a processing cycle (i.e., processing the preset number of objects), a new carrier is moved from the preparation area to the transfer area to begin scheduling for the next cycle. This periodic, cyclical processing method ensures continuous and stable operation of the production process.

[0086] In one embodiment, the implementation steps of the method for generating the optimal scheduling path provided by the present invention include:

[0087] (1) Model building (i.e., scheduling path generation model) to generate scheduling paths that meet process parameters and do not deadlock;

[0088] (2) Controller solution, that is, the optimal scheduling time interval is determined based on the controller (such as the controller solution method based on the discrete event system supervision control theory (abbreviated as supcon) to synthesize the supervisory controller). That is, under the premise of ensuring smooth resource allocation and avoiding deadlock, the scheduling time interval of each node is flexibly adjusted to iteratively solve the corresponding optimal (shortest) time interval.

[0089] (3) For the scheduling path obtained by the controller, the scheduling path with the shortest total processing time for each production cycle is taken as the optimal scheduling path.

[0090] In one embodiment, when considering the priority of different process steps, based on the agreed-upon process step priority division (ensuring that the events most needed to occur immediately occur promptly according to the priority of different process steps, thereby minimizing the total processing time and maximizing efficiency), the scheduling paths obtained by the controller are prioritized to obtain a single path. That is, when selecting the optimal scheduling path, scheduling paths that do not meet the event priority are pruned from the scheduling paths obtained by the controller.

[0091] In one embodiment, the scheduling strategy for generating scheduling paths that satisfy process parameters and do not deadlock includes at least one of the following:

[0092] (1) Scheduling strategy 1: Single-threaded scheduling based on for loop

[0093] In the single-threaded scheduling strategy based on a for loop provided in this embodiment of the invention, the characteristics of the for loop are utilized to achieve sequential execution and precise control of tasks under a single thread. The specific implementation process is as follows:

[0094] Define a for loop structure in the program. This for loop executes scheduled tasks, such as a robotic arm picking up and placing a carrier, in a predetermined order by the same thread. The for loop has a clear start condition, termination condition, and step size setting. When the program executes, it first starts from the start condition of the for loop, which defines the initial value of the loop variable (index variable).

[0095] The program then increments (or decrements) the index variable (such as dwell time) according to the set step size. The size of the step size determines the magnitude of the change in the index variable in each loop iteration. By setting the step size appropriately, the execution speed and range of the loop can be flexibly controlled.

[0096] In each iteration of the for loop, the program executes the instructions within the loop body once. These instructions can be various operations, such as device control, and are executed sequentially according to a predetermined logical order.

[0097] The program will continue to iterate in a loop until the loop termination condition is met. The loop termination condition is usually related to the value of the index variable. The for loop ends when the value of the index variable reaches or exceeds (or falls below) the threshold set by the termination condition, or when a scheduling path is found.

[0098] (2) Scheduling strategy 2: Multi-threaded scheduling based on for loop

[0099] In the multi-threaded scheduling strategy based on for loop provided in the embodiments of the present invention, by reasonably modifying the for loop, multi-threaded parallel execution can be achieved, and task division, thread scheduling and synchronization mechanisms can be effectively controlled to ensure efficient and stable operation in a multi-threaded environment.

[0100] Specifically, the tasks covered by the for loop can be finely broken down, allowing each thread to independently handle a portion of the tasks within the for loop, or a dedicated thread can be assigned to handle each iteration of the for loop. This modification fully leverages the advantages of multi-threaded parallel computing, aiming to significantly improve task processing efficiency. The processing logic for each iteration of the for loop is encapsulated as an independent task and submitted to a thread pool for unified management.

[0101] For example, in optimizing the cycle period parameters of an intelligent scheduling system, a parallel computing strategy can be employed to find the cycle period value that maximizes scheduling efficiency. Specifically, the scheduling path calculation process is initiated simultaneously for multiple different preset cycle period values. After completing the scheduling path calculation for each cycle number, the optimal scheduling path is selected based on the path results, and the cycle period value corresponding to this optimal path is determined as the cycle period parameter used in practical applications.

[0102] A thread pool acts as a highly efficient resource allocation center, intelligently distributing threads to execute tasks based on the current system load and thread idle status. This approach not only avoids the significant system overhead of frequently creating and destroying threads, but also allows for precise control of system concurrency through proper configuration of thread pool parameters (such as the number of threads and task queue capacity), thereby optimizing overall performance.

[0103] In one embodiment, the high-throughput-oriented periodic scheduling method proposed in this invention can be applied to the fabrication of perovskite semiconductor materials. Perovskite semiconductor materials have become an important research direction in recent years due to their excellent photoelectric properties, low-cost fabrication processes, and broad application prospects in solar cells and photodetectors. In their fabrication process, glass slides, as substrates, undertake crucial physical and chemical processing steps.

[0104] For perovskite semiconductor materials, such as Figure 2 As shown, the division of the work area specifically includes: preparation area, transfer area, core processing area (spin coating operation area), secondary processing area (flash evaporation operation area, annealing operation area), analysis area and end area. In specific operations, one can directly jump from the transfer area to the analysis area or directly from the core processing area to the analysis area. Each work area also includes multiple sub-areas to achieve synchronous processing of processing objects (wave plates) on different layers of the carrier. Figure 2In this context, n, m, p, q, r, and s represent the number of sub-regions for each work area.

[0105] For example, the entire work area for perovskite semiconductor materials is divided into: preparation area 1, preparation area 2, preparation area 3, preparation area 4, coating area 1, coating area 2, coating area 3, coating area 4, spin coating area 1, spin coating area 2, flash evaporation area 1, flash evaporation area 2, annealing area 1, annealing area 2, annealing area 3, annealing area 4, photography area, and finishing area. The preparation area involves pre-preparing the pre-set carrier layers and using a robotic arm to move the carrier to the coating area, awaiting the spin coating process. When the spin coating time is up, the robotic arm moves the glass slide from the carrier in the coating area. Move the slides to the spin coating area and number them on the carrier. Odd-numbered slides move to spin coating area 1, and even-numbered slides move to spin coating area 2. Slides in the spin coating area require air blowing, liquid removal, and liquid spraying operations. Air blowing events are differentiated based on the spin coating area: air blowing on slides in spin coating area 1 is designated as Air Blow 1, and air blowing on slides in spin coating area 2 is designated as Air Blow 2, representing different operations. Slides after spin coating are sequentially moved to the flash evaporation area and annealing area via a robotic arm for further processing; both can be skipped. The processing sequence is spin coating, flash evaporation, and annealing. If there is no flash evaporation but annealing, the slide moves to the annealing area for annealing, and the next slide can be processed while waiting for annealing. If there is flash evaporation but no annealing, the slide moves to the flash evaporation area for flash evaporation. If there is neither flash evaporation nor annealing, the slide moves directly to the carrier. After processing one layer of slides, the slide can be moved to the photography area for taking pictures. After taking pictures, it can be moved to the end area. After processing several slides in a cycle, the slide can be moved from the preparation area to the coating area for the next cycle.

[0106] Furthermore, the scheduling paths for processing slides differ depending on the cycle. The use of different regions also varies. When processing 16 slides at once, only a maximum of four areas need to be used: Coating Area 1, Coating Area 2, Annealing Area 1, and Annealing Area 2. When processing 32 or more slides at once, a maximum of four areas need to be used: Coating Area 1, Coating Area 2, Coating Area 3, Coating Area 4, Annealing Area 1, Annealing Area 2, Annealing Area 3, and Annealing Area 4.

[0107] In one embodiment, the high-throughput-oriented periodic scheduling method proposed in this invention can be applied to materials processing in the pharmaceutical field, such as... Figure 3As shown, its operating area can be divided into: Central Storage, Hotel, Dispenser, Shaker, Incubator, and Reader; for example, for enzyme-linked immunosorbent assay (ELISA) experiments, the microplate is first stored in the Central Storage, which is the waiting area for the microplate before entering the system. Storage corresponds to the preparation area. Microplates are retrieved from the central storage area and sent to the Hotel (transfer area), which temporarily stores them and prepares them for subsequent processing steps. The microplates are then sent to the Dispenser, where the required liquid reagents are precisely dispensed into each well for pipetting. The dispensed microplates are then sent to the Shaker, which ensures thorough mixing of the liquid reagents for subsequent experimental steps, i.e., mixing. The mixed microplates are then placed in the Incubator, which provides suitable environmental conditions, such as temperature and time, to promote the reaction between the reagents and samples. This area corresponds to the core processing area. Finally, the incubated microplates are sent to the Reader for data analysis. The Reader measures the absorbance or other relevant parameters in each well to evaluate the experimental results. This area corresponds to the analysis area.

[0108] In one embodiment, the high-throughput cycle scheduling method proposed in this invention can be applied to lithium-ion battery manufacturing. Lithium-ion batteries, as a mainstream energy storage technology, involve multiple steps in their manufacturing process, including material preparation, electrode coating, assembly and packaging, electrolyte injection, formation, and capacity testing. Among these, the assembly stage is crucial for ensuring battery performance stability and safety, making its automation, precision, and consistency a key focus. In the field of lithium battery manufacturing, the overall process flow can be divided into an electrode preparation stage and a battery assembly stage, each specifically including:

[0109] (1) Electrode sheet preparation stage

[0110] This stage is a separate process. Electrode preparation is one of the core technologies of lithium battery technology, encompassing many key steps and technical aspects, including:

[0111] 1-1) Selection and Treatment of Active Materials: Based on battery performance requirements, carefully select suitable positive and negative electrode active materials. For example, the positive electrode may use lithium cobalt oxide, lithium manganese oxide, lithium iron phosphate, etc., while the negative electrode commonly uses graphite, silicon-based materials, etc. Refined treatment of the active materials, including particle size control and surface modification, is performed to optimize their electrochemical performance.

[0112] 1-2) Addition of conductive agents and binders: To improve the conductivity and structural stability of the electrode, appropriate amounts of conductive agents (such as carbon black, carbon nanotubes, etc.) and binders (such as polyvinylidene fluoride PVDF, etc.) need to be added. The proportions of each component and the mixing process should be optimized to ensure uniform dispersion of the electrode material, forming a good conductive network and mechanical structure.

[0113] 1-3) Slurry preparation: The active material, conductive agent, binder and solvent (such as N-methylpyrrolidone NMP) are mixed in a specific ratio and a uniform and stable electrode slurry is prepared by using advanced stirring equipment and process.

[0114] 1-4) Coating process: The prepared electrode slurry is uniformly coated onto the surface of the current collector (e.g., aluminum foil for the positive electrode and copper foil for the negative electrode). The coating thickness and uniformity directly affect the battery performance. The main focus is on the precision control of the coating equipment, the adjustment of the coating speed and tension, etc., to achieve high-precision and high-quality coating.

[0115] 1-5) Drying and Rolling: After coating, the electrode sheet needs to be dried to remove the solvent. This process requires precise control of the drying temperature, time, and atmosphere to avoid damage to the electrode material structure. After drying, the electrode sheet undergoes a rolling process to adjust its thickness and density, thereby increasing the electrode's volumetric energy density.

[0116] (2) Battery assembly stage

[0117] After the electrode sheet fabrication is completed, the battery assembly stage begins, which involves several key assembly steps:

[0118] Assembly preparation: Prepare the necessary components for assembly, including electrode shells, gaskets, positive electrode plates, separators, electrolytes, negative electrode plates, battery casings, etc.

[0119] Electrode housing placement: The electrode housing is placed at the assembly station as the basic structure for battery assembly. Its shape, size, and material must meet the requirements of battery assembly and use. Electrode housing design often focuses on its structural strength, sealing performance, and compatibility with subsequent assembly components.

[0120] Gasket Installation: Place a gasket in a suitable position inside the electrode housing. The gasket's function is to isolate the electrode plate from the electrode housing, prevent short circuits, and provide some cushioning and support. Key considerations include the gasket material selection (e.g., PTFE), dimensional accuracy, and the precision of the installation position.

[0121] Positive electrode placement: The prepared positive electrode is precisely placed in the designated position within the electrode shell, ensuring good contact between the positive electrode and the electrode shell and gasket, while avoiding interference with other components. The positive electrode placement process involves the accuracy of the placement equipment, the positioning method, and the stability control of the placement process.

[0122] Separator Placement: A separator is placed on the surface of the positive electrode. The separator's function is to isolate the positive and negative electrodes, prevent short circuits, and allow lithium ions to pass through. The separator's material (such as polyethylene PE, polypropylene PP, etc.), thickness, porosity, and other performance parameters have a significant impact on battery performance.

[0123] Electrolyte addition: Using precise addition equipment, an appropriate amount of electrolyte is added dropwise to the surface of the separator and the gaps between the electrode plates. The composition, concentration, amount added, and rate of addition of the electrolyte are crucial to battery performance and safety.

[0124] Negative electrode placement: Place the negative electrode on top of the separator, ensuring that the negative electrode is accurately aligned with the separator and the positive electrode to form a good electrode structure.

[0125] Battery casing encapsulation: The battery casing is placed over the assembled electrode assembly, and the battery casing is sealed through sealing processes (such as laser welding, glue sealing, etc.) to prevent electrolyte leakage and external environmental influences on the battery's internal structure.

[0126] Pressure treatment: The assembled battery is placed in a tablet press and a certain pressure is applied to make the internal components of the battery come into close contact, reduce contact resistance, and improve battery performance and stability.

[0127] In one embodiment, for perovskite semiconductor material processing, this embodiment of the invention provides the following two specific scheduling strategies for high-throughput periodic scheduling methods:

[0128] 1. A high-throughput periodic scheduling strategy for a dual-channel spin coater based on a discrete event system supervisory control theory model.

[0129] For the glass slides to be processed, this embodiment models different processing techniques as different automated machine models, including: pipetting, spin coating, annealing, flash evaporation, picking / applying TIP, entering the carrier for coating area, exiting the carrier for coating area, and taking pictures.

[0130] (1-2) Automata Model

[0131] Naming rules for robotic arm slide picking and placing events: For example, the event "Pick_Glass_n_From_Place_m" indicates that the robotic arm picks up the nth slide in region m of Place; the event "Put_Glass_n_Into_Place_m" indicates that the robotic arm places the nth slide in region m.

[0132] The naming convention for the "Move_Carrier_from_Place_to_Place" event is as follows: the first "Place" indicates the position of the carrier before it is moved, and the second "Place" indicates the position of the carrier after it is moved.

[0133] Naming rules for events that mark the end of a moving carrier: For example, “Move_Carrier_from_Place_to_Place_fn”, where “fn” means end, indicating that the event “Move_Carrier_from_Place_to_Place” has ended.

[0134] Naming rules for process steps (also known as process events): For example, the event "Coating_n" indicates that the nth spin coater performs the spin coating process.

[0135] Naming rules for the end of a process event: For example, "Coating_n_fn" indicates that the "Coating_n" event has ended.

[0136] Spin coater automatic machine model:

[0137] See Figure 4 The slides on the carrier are numbered with odd and even numbers. Odd-numbered slides are processed by spin coater 1, while even-numbered slides are processed by spin coater 2. Taking the Coat_1 model as an example, the execution time of this model includes the processing time of events such as picking up the odd-numbered slides from the carrier, pipetting (air blowing, suction, spraying), spin coating, and placing the odd-numbered slides onto the hot stage (or to the flash evaporation zone, or back onto the carrier). The event Pick_Glass_n_From_Carrier represents picking the nth slide (1≤n≤8) from the carrier, and the event Put_Glass_n_Into_Coater_m represents placing the picked-up nth slide into spin coater m (1≤m≤2). When the slide in spin coater 1 is subjected to air blowing, the event is defined as AirBlow_1; similarly, when the slide in spin coater 2 is subjected to air blowing, the event is defined as AirBlow_2. The event AirBlow_m_fn indicates the end of the air blowing process at spin coater m. Figure 4 The circles 1 to n represent the n states of the spin coater automaton model.

[0138] The time span between the start of the spin coater for the current odd-numbered slide and the start of the spin coater for the next odd-numbered slide is defined as the periodic time, T1.

[0139] Model of an automated pipetting machine:

[0140] See Figure 5This automaton model represents the execution of three processing steps in sequence: air blowing, liquid aspiration, and liquid spraying, when a pipetting event occurs (after the TIP event is completed). The liquid aspiration event is represented by Pip_a_Suck_in_Liquid_Zone_b_Place_c, where 'a' indicates which pipette tip aspirates liquid; 'b' indicates which pipetting zone is selected; 'c' indicates which well position within the pipetting zone; and 'Time_x_second' indicates a time interval of x seconds. Figure 5 The circles 1 to n in the diagram represent the n states of the pipetting automaton model.

[0141] Pick / Pick TIP Model

[0142] See Figure 6 The TIP retrieval event is executed once the Pick_Glass_From_Carrier event finishes. The number of TIPs retrieved at once by the pipette can be determined based on the TIP retrieval cycle and adjusted according to actual needs. The event Pips_d_to_e_Pick_f_ul_tips_g_to_h represents the coordinate range of the TIPs retrieved by the pipette; f indicates the type of TIP selected; and g_to_h represents the coordinate range of the TIP area. Figure 6 The circles 1 to n in the diagram represent the n states of the TIP model.

[0143] Annealing Automaton Model

[0144] See Figure 7 After the glass slide is spin-coated, an annealing process is performed. This automatic machine represents the annealing process flow. The event Put_Glass_n_Into_Heater indicates that the glass slide is picked up and placed on the heating stage. The event Heating_n indicates that the annealing process has started. The event Heating_n_fn indicates that the annealing process has ended. The event Pick_Heated_Glass_n indicates that the nth annealed glass slide is picked up and placed on the carrier. When the event Pick_Heated_Glass_n is completed, the annealing process is considered finished. Figure 7 The circles 1 to m in the diagram represent the n states of the annealing automaton model.

[0145] Model placed in the painting area

[0146] See Figure 8The event Move_Carrier_from_Prepare_Zone_Level_4_to_Waiting_Zone_1 indicates that the robotic arm has moved the fourth-level carrier from the preparation zone to the painting zone. The suffix "_fn" indicates that the event has been completed. Move_Carrier_from_Prepare_Zone_to_Waiting_Zone_fn indicates that the task of moving the carrier to the painting zone has been completed. Figure 8 The circles 1 to n+1 represent the n states of the carrier model in the area to be painted.

[0147] Model of the painting area

[0148] See Figure 9 Each carrier layer holds 8 slides. When the Put_Glass_8_back_to_Carrier event occurs, it means that all slides in one carrier layer have been processed and this carrier layer can be transported to the photography area. Move_Carrier_from_Waiting_Zone_1_to_Photo_Taking_Zone_fn indicates that the robot arm has completed the action of moving the carrier layer to be coated to the photography area. Figure 9 The circles 1 to n represent the n states of the carrier model in the area to be painted.

[0149] Photo model

[0150] The event Move_Carrier_from_Photo_Taking_Zone_to_Transit_Zone_fn indicates the completion of the robotic arm's action of moving the carrier from the area to be painted to the photo-taking area. The event Move_Carrier_from_Photo_Taking_Zone_to_Transit_Zone indicates the action of the robotic arm moving the carrier from the photo-taking area to the preparation area.

[0151] The event Move_Carrier_from_Photo_Taking_Zone_to_Transit_Zone_fn indicates the completion of the Move_Carrier_from_Photo_Taking_Zone_to_Transit_Zone event. The event Photo_Taking_1 indicates that a photographing operation is performed on the carrier, and Photo_Taking_1_fn indicates that the Photo_Taking_1 event has been completed, all the processing of the slides is finished, and the carrier can be moved to the preparation area. Figure 10 The numbers in the circles represent multiple states of the photographic model.

[0152] (1-3) Scheduling Ideas

[0153] For industrial scheduling with the same processing process and cycle, only high spin coating efficiency can lead to high overall efficiency. Therefore, the spin coater is the core processing machine, and in this embodiment, its "spin coating" process is the main control object, while the robotic arm, pipetting, and photography are constraints (due to limited resources).

[0154] The shortest total processing time is determined by minimizing the dwell time (excluding necessary time) after completing a process or operation at any node. This embodiment specifically employs a double-layer cyclic scheduling algorithm for this purpose.

[0155] Flexible model settings (taking a single-layer support structure that can support 8 glass slides as an example):

[0156] The outer loop addresses the constraint problem of the processing sequence of odd-numbered slides between carriers, specifically, when to schedule the next odd-numbered slide after the current odd-numbered slide has finished processing. T1 is defined as the cycle time from the start of processing the current odd-numbered slide to the start of processing the next odd-numbered slide. The outer loop dynamically scales and adjusts T1 elastically, ensuring no deadlock occurs during system resource allocation. Each iteration checks whether a valid and feasible scheduling path has been successfully found. The goal of this process is to continuously try and optimize under limited resource conditions to find the shortest cycle T1 that achieves the objective, thereby maximizing resource utilization while ensuring production efficiency.

[0157] The inner loop addresses the constraint problem of the processing order of adjacent odd and even glass slides. Under the rule that odd-numbered slides are processed by spin coater 1 and even-numbered slides are processed by spin coater 2, T2 is defined as the cycle time from the start of spin coating on an odd-numbered slide to the start of spin coating on an adjacent even-numbered slide. By gradually extending T2, the controller can solve for the scheduling path while ensuring that no deadlock occurs during the processing of odd and even slides.

[0158] Example: When T1 = 5s, perform a calculation of the inner and outer loops to determine whether the controller has successfully found a scheduling path. If the controller is empty, T1 is incremented by 1, and the above operation continues until the controller finds a valid scheduling path.

[0159] The goal is for the controller to find the most efficient and time-efficient scheduling path when T1 and T2 are minimized. T2 has no impact on efficiency, while the spring extension / retraction during T1 affects efficiency.

[0160] (1-4) Priority

[0161] Pruning the states of a given automaton based on an event priority mechanism to construct a closed-loop system that satisfies priority constraints is an important task aimed at optimizing system performance and behavior. The core objective of this process is to eliminate unnecessary state transitions triggered by low-priority events, thereby significantly improving the system's operational efficiency and response speed while ensuring the integrity of its functionality.

[0162] Specifically, automata, as mathematical models describing system states and their transitions, often face complex and varied event inputs in semiconductor material production. These events, such as a spin coater starting spin coating or a robotic arm picking up a glass slide, may have different levels of importance and urgency, i.e., priorities. Traditional automata processing mechanisms typically do not consider event priorities and process all events uniformly. This often leads to the system consuming excessive resources and time when processing low-priority events, thus affecting the response speed and processing capacity for high-priority events.

[0163] To address this issue, this embodiment introduces a pruning mechanism based on event priority. This mechanism first requires prioritizing each event in the automaton to determine which events are critical and require priority processing, and which are secondary. Then, based on this priority information, the automaton's state transition rules are adjusted and optimized, deleting or merging state transitions triggered by low-priority events that have a minor impact on the overall system behavior.

[0164] Priority definition rule: To maximize the efficiency of event execution while ensuring normal process operation.

[0165] In the highly precise and complex semiconductor manufacturing process, the prioritization of various events is meticulously planned. Among these, the highest priority event is the spin-coating completion event (coat_fn). Spin-coating plays a crucial role in the entire semiconductor manufacturing process, arguably the most important step. It directly affects the smooth execution of subsequent process steps. During spin-coating, a specific material needs to be uniformly coated onto the surface of the semiconductor substrate. This process demands extremely high standards for material uniformity, thickness control, and adhesion to the substrate. Only when the spin-coating event ends, ensuring the material is accurately adhered to the substrate, can subsequent processing steps have a solid foundation. Therefore, the completion of the spin-coating event is given the highest priority; its completion signifies that the entire process has entered a new critical stage.

[0166] In contrast, the execution priority of other process completion events is lower. These processes encompass a series of operations such as annealing, air blowing, liquid suction, and liquid spraying. Each process step has its specific goals and requirements; they are interconnected and influence each other, collectively constructing the functional structure of semiconductor products. Although their importance is not as great as spin coating, they are equally indispensable in the entire process flow. When these processes are completed sequentially, they need to be confirmed and processed according to a predetermined priority order to ensure the orderly progress of the entire process flow.

[0167] Next comes the robotic arm movement event. Robotic arms play a crucial role in slide handling and carrier movement. They need to move precisely between different processing devices to accurately place the slides in their respective processing positions. The efficiency and accuracy of the robotic arm's movement directly affect the efficiency and stability of the entire process flow. However, compared to spin coating events and other processing steps, robotic arm movement events have a relatively lower priority. This is because the movement of the robotic arm mainly plays an auxiliary and connecting role; its operation can be paralleled to other process steps to a certain extent or slightly delayed, as long as it does not affect the overall process progress and product quality.

[0168] Finally, the "Time_x_second" event has the lowest priority. The "Time_x_second" event can be understood as a marker of a time point, representing the passage of time. In semiconductor manufacturing processes, time is only counted as consumed when an actual process step begins execution. In other words, the "Time_x_second" event exists in relation to other process steps, hence its lowest priority. This priority setting ensures that the entire process flow is carried out in the most rational and efficient manner, guaranteeing that semiconductor materials can be manufactured in a high-quality, high-efficiency production environment.

[0169] In this way, this embodiment can construct a more streamlined and efficient closed-loop system. In this system, high-priority events can be responded to and processed in a timely manner, while low-priority events will not cause excessive interference or burden on the normal operation of the system. This not only improves the overall performance of the system, but also makes the system's behavior more predictable and controllable.

[0170] While meeting the requirements of the processing technology, the priorities of scheduling events are clearly defined by comprehensively considering factors such as the sequence of processing steps, resource utilization efficiency, and time constraints. With these priorities assigned, the scheduling path of the double-loop solution is optimized to find the most efficient path with the shortest total processing time.

[0171] In this embodiment, the code implementation process of the priority algorithm can be described as follows:

[0172]

[0173]

[0174] 2. High-throughput periodic scheduling strategy for a single-channel spin coater based on supervisory control theory

[0175] Considering the glass slide as the processing object, this embodiment models various processing flows as different automated machine models, specifically including: liquid transfer process, spin coating process, flash evaporation process, annealing process, pick-up / tip operation, movement of the carrier into the coating area, movement of the carrier away from the coating area, and photography process, totaling eight processing steps.

[0176] (2-1) Automata Model

[0177] Naming rules for robotic arm slide picking and placing events: For example, the event "Pick_Glass_n_From_Place" indicates that the robotic arm picks up the nth slide in the place area; the event "Put_Glass_n_Into_Place" indicates that the robotic arm places the nth slide in the place area.

[0178] The naming convention for the "Move_Carrier_from_Place_to_Place" event is as follows: the first "Place" indicates the position of the carrier before it is moved, and the second "Place" indicates the position of the carrier after it is moved.

[0179] Naming rules for events that mark the end of a moving carrier: For example, “Move_Carrier_from_Place_to_Place_fn”, where “fn” means end, indicating that the event “Move_Carrier_from_Place_to_Place” has ended.

[0180] Naming rules for process events: For example, the event "Coating_n" indicates that the nth spin coater performs a spin coating process.

[0181] Naming rules for the end of a process event: For example, "Coating_n_fn" indicates that the "Coating_n" event has ended.

[0182] Spin coating automatic machine model

[0183] The slides on the carrier are numbered, and the slides are processed using a spin coater. The execution time of the Coat model includes the processing time for events such as removing the slide from the carrier, pipetting (air blowing, suction, spraying), spin coating, and placing the slide onto the hot stage. The time span from the moment the spin coater is started for the currently numbered slide to the moment the spin coater is started for the next numbered slide is defined as the cycle time, T1.

[0184] Pipette Automaton Model

[0185] This automaton model represents the execution of three processing steps in sequence: air blowing, liquid aspiration, and liquid spraying, when a pipetting event occurs (after the TIP event is completed). When the slide in the spin coater is subjected to the air blowing process, the event is defined as AirBlow; the liquid spraying event is defined as Liquid_Spray; and the liquid aspiration event is represented by Pip_a_Suck_in_Liquid_Zone_b_Place_c. 'a' indicates which pipette tip aspirates liquid; 'b' indicates which pipetting zone is selected; and 'c' indicates which well position within the pipetting zone.

[0186] Pick / Pick TIP Model

[0187] The TIP retrieval event is executed once the Pick_Glass_from_Carrier event finishes. The number of TIPs retrieved at once by the pipette can be determined based on the TIP retrieval cycle and adjusted according to actual needs.

[0188] The event Pips_d_to_e_Pick_f_ul_tips_g_to_h is used, where d_to_e represents the coordinate range of the TIP picked by the pipette; f represents the type of TIP selected; and g_to_h represents the coordinate range of the TIP area.

[0189] Annealing Automaton Model

[0190] After spin coating is applied to the glass slide, an annealing process is performed. This automated machine represents the annealing process flow, which includes picking the slide onto the heating stage (Put_Glass_Into_Heater), starting the annealing process (Heating_n), and ending it (Heating_n_fn), where n represents the annealing event for that slide. The occurrence of the event Pick_Heated_Glass_n indicates the end of the slide processing.

[0191] Model placed in the painting area

[0192] This indicates that the carrier from the preparation area will be transported to the coating area to facilitate subsequent spin coating processes.

[0193] Model of the painting area

[0194] Each shelf can hold 8 slides. When the Pick_Heated_Glass_8 event occurs, it means that all the slides in one shelf have been processed and this shelf can be transported to the photography area.

[0195] Photo model

[0196] Take photos of the carrier, namely Photo_taking_1 and Photo_taking_2.

[0197] (2-2) Scheduling Ideas

[0198] For semiconductor material production tasks with the same processing flow and cycle, the key to improving overall efficiency lies in spin coating efficiency. Therefore, the spin coater, as the core processing equipment, is identified as the main control object for its spin coating process, while robotic arm operation, liquid transfer, and photography are considered as limited resources.

[0199] Following the principle of minimizing the dwell time of each node after completing a process or operation (excluding necessary time), this embodiment aims to solve for the shortest total processing time. To achieve this goal, this scheme specifically employs a cyclic scheduling algorithm for solving the problem.

[0200] Elastic model (taking a single-layer support structure capable of supporting 8 glass slides as an example):

[0201] The iterative algorithm is designed to solve the complex constraint problem of the slide processing sequence between shelves, which is directly related to the optimization of production efficiency and resource management. Specifically, it deals with how to determine the optimal start time for scheduling slides on the next shelf after all slide processing tasks on the current shelf have been completed. To quantify this process, this embodiment introduces the concept of T1, which represents the entire cycle time from the start of processing the first slide on the current shelf to the start of processing the first slide on the next shelf.

[0202] The core of this loop algorithm lies in its ability to flexibly adjust T1. By dynamically and elastically scaling T1, the system can make efficient scheduling decisions while ensuring smooth resource allocation and avoiding deadlock. Each loop iteration includes a judgment step, in which the system evaluates whether a valid and feasible slide scheduling path can be successfully found under the current T1 constraint.

[0203] The fundamental purpose of this process is to find the shortest possible time (T1) to achieve the processing target under given and limited resource conditions through continuous trial and path optimization. This approach not only helps improve production efficiency and ensures that the slides on each shelf can be processed as quickly as possible, but also maximizes the use of existing resources and reduces unnecessary waiting time and resource waste.

[0204] To illustrate with a specific example, when T1 is initially set to 0 seconds in this embodiment, the system will initiate a complete cyclic calculation process, which includes evaluating whether the controller can successfully find a valid scheduling path under the current T1 setting. If the controller solves for an empty path, i.e., it fails to find a scheduling scheme that meets the conditions, the cyclic scheduling algorithm will increment T1 and re-execute the above process until the controller can solve for a valid scheduling path.

[0205] In a discrete event system, when the system state changes, there may be multiple executable subsequent events that cause different transitions in the system state. These events are called path bifurcation events.

[0206] The "pairwise deprivation" strategy achieves deterministic operation of discrete-event systems by precisely prioritizing path-branching events and combining this with preemption control methods. The core of this strategy lies in its ability to effectively manage and coordinate the interactions between various events within the system, ensuring that the system's behavior remains controllable and predictable during complex state transitions. Through this strategy, the system can avoid uncertainties caused by disordered execution or event conflicts, thereby significantly improving the system's stability and reliability.

[0207] In the semiconductor material production process, such as in the glass slide processing, it is stipulated that the moving carrier event always precedes the robotic arm grasping the glass slide event. That is, when these two path forks in the scheduling path, the moving carrier event is executed first, and then the robotic arm grasping the glass slide event is executed.

[0208] While meeting the requirements of the processing technology, the scheduling path solved by the loop algorithm is optimized by comprehensively considering factors such as the order of processing, resource utilization efficiency, and time constraints, under the priority relationship of pairwise deprivation, to find the scheduling path with the highest efficiency and the shortest total processing time.

[0209] In this embodiment, the code implementation process of the pairwise preemption algorithm can be described as follows:

[0210]

[0211]

[0212] Example

[0213] Taking semiconductor fabrication as an example, the specific implementation steps of the high-throughput cycle scheduling method provided in this embodiment include:

[0214] (1) Input production data, including: the number of processing objects (num), the division of the work area for process processing, equipment information, and process flow information;

[0215] For example, the work area division information includes: preparation area, transfer area, core processing area, secondary processing area, analysis area, and final processing area. Equipment information includes: processing equipment (2 spin coaters, 4 annealing tables), auxiliary equipment (two robotic arms), and loading devices (carriers) for loading the processed objects.

[0216] The process flow information (glass slide process flow) and the process flow information (carrier process flow) of the processing object, based on the description of the work area and process steps, are as follows:

[0217] Glass slide process flow: Area to be coated = a => Spin coater = b => Flash evaporator = c => Hot stage = d => Area to be coated, where "Flash evaporator = c => Hot stage = d =>" are skippable process steps;

[0218] Carrier process flow: Preparation area = A => Coating area = B => Photo taking area = C => Transfer area;

[0219] The duration of each action in the designed process steps includes: time to place the slide into the spin coater, time to place the slide into the carrier, spin coating time, time to place the slide into the hot stage, annealing time, time to remove the distributor, time to remove the distributor, air blowing time, liquid aspiration time, liquid spraying time, time to move the carrier from the preparation area to the coating area, time to move the carrier from the coating area to the photography area, time to move the carrier from the photography area to the transfer area, flash evaporation time, and time to place the slide into the flash evaporator, etc.

[0220] (2) Based on the discrete event monitoring system, a process flow automaton model is constructed for each process flow, and the work area, process step and equipment are used as scheduling path nodes to generate candidate scheduling paths that meet the process requirements and do not cause deadlock; the candidate scheduling path with the shortest total processing time for each production cycle is taken as the optimal scheduling path; among them, the process flow automaton model represents each process step according to a unified naming rule.

[0221] The construction of the automated machine model for the process steps specifically includes the following steps:

[0222] (2a-1) Multiple sets of first combined automated machines for generating process steps:

[0223] (2a-1-1) Generate a core processing automaton for each processing object, which is a single process step (e.g., spin coating process step); where each process step corresponds to one core processing automaton;

[0224] (2a-1-2) Generate auxiliary operation automata (such as air blowing, liquid picking, liquid spraying, etc.) for each core processing automaton;

[0225] (2a-1-3) Divide all core processing automatic machines of the same process step of all processing objects into several groups (grouping according to the number of core processing equipment that the work area can provide);

[0226] (2a-1-4) Perform a synchronous product operation on each group of automata (including each core processing automata and its auxiliary operation automata) for each process step to obtain the first joint automata of the group (also known as the first process step model);

[0227] (2a-2) Multiple sets of second combined automated machines for generating process steps:

[0228] (2a-2-1) Generate a sub-processing automaton for each processing object with a single process step (such as optional flash evaporation, optional annealing, etc.);

[0229] (2a-2-2) Generate auxiliary operation automata (such as spraying, picking, pipetting, etc.) for each sub-processing automaton;

[0230] (2a-2-3) Generation of the photo-taking automaton (also a type of sub-processing automaton)

[0231] (2a-2-4) Divide all automated machines for the same process step of all processed objects into several groups (based on the number of flash evaporation / annealing equipment, i.e., the number of equipment that can perform the current process step based on the input).

[0232] (2a-2-5) Perform a synchronous integration operation on each group of automata (subprocessing automata and auxiliary operation automata involved in the current process step) for each process step to obtain the second joint automata of the group (also known as the second process step model);

[0233] In this embodiment, the corresponding process steps are divided into core processing and secondary processing based on the work area to which each process step of the processing object belongs.

[0234] (2a-3) Construct a model for the loading and unloading process of the carrier:

[0235] (2a-3-1) Automatic machine for the carrier to enter the core processing area;

[0236] (2a-3-2) The carrier leaves the core processing area and enters the secondary processing area automaton;

[0237] (2a-3-3) Automatic machine for carrier to leave secondary processing area;

[0238] (2a-3-4) Synchronize the above-mentioned automatic loading and unloading machines to form an overall loading and unloading model of the carrier, that is, the loading and unloading model of the process flow;

[0239] (2a-4) Construct global constraints:

[0240] (2a-4-1) Robotic arm operation constraint automatic machine

[0241] (2a-4-2) Core Processing Area Device Access Constraint Automaton

[0242] (2a-4-3) Sub-processing area equipment access constraint automaton

[0243] (2a-4-4) Synchronously integrate all the constrained automata involved in each process flow with the model (loading and unloading model, first and second process step model) to generate a global controlled system model, i.e., the process flow automata model, denoted by SPEC (abbreviation for control specification).

[0244] Therefore, a globally controlled model can be synthesized using Supermajor Controllable Sublanguage to obtain the optimal scheduling path. In this embodiment, the output optimal scheduling path is a scheduling string, which can then generate its Gantt chart or other visualization results.

[0245] In this embodiment, when generating the optimal scheduling path (final_string_found), a serial for loop strategy or a multi-threaded parallel strategy can be adopted. The specific implementation steps are as follows:

[0246] (2b-1) Serial for loop strategy

[0247] 2b-1-1) Initialize the optimal scheduling path "final_string_found" to the error "false", and initialize the outer loop boundary, that is, initialize the value ranges of the time intervals T1 and T2 respectively: T1∈[out_lower_bound, out_upper_bound], T2∈[in_lower_bound, in_upper_bound];

[0248] 2b-1-2) The outer loop `for T1 = out_lower_bound to out_upper_bound` is executed serially:

[0249] Based on the set step value, the current value of time interval T1 is assigned to the start interval Offset of adjacent workpieces on the same device;

[0250] Under the start interval Offset, the first process step model of each group is called, and the first scheduling sequence Coat_String is generated based on the synchronization product of the called first process step model;

[0251] 2b-1-2) Inner for T2 = in_lower_bound to in_upper_bound executed serially:

[0252] Based on the set step value, the current value of time interval T2 is assigned to the starting offset Offset_between_coaters of adjacent workpieces on different devices;

[0253] Under the starting offset Offset_between_coaters, the first scheduling sequence Coat_String is synchronized to obtain the initial scheduling plan Plant1;

[0254] Then, based on the synchronous product of Plant1, the second process step model, and the loading and unloading model, the second scheduling plan Plant2 is obtained;

[0255] Finally, a supervisory control check is performed based on Plant2 and the control specification SPEC. If the check passes, the final scheduling result Super is output.

[0256] For example, by inputting Plant2 and the control specification SPEC into the supervisory controller synthesis software SupCon, the corresponding scheduling result Super can be obtained, that is, setting the optimal scheduling path "final_string_found" to the error "true", and setting the current T1 and T2 to the final search values ​​found_out_bound and found_in_bound, respectively.

[0257] 2b-1-3) When "final_string_found = true", the outer loop will be exited;

[0258] 2b-1-4) Returns (found_out_bound, found_in_bound) and the corresponding candidate scheduling path string Schedule_String;

[0259] 2b-1-5) By depriving candidates of their rights in pairs, the optimal scheduling path is found among the candidate scheduling paths.

[0260] (2b-2) Multi-threaded parallel strategy

[0261] Objective: To schedule all (T1,T2) combinations in parallel, collect feasible solutions, and select the smallest (T1,T2) pair.

[0262] 2b-2-1) Input:

[0263] The range of time interval T1: T1∈[out_lower_bound, out_upper_bound];

[0264] The range of time interval T2 is: T2∈[in_lower_bound, in_upper_bound];

[0265] 2b-2-2) Generate a list of all (T1,T2) combinations based on the set value step;

[0266] 2b-2-3) Perform the following in parallel for each combination (T1,T2):

[0267] Assign the current value of time interval T1 to Offset, and call the first process step model of each group under Offset. Generate the first scheduling sequence Coat_String(T1) based on the synchronization product of the called first process step model.

[0268] Under Offset_between_coaters, calculate the synchronization product of Coat_String(T1) to obtain the initial scheduling plan Plant1(T1,T2);

[0269] The second scheduling plan Plant2(T1,T2) is obtained by synchronizing Plant1(T1,T2), the second process step model, and the loading and unloading model.

[0270] The supervisory controller is verified based on Plant2(T1,T2) and the control specification SPEC. If the check passes, the final scheduling result Super r(T1,T2) is output. If it is not empty, the feasible solution (T1,T2,Scheduling_String) is output.

[0271] Finally, by stripping away items one by one, a unique path was found;

[0272] 2b-2-4) After the parallel tasks are completed, collect the set of all feasible solutions;

[0273] 2b-2-5) Select the smallest (T1,T2) combination (lexicographical order or T1 first, then T2);

[0274] 2b-2-6) Returns the minimum (T1,T2) and the corresponding Schedule_String.

[0275] (3) Distribute the corresponding scheduling information to each work equipment in the work area based on the optimal scheduling path, and execute the processing operation of the processing object based on the received scheduling information.

[0276] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0277] Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include at least one of those features.

[0278] Any process or method described in the flowchart or otherwise in this specification can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0279] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of this application.

Claims

1. A periodic scheduling method for high throughput, characterized in that, Includes the following steps: Input production data, including: work area division information for process processing, equipment information, and process flow information; The equipment information includes processing equipment, auxiliary equipment, and loading devices for loading the processed objects. The process flow information includes: process flow information of the processed object and process flow information of the loading device based on the description of the work area and process steps; wherein, the process flow information includes: the working equipment, the number of working equipment, the logical relationship between the working equipment and the process steps, and the duration of each action in each process step in each work area. Based on the discrete event system supervisory control simulation platform, a process flow automaton model is constructed for each process flow. The work area, process step and equipment are used as scheduling path nodes to generate candidate scheduling paths that meet the process requirements and are not deadlocked. The candidate scheduling path with the shortest total processing time for each production cycle is selected as the optimal scheduling path. The process automaton model represents each process step according to a unified naming rule to represent the name, action, working equipment, position and status of each process step. Based on the optimal scheduling path, the corresponding scheduling information is distributed to each working device within the working area, and the relevant scheduling information of each working device is distributed to execute the processing operation of the processing object based on the received scheduling information; Specifically, a two-level cyclic scheduling algorithm is used to generate candidate scheduling paths that meet process requirements and are not deadlocked. The candidate scheduling path with the shortest total processing time for each production cycle is selected as the optimal scheduling path. This includes: When executing the process steps in the designated work area, two processing equipment are configured for the process steps in the current designated work area. The processing objects on the same layer of the loading device are divided into two parts according to odd and even. The time interval between the start of processing of two adjacent processing objects on the same layer is defined as T1, and the time interval between the start of the two processing equipment is defined as T2. Within the boundary range of T1, the outer loop dynamically adjusts T1 to obtain the shortest time interval T1 under the condition of at least one candidate scheduling path that meets process requirements and avoids deadlock. Within the boundary range of T2, the inner and outer loops dynamically adjust the time interval T2 to obtain the shortest time intervals T1 and T2 under the condition of at least one candidate scheduling path that meets the process requirements and is not deadlocked. Based on the shortest time intervals T1 and T2, the optimal scheduling path is selected from the candidate scheduling paths based on the shortest total processing time.

2. The method of claim 1, wherein, The construction of a process flow automata model specifically includes: For each process step in the process flow, generate an automated processing machine, and divide the generated automated processing machines into several groups based on the number of available processing equipment; perform a synchronous product operation on each group to obtain the process step model for each group. Generate a first automatic loader entering the work area, a second automatic loader leaving the current work area and entering the next work area, and a third automatic loader leaving the work area for the loading device; perform synchronous integration on the automatic loaders involved in the current process flow to form a loading entry and exit model of the process flow; Construct auxiliary equipment operation constraint automata and processing equipment access constraint automata, and generate process flow automaton model based on the loading and unloading model of the current process flow, the process step model involved in the current process flow, and the synchronous product of constraint automata.

3. The method of claim 1, wherein, Candidate scheduling paths that meet process requirements and are not deadlocked will be generated. The candidate scheduling path with the shortest total processing time for each production cycle will be used as the optimal scheduling path. Define the time interval between the start of processing of the first processing object in the adjacent layer of the loading device as T0; Within the adjustable range of the set time interval T0, the time interval T0 is iteratively and elastically adjusted to obtain the shortest time interval T0 under the condition that it meets the process requirements of the process flow and is not deadlocked. Based on the shortest time interval T0, the optimal scheduling path is selected from the candidate scheduling paths based on the shortest total processing time. When executing the process steps in a designated work area, the same layer of the loading device is divided into multiple parts, and each part is executed synchronously by multiple working devices.

4. The method of claim 1, wherein, The scheduling strategies used to generate candidate scheduling paths include at least one of the following: Single-threaded scheduling strategy based on for loop: Set up inner and outer for loops, where the outer for loop is used to serially traverse the different values ​​of the time interval T1 starting from the lower limit of the boundary range, and the inner for loop is used to serially traverse the different values ​​of the time interval T2 starting from the lower limit of the boundary range. Each iteration of the outer for loop must wait for the inner for loop to complete before continuing to iterate over the next value of T1. For each iteration of the inner for loop, it is necessary to wait for the supervisory controller to verify before continuing to iterate to the next value of T2; the supervisory control check is used to determine whether the current time intervals T1 and T2 can generate a candidate scheduling path that meets the process requirements and is not deadlocked. The first pair of time intervals (T1, T2) verified by the supervisory controller is used as the shortest time intervals T1 and T2. The optimal scheduling path is then obtained based on the candidate scheduling paths corresponding to the shortest T1 and T2. Multi-threaded parallel scheduling strategy: Based on the boundary range of T1 and T2 and the step size of value traversal, generate all possible value combinations (T1, T2) between intervals T1 and T2; For all value combinations (T1, T2), candidate scheduling paths are generated that meet the process requirements and are not deadlocked. The optimal scheduling path is then obtained by selecting the candidate scheduling path corresponding to the smallest (T1, T2).

5. The method of claim 1, wherein, The process flow information also includes the priority of each process step. When selecting the optimal scheduling path, candidate scheduling paths that do not meet the step priority are cut off.

6. The method of claim 1, wherein, It also includes path optimization of the generated candidate scheduling paths. When adjacent path nodes in the scheduling path are loading device moving steps and auxiliary equipment operation steps, the priority of the loading device moving steps is set to be higher than that of the auxiliary equipment operation steps. When the system state changes and multiple path fork events exist, a pairwise preemption strategy is adopted to sort the path fork events by priority. Specifically, the pairwise preemption strategy is as follows: traverse the list of events to be executed, if the priority of path fork event A is higher than that of path fork event B, then return the execution order of events as A before B; otherwise, return B before A; the pairwise preemption strategy is terminated when there are no cross-fork events in the scheduling path.

7. The method of claim 1, wherein, The work area includes a preparation area, a transfer area, a core processing area, a secondary processing area, an analysis area, and a final processing area. The preparation area is used to transfer the workpiece to the loading device; The transfer area is used to allocate the loading device for loading the processed objects to the corresponding work area for processing, according to the cycle of the processed objects; The core processing area is used to perform designated core process processing on the objects being processed on the loading devices located in this work area; The secondary processing zone is used for secondary processing of the workpieces on the loading devices located in this work area, and all workpieces on the loading devices located in the secondary processing zone have completed the core processing. The analysis area is used to analyze and inspect the processing quality of the workpieces on the loading device located in this work area, and all workpieces on the loading device located in the analysis area have completed the next process. The loading device performs analysis and testing operations. After the processing object on the loading device completes the processing in the core processing area and the secondary processing area, it is moved to the analysis area. The processing quality of the processing object is comprehensively tested using the corresponding testing equipment and analysis methods. The end area serves as a storage area, used to store loading devices for processed objects that have passed quality inspection.

8. The method according to any one of claims 1 to 7, characterized in that, The operational scenarios corresponding to the production data include semiconductor manufacturing, biomedical research and development, or lithium battery manufacturing.

Citation Information

Patent Citations

  • Symmetric discrete-event system online control law reconfiguration method

    CN106227036A

  • Workflow modeling method suitable for multi-scale high-throughput material calculation

    CN113723931A