A Method and System for Production Scheduling of Display Panels Based on Agent Collaboration

Abstract

This invention discloses a method and system for scheduling display panel production based on agent collaboration. The method first performs real-time state collaborative update operations on multiple agents in the display panel production system to obtain a collaborative state set containing agent identification information and corresponding production-related state descriptions. Next, based on this set, a set of inter-agent collaboration strategies is generated, containing information interaction rules and task allocation priority descriptions. Then, a pre-configured negotiation mechanism is invoked to perform conflict detection and negotiation processing on the collaboration strategy set, resulting in an optimized collaboration strategy after conflict resolution. Finally, a dynamic scheduling operation for display panel production is executed according to the optimized collaboration strategy, generating a set of production scheduling instructions containing process execution sequences and equipment allocation schemes. This invention achieves optimal scheduling for display panel production through agent collaboration and dynamic scheduling, improving production efficiency and quality.

Description

Technical Field

[0001] This invention relates to the field of computer technology, specifically to a method and system for scheduling the production of display panels based on intelligent agent collaboration. Background Technology

[0002] In the display panel manufacturing industry, with the continuous expansion of production scale and the increasing complexity of production processes, the requirements for production scheduling are also becoming more demanding. Traditional display panel production scheduling methods mainly rely on human experience for decision-making, which is difficult to adapt to the needs of modern large-scale production. Under this traditional approach, the flow of information between various links in the production process is not smooth, leading to information lag and inaccuracy. For example, the operating status of equipment cannot be fed back to the scheduler in a timely manner, making it impossible for the scheduler to make reasonable scheduling decisions based on the actual situation, easily resulting in idle or overused equipment and reduced production efficiency.

[0003] Meanwhile, traditional methods lack a scientific basis for task allocation, often employing fixed allocation patterns that cannot be dynamically adjusted based on the urgency of orders and the actual workload of agents. This results in some urgent orders not being processed in a timely manner, while some agents are burdened with too many tasks, causing task backlogs and production delays.

[0004] Furthermore, traditional methods lack effective conflict resolution mechanisms in agent collaboration. When conflicts arise among multiple agents regarding resource allocation and task scheduling, manual intervention is often required, which is not only inefficient but also prone to subjectivity and irrationality in decision-making. These problems severely restrict the efficiency and quality of display panel production and increase production costs, thus urgently requiring a more intelligent and efficient production scheduling method to address these issues. Summary of the Invention

[0005] This invention provides a method and system for scheduling the production of display panels based on intelligent agent collaboration.

[0006] In a first aspect, embodiments of the present invention provide a display panel production scheduling method based on intelligent agent cooperation, applied to a display panel production scheduling system, the method comprising:

[0007] A real-time state collaborative update operation is performed on multiple intelligent agents in the display panel production system to obtain a collaborative state set for each intelligent agent. The collaborative state set includes intelligent agent identification information and corresponding production-related state descriptions.

[0008] Based on the set of cooperative states, a set of cooperative strategies among agents is generated. The set of cooperative strategies includes information interaction rules and task allocation priority descriptions among agents.

[0009] The pre-configured negotiation mechanism is invoked to perform conflict detection and negotiation on the set of cooperation strategies, and an optimized cooperation strategy after conflict resolution is generated.

[0010] The optimized collaboration strategy is used to perform dynamic scheduling operations for display panel production, generating a set of production scheduling instructions that includes process execution sequences and equipment allocation schemes.

[0011] Secondly, embodiments of the present invention provide a display panel production scheduling system, comprising:

[0012] processor;

[0013] Storage device, on which computer programs are stored,

[0014] When the computer program is executed by the processor, the processor implements any of the aforementioned agent-based display panel production scheduling methods.

[0015] This invention provides a readable storage medium storing a program or instructions, which, when executed by a processor, implement the steps of the display panel production scheduling method based on intelligent agent cooperation.

[0016] Therefore, the embodiments of the present invention significantly improve the efficiency and quality of display panel production scheduling, and realize the intelligent, dynamic and efficient production process.

[0017] First, real-time status collaborative update operations are performed on multiple intelligent agents in the display panel production system, which comprehensively and timely grasps the production-related status of each intelligent agent. This provides an accurate and real-time data foundation for subsequent scheduling decisions, enabling the scheduling system to formulate strategies based on accurate information and avoiding production chaos caused by information lag or inaccuracy.

[0018] Secondly, the set of collaborative strategies among agents, generated based on the set of cooperative states, clarifies the rules for information interaction and the priority of task allocation among agents, effectively coordinating the work among them. Reasonable information interaction rules ensure that agents can share key information in a timely and accurate manner, improving information flow efficiency; scientific task allocation priorities ensure that tasks are reasonably assigned to appropriate agents, avoiding unreasonable task accumulation and resource waste.

[0019] Furthermore, by invoking a pre-configured negotiation mechanism to detect and resolve conflicts in the set of collaboration strategies, potential conflicts that may arise during agent collaboration are identified and resolved in a timely manner. This not only avoids conflicts hindering production progress but also improves the collaboration efficiency between agents and the overall stability of production by optimizing collaboration strategies.

[0020] Finally, based on the optimized collaboration strategy, dynamic scheduling operations for display panel production are executed, generating a set of production scheduling instructions containing process execution sequences and equipment allocation schemes. This instruction set can be dynamically adjusted according to the actual production situation, ensuring the reasonable execution sequence of processes and the effective allocation of equipment. Dynamic scheduling enables the production process to respond quickly to various changes, such as equipment failures and order changes, improving production flexibility and adaptability. Ultimately, optimal scheduling for display panel production is achieved, improving production efficiency and product quality while reducing production costs. Attached Figure Description

[0021] Figure 1 This is a flowchart of a display panel production scheduling method based on intelligent agent collaboration, provided in an embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram of the basic structure of a display panel production scheduling system provided in an embodiment of the present invention. Detailed Implementation

[0023] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] See Figure 1 As shown, this figure is a flowchart of a display panel production scheduling method based on intelligent agent cooperation provided by an embodiment of the present invention. This method can be applied to a display panel production scheduling system. Figure 1 As shown, the method may include steps 110-140.

[0025] Step 110: Perform real-time state collaborative update operation on multiple agents in the display panel production system to obtain a collaborative state set of each agent. The collaborative state set includes agent identification information and corresponding production-related state descriptions.

[0026] In the display panel production scheduling scenario of this invention embodiment, multiple intelligent agents are deployed in the display panel production system. Each of these agents undertakes different production tasks, such as equipment operation, material transportation, and quality inspection. To accurately grasp the real-time status of each intelligent agent, the display panel production scheduling system initiates a real-time status collaborative update operation.

[0027] In detail, the display panel production scheduling system has established stable data transmission channels with each intelligent agent, continuously collecting status information of the agents through sensors, monitoring equipment, and communication interfaces. For agents responsible for equipment operation, it collects information such as the current operating mode, operating speed, and whether there are any abnormal alarms; for agents responsible for material transportation, it obtains information such as the transportation location, transportation progress, and remaining transportation volume; for agents responsible for quality inspection, it collects information such as the number of inspected products, the number of qualified products, and the number of unqualified products.

[0028] After collecting this raw information, the display panel production scheduling system performs preliminary cleaning and organization. Noisy, duplicate, and invalid data are removed to ensure accuracy and completeness. Then, the identification information of each agent is associated with its corresponding production-related state description and stored in a data structure to form a collaborative state set. Each record in this set comprehensively reflects the identity of an agent and its current production status.

[0029] Step 120: Generate a set of cooperation strategies among agents based on the set of cooperative states. The set of cooperation strategies includes information interaction rules and task allocation priority descriptions among agents.

[0030] After obtaining the collaborative state set of each agent, the display panel production scheduling system needs to further formulate a set of cooperation strategies among the agents. The system analyzes the information in the collaborative state set, considering the relationships between different agents and the overall needs of the production process. The formulation of information interaction rules must ensure that agents can share necessary information in a timely and accurate manner, avoiding production problems caused by information delays or inconsistencies. The task allocation priority description specifies how tasks should be allocated to appropriate agents under different circumstances to ensure that production tasks are completed on time and with high quality.

[0031] In an optional embodiment, step 120 includes:

[0032] Step 121: Analyze the production-related state descriptions of each agent in the collaborative state set, and extract a subset of state features including equipment operating status, order completion progress, and process waiting queue length.

[0033] The display panel production scheduling system performs in-depth analysis of the collaborative state set, and filters out information related to equipment operating status, order completion progress, and process waiting queue length from the production-related state descriptions of each intelligent agent.

[0034] For equipment operating status, a detailed analysis is conducted to determine whether the equipment is in normal operation, under maintenance, or in standby mode, as well as indicators such as operating efficiency and energy consumption. For order completion progress, the completed workload, remaining workload, and estimated completion time for each order are determined. For process waiting queue length, the number of tasks waiting to be processed before each process is tallied to reflect the busyness and load of that process.

[0035] The selected information is organized and categorized to form a subset of state features, which highlights key information in the panel production process.

[0036] Step 122: Perform availability assessment processing on the device operating status in the aforementioned state feature subset to generate a time distribution matrix of idle and busy periods.

[0037] The display panel production scheduling system performs availability assessments on the operating status of equipment within a subset of status characteristics, taking into account factors such as equipment maintenance plans, fault history, and current operational tasks.

[0038] By analyzing the equipment's operation records and maintenance plans, the availability of the equipment in different time periods can be determined. If the equipment is under maintenance or has a planned downtime, it is unavailable during that period; if the equipment is performing a task, it is busy during the task execution period; the equipment is idle only when it has no task and is not under maintenance.

[0039] Arrange the idle and busy periods of the equipment in chronological order to form a time distribution matrix. This matrix accurately represents the status of the equipment in different time periods and can be used to rationally arrange production tasks and avoid overuse or idleness of the equipment.

[0040] Step 123: Prioritize the order completion progress in the subset of state features and generate an order priority sequence based on the order delivery time requirements and product type complexity.

[0041] The display panel production scheduling system prioritizes orders based on their completion progress within a subset of status characteristics. It considers two key factors: delivery time requirements and product complexity. Orders with tighter delivery times are given higher priority; orders with more complex products, due to their greater production difficulty, also receive a slightly higher priority. By comprehensively evaluating these two factors, all orders are ranked, generating an order priority sequence. During the ranking process, the delivery time and product complexity of each order are quantitatively evaluated. For example, delivery time is scored based on its proximity to the current time, and product complexity is scored based on factors such as process difficulty and required materials. Orders are then ranked according to these scores to ensure that higher-priority orders are processed first.

[0042] Step 124: Perform load balancing analysis on the process waiting queue length in the state feature subset, and calculate the average waiting time and queue backlog rate for different agents corresponding to the processes.

[0043] The display panel production scheduling system performs load balancing analysis on the waiting queue lengths of processes within a subset of state characteristics. It calculates the average waiting time by statistically analyzing the waiting queue length for each process corresponding to each agent and considering the processing speed of that process. The average waiting time reflects how long tasks before a process can be processed. Simultaneously, it calculates the queue backlog rate, which is the ratio of the waiting queue length to the maximum capacity of the process. This ratio measures the load pressure on the process. By comparing the average waiting time and queue backlog rate for different agents across processes, it can identify which processes are overloaded and which have remaining processing capacity. This information can be used for reasonable task allocation and load balancing.

[0044] Step 125: Construct information interaction rules based on the time distribution matrix, the order priority sequence, and the average waiting time. The information interaction rules include state data types and interaction frequency parameters that are shared in real time among intelligent agents.

[0045] The display panel production scheduling system constructs information interaction rules based on the previously obtained time distribution matrix, order priority sequence, and average waiting time. When determining the types of state data to be shared in real-time between agents, key information in the production process is considered. For example, information such as equipment operating status, order priority, and process waiting time needs to be shared promptly. This information helps agents understand the status of other agents, thereby better coordinating their work. The interaction frequency parameter is determined based on actual production needs and equipment performance. For critical information, such as equipment fault information and urgent order information, real-time sharing is required, resulting in a higher interaction frequency. For relatively stable information, such as long-term equipment operating statistics, the interaction frequency can be appropriately reduced. By rationally setting information interaction rules, smooth information flow between agents can be ensured, improving the efficiency of production scheduling.

[0046] Step 126: Combine the queue backlog rate and the order priority sequence to generate a task allocation priority description, which includes a rule for prioritizing urgent orders and a rule for prioritizing low-load agents.

[0047] The display panel production scheduling system combines queue backlog rate and order priority sequence to generate task allocation priority descriptions. For urgent orders, due to their tight delivery times, they are prioritized for allocation to suitable agents. When selecting agents, the system considers their processing capacity, current load, and suitability for the order. Simultaneously, a rule prioritizing low-load agents is followed. Agents with low queue backlog rates indicate lighter current loads and more processing capacity to handle new tasks. Therefore, tasks are prioritized for these low-load agents to achieve balanced task allocation and improve overall production efficiency.

[0048] Step 130: Invoke the pre-configured negotiation mechanism to perform conflict detection and negotiation on the set of cooperation strategies, and generate an optimized cooperation strategy after conflict resolution.

[0049] The display panel production scheduling system invokes a pre-configured negotiation mechanism to detect and resolve conflicts in the set of collaboration strategies. During the collaboration process of intelligent agents, various conflicts may arise, such as conflicts over information interaction rules and task allocation.

[0050] The negotiation mechanism performs a detailed check on each rule and description in the set of cooperation strategies. For information exchange rules, it checks for conflicts in the frequency and data types of interactions between different agents; for task allocation priority descriptions, it checks for situations such as multiple tasks being assigned to the same agent simultaneously or an agent being burdened with too many tasks.

[0051] When a conflict is detected, the negotiation mechanism initiates a negotiation process. This involves the relevant agents exchanging information and negotiating to find the best solution to the conflict. For example, if inconsistent information exchange frequencies between two agents cause data delays, the interaction frequency will be adjusted through negotiation to achieve consistency.

[0052] After conflict detection and negotiation, an optimized collaboration strategy is generated to resolve the conflict. This strategy can better coordinate the work between agents, avoid conflicts, and improve productivity.

[0053] As one implementation, step 130 includes:

[0054] Step 131: Perform conflict type identification processing on the information interaction rules in the collaboration strategy set, and extract the state data lag conflict caused by mismatch in interaction frequency and the information comprehension deviation conflict caused by inconsistency in shared data types.

[0055] The display panel production scheduling system identifies conflict types in the information interaction rules within the collaboration strategy set. It carefully analyzes two key factors: interaction frequency and shared data type. For interaction frequency, it compares the consistency of interaction frequencies between different agents. If an agent's interaction frequency is too high or too low, it can lead to lag in state data. For example, if one agent sends data at a high frequency while another receives data at a low frequency, the receiver may not receive the latest data in a timely manner, resulting in biased production decisions. For shared data type, it checks whether the data types shared by different agents are consistent. If the data type sent by one agent is inconsistent with the data type expected to be received by other agents, it can lead to misunderstandings. For example, if one agent sends energy consumption data while another expects to receive operating speed data, the receiver may misunderstand the information, affecting production coordination. These identified conflict types are organized and recorded for subsequent negotiation and resolution.

[0056] Step 132: Perform conflict scenario localization processing on the task allocation priority description in the collaboration strategy set to identify equipment competition conflicts caused by overlapping allocation of urgent orders and task backlog conflicts caused by insufficient handling capacity of low-load intelligent agents.

[0057] The display panel production scheduling system identifies conflict scenarios by analyzing the task allocation priority descriptions in the collaboration strategy set. It focuses on the allocation of urgent orders and the processing capacity of low-load agents. For overlapping urgent order allocations, it checks whether multiple urgent orders are simultaneously assigned to the same device or agent. If this occurs, it can lead to device contention, preventing the device from processing multiple urgent orders concurrently and impacting delivery time. For low-load agents with insufficient processing capacity, it analyzes whether they can complete new tasks on time. If a low-load agent's processing capacity is limited and it cannot complete newly assigned tasks within the specified time, it will cause task backlog and conflict, affecting overall production progress. When identifying conflict scenarios, it records detailed information about the agent identifiers, task identifiers, and specific details of the conflict.

[0058] In an optional embodiment, the conflict scenario localization processing of the task allocation priority descriptions in the collaboration strategy set, identifying equipment competition conflicts caused by overlapping allocation of urgent orders and task backlog conflicts caused by insufficient handling capacity of low-load intelligent agents, includes:

[0059] Step 1321: Extract the target agent allocation list of emergency orders from the task allocation priority description, and count the number of times the same device is simultaneously allocated by multiple emergency orders as the device overlap allocation frequency.

[0060] The display panel production scheduling system extracts the target agent allocation list for urgent orders from the task allocation priority description. It then examines which agents each urgent order is assigned to and the equipment used by those agents. Next, it counts the number of times the same equipment is simultaneously assigned to multiple urgent orders; this number is the equipment overlap allocation frequency, reflecting the intensity of equipment competition. A high equipment overlap allocation frequency indicates that the equipment is being overused in urgent order allocation, easily leading to equipment competition conflicts.

[0061] Step 1322: Calculate the severity parameter of the equipment competition conflict based on the frequency of equipment overlap allocation and the maximum number of orders processed by the equipment in a single transaction.

[0062] The display panel production scheduling system calculates the severity of equipment contention based on the frequency of equipment overlap allocation and the maximum number of orders a single equipment can process. The maximum number of orders a single equipment can process is an inherent attribute of the equipment, representing the upper limit of the number of orders the equipment can handle simultaneously. By comparing the frequency of equipment overlap allocation and the maximum number of orders a single equipment can process, the severity of equipment contention can be assessed. If the frequency of equipment overlap allocation exceeds the maximum number of orders a single equipment can process, the contention is relatively severe; if the frequency is close to the maximum number of orders a single equipment can process, the contention is somewhat risky; if the frequency is much lower than the maximum number of orders a single equipment can process, the contention is relatively mild.

[0063] Step 1323: Extract the process processing capability parameters of the low-load agent from the task allocation priority description. The process processing capability parameters include the number of processes that can be processed per unit time and the maximum continuous working time.

[0064] The display panel production scheduling system extracts process processing capability parameters from the task allocation priority description of low-load agents. It focuses on two key parameters: the number of processes a low-load agent can process per unit time and its maximum continuous working time. The number of processes that can be processed per unit time reflects the number of processes the agent can complete within a given time, demonstrating its processing efficiency. The maximum continuous working time indicates the longest time the agent can work continuously; beyond this time, the agent may need to rest or undergo maintenance. By extracting these parameters, the processing capability of low-load agents can be accurately assessed.

[0065] Step 1324: Calculate the number of tasks assigned to the low-load agent and the total processing time of the tasks, and calculate the difference between the total processing time of the tasks and the maximum continuous working time as the remaining capacity.

[0066] The display panel production scheduling system tracks the number of tasks assigned to low-load agents and their total processing time. It examines the tasks currently being handled by each low-load agent and calculates their total processing time. Then, it subtracts the total processing time from the maximum continuous working time to obtain the remaining processing capacity. This remaining capacity represents the amount of task processing time a low-load agent can still handle without exceeding its maximum continuous working time. A large remaining capacity indicates that the low-load agent has ample processing capacity to handle new tasks; a small remaining capacity indicates that the low-load agent's processing capacity is nearing its limit, and new tasks should be assigned with caution.

[0067] Step 1325: When the remaining capacity is less than the preset capacity value, it is determined that there is a task backlog conflict caused by insufficient capacity of the low-load agent, and the agent identifier and task identifier involved in the task backlog conflict are recorded.

[0068] The display panel production scheduling system compares the remaining processing capacity with a preset processing capacity value. This preset capacity value is a threshold pre-set based on production experience and equipment performance. If the remaining processing capacity is less than the preset value, it indicates that the processing capacity of the low-load agents is insufficient and they may not be able to complete newly assigned tasks on time, leading to task backlog and conflicts. When a task backlog and conflict is detected, the system records the agent and task identifiers involved in detail. These records help with subsequent negotiation and task adjustments.

[0069] Step 1326: Combine the severity parameters of the device competition conflict and the scope of the task backlog conflict to generate a conflict scene location result that includes the conflict location coordinates and the degree of impact.

[0070] The display panel production scheduling system combines the severity parameters of equipment contention conflicts with the scope of task backlog conflicts to generate conflict scenario localization results. The conflict location coordinates pinpoint the exact location of the conflict, including the involved agents and equipment. The degree of impact is determined based on the severity parameters of the equipment contention conflict and the scope of the task backlog conflict. For example, if the equipment contention conflict is severe and the task backlog conflict involves multiple agents and tasks, the impact will be high; if the equipment contention conflict is minor and the task backlog conflict only involves a few agents and tasks, the impact will be low. This conflict scenario localization result provides detailed information for subsequent negotiation and processing, helping to quickly resolve the conflict.

[0071] Step 133: Input the state data lag conflict, the information comprehension bias conflict, the device competition conflict, and the task backlog conflict into the pre-configured conflict classification model to generate conflict analysis results that include conflict type identifiers and scope of impact.

[0072] The display panel production scheduling system inputs conflicts such as status data lag, information misunderstanding, equipment competition, and task backlog into a pre-configured conflict classification model. This model is a trained neural network capable of classifying and analyzing the input conflict information. Input data is received at the model's input layer and then processed through hidden layers. In the hidden layers, the model extracts and analyzes features of different types of conflicts, identifying key factors. Finally, at the output layer, the model generates conflict analysis results containing conflict type identifiers and their impact scope. The conflict type identifier clarifies the specific type of conflict, such as status data lag or equipment competition; the impact scope describes the degree and extent of the conflict's influence on the production process.

[0073] Step 134: Based on the conflict analysis results, invoke the auction negotiation sub-mechanism, whereby the agent involved in the conflict analysis results submits bidding parameters including equipment availability time, process efficiency, and resource consumption cost, and adjusts the task allocation priority based on the comprehensive evaluation results of the bidding parameters.

[0074] The display panel production scheduling system invokes the auction negotiation sub-mechanism based on conflict analysis results. Agents involved in the conflict analysis are required to submit bidding parameters, including equipment availability time periods, process efficiency, and resource consumption costs. Equipment availability time periods indicate the time periods during which an agent's equipment can be used; process efficiency reflects the speed at which an agent completes a process; and resource consumption costs include the energy consumption and raw material consumption costs of the equipment. The display panel production scheduling system comprehensively evaluates these bidding parameters, considering factors such as whether the equipment availability time periods match the task's time requirements, whether the process efficiency is high, and whether the resource consumption costs are low. Based on the comprehensive evaluation results, the task allocation priority is adjusted. Agents with excellent bidding parameters have their task allocation priority increased, and tasks are assigned to them preferentially; agents with poor bidding parameters have their task allocation priority decreased.

[0075] Step 135: Invoke the contract network negotiation sub-mechanism to negotiate and modify the information interaction rules. The leading agent publishes the interaction rules to adjust the bidding information, and the non-leading agents respond and provide feedback on constraints including data processing capabilities and communication bandwidth limitations. The information interaction rules are optimized based on the compatibility analysis results of the constraints.

[0076] The display panel production scheduling system invokes the contract network negotiation sub-mechanism to negotiate and modify information interaction rules. A leading agent is selected to coordinate globally. This leading agent generates bidding information to adjust interaction rules, including requirements for adjusting interaction data types and frequencies. This bidding information is then broadcast to non-leading agents. Non-leading agents generate constraint feedback information based on their data processing capabilities and communication bandwidth limitations. Data processing capability indicates the types and amounts of data an agent can handle; communication bandwidth limitations specify the amount of data an agent can transmit per unit of time. The display panel production scheduling system performs compatibility analysis on the constraint feedback information. It calculates the matching degree between interaction data type adjustment requirements and the data processing capabilities of each agent, as well as the adaptability of interaction frequency adjustment requirements and the communication bandwidth limitations of each agent. Based on the compatibility analysis results, the information interaction rules are optimized. For example, agents with high matching and adaptability are given priority in adjusting interaction rules; agents with low matching and adaptability are further negotiated to find suitable solutions.

[0077] In a preferred embodiment, step 135 includes:

[0078] Step 1351: Select the agent responsible for global coordination in the current production stage as the leading agent, and have the leading agent generate bidding information for adjusting interaction rules, including the requirements for adjusting the data type of interaction and the requirements for adjusting the frequency of interaction.

[0079] The display panel production scheduling system selects a leading agent as the overall coordinating agent for the current production stage. This leading agent possesses more comprehensive information and stronger coordination capabilities. Based on the actual situation and needs during production, the leading agent generates interaction rules to adjust the bidding information. Adjustments to interaction data types take into account which data types need to be added, reduced, or modified; adjustments to interaction frequency are determined based on data timeliness requirements and equipment performance. These adjustment requirements are then compiled into bidding information for distribution to non-leading agents.

[0080] Step 1352: Broadcast the interactive rule adjustment bidding information to the non-dominant intelligent agent, which then generates constraint feedback information based on its own data processing capabilities and communication bandwidth limitations.

[0081] The leading agent broadcasts the bidding information regarding the adjustment of interaction rules to all non-leading agents. Upon receiving the bidding information, each non-leading agent generates constraint feedback information based on its own data processing capabilities and communication bandwidth limitations. Data processing capabilities determine the types and amounts of data the agent can handle; if the data types in the bidding information exceed the agent's processing capabilities, the agent may be unable to process the data correctly. Communication bandwidth limitations restrict the amount of data an agent can transmit per unit of time; excessively high interaction frequency may lead to communication congestion. The non-leading agents then feed these constraints back to the leading agent for subsequent negotiation.

[0082] Step 1353: Perform compatibility analysis on the constraint feedback information, calculate the matching degree between the interaction data type adjustment requirements and the data processing capabilities of each agent, and the adaptability between the interaction frequency adjustment requirements and the communication bandwidth limitations of each agent.

[0083] The display panel production scheduling system performs compatibility analysis on constraint feedback information. It compares the interaction data type adjustment requirements with the data processing capabilities of each agent, calculating the matching degree. If the interaction data type adjustment requirements and the agent's data processing capabilities are perfectly matched, the matching degree is high; if they are partially matched, the matching degree is medium; and if they are completely mismatched, the matching degree is low.

[0084] Similarly, the interaction frequency adjustment requirements are compared with the communication bandwidth limitations of each agent to calculate the fit. If the interaction frequency adjustment requirements are within the agent's communication bandwidth limitations, the fit is high; if they are close to the communication bandwidth limitations, the fit is medium; and if they exceed the communication bandwidth limitations, the fit is low.

[0085] Step 1354: Select agents whose matching degree and fitness degree both exceed the preset fitness degree as rule-accepting agents, and calculate the average data processing capability and average communication bandwidth limit of the rule-accepting agents.

[0086] The display panel production scheduling system filters out agents whose matching and adaptation scores both exceed a preset adaptation score, designating them as rule-accepting agents. The preset adaptation score is a pre-defined threshold used to determine whether an agent can accept the adjusted interaction rules. The system also statistically analyzes the data processing capabilities and communication bandwidth limitations of these rule-accepting agents. The average data processing capabilities and average communication bandwidth limitations of these agents are calculated; these averages reflect the overall processing and communication capabilities of the rule-accepting agents.

[0087] Step 1355: Adjust the interaction data type based on the average data processing capability, retaining data types that can be processed by all agents and eliminating data types that do not match the data processing capability.

[0088] The display panel production scheduling system adjusts the types of interactive data based on the average data processing capability. It analyzes whether each data type can be supported by the data processing capability of the rule-receiving agents. Data types that can be processed by all agents are retained, ensuring that all rule-receiving agents can process interactive information correctly. Data types that do not match the data processing capability are eliminated to avoid information processing problems caused by data type mismatch.

[0089] Step 1356: Adjust the interaction frequency based on the average communication bandwidth limit, and set the interaction frequency to the maximum supportable frequency that does not exceed the communication bandwidth limit of each agent.

[0090] The display panel production scheduling system adjusts the interaction frequency based on the average communication bandwidth limit. It determines a maximum supportable frequency that does not exceed the communication bandwidth limit of each agent. Setting the interaction frequency to this maximum supportable frequency ensures timely information exchange without causing communication congestion, thus improving the efficiency and stability of information exchange.

[0091] Step 1357: Generate optimized information interaction rules that include the adjusted interaction data types and adjusted interaction frequencies.

[0092] The display panel production scheduling system organizes the adjusted types and frequencies of interactive data to generate optimized information interaction rules. These rules are better adapted to the data processing capabilities and communication bandwidth limitations of the intelligent agent, thereby improving the effectiveness of information interaction.

[0093] Step 136: Integrate the processing results of the auction negotiation sub-mechanism and the contract network negotiation sub-mechanism to generate an optimized collaboration strategy after conflict resolution.

[0094] The display panel production scheduling system integrates the processing results of the auction negotiation sub-mechanism and the contract network negotiation sub-mechanism. It combines the adjusted task allocation priorities from the auction negotiation sub-mechanism with the optimized information interaction rules from the contract network negotiation sub-mechanism. After integration, an optimized collaboration strategy is generated to resolve conflicts. This strategy fully considers the processing capabilities, resource consumption, and information interaction needs of the agents, making collaboration between agents more efficient and stable.

[0095] Step 140: Execute dynamic scheduling operations for display panel production according to the optimized collaboration strategy, and generate a set of production scheduling instructions containing process execution sequences and equipment allocation schemes.

[0096] The display panel production scheduling system executes dynamic scheduling operations for display panel production based on optimized collaboration strategies. It combines information exchange rules and task allocation priorities from the optimized collaboration strategies to schedule the production process in real time. It determines the execution order and time schedule of each process, forming a process execution sequence. Simultaneously, based on equipment availability and task requirements, it allocates equipment to appropriate processes, forming an equipment allocation scheme. The process execution sequence and equipment allocation scheme are then organized and integrated to generate a production scheduling instruction set. This set contains detailed production scheduling information, guiding the actual production operations of the display panel production system.

[0097] In one exemplary embodiment, step 140 includes:

[0098] Step 141: Extract the information interaction rules from the optimized collaboration strategy, and obtain the real-time updated status data of each agent based on the information interaction rules. The real-time updated status data includes the current operating status of the equipment and the real-time length of the process queue.

[0099] The display panel production scheduling system extracts and optimizes information interaction rules from the collaboration strategy. Based on these rules, it interacts with various intelligent agents to obtain their real-time updated status data. For the current operating status of equipment, it understands whether the equipment is in normal operation, fault repair, or standby mode. For the real-time length of the process queue, it counts the number of tasks waiting to be processed before each process. This real-time updated status data reflects the actual status of the intelligent agents at the current moment, providing the latest information for subsequent scheduling decisions.

[0100] Step 142: Extract the task allocation priority description from the optimized collaboration strategy. Based on the emergency order priority allocation rule and the low-load intelligent agent priority undertaking rule in the task allocation priority description, perform preliminary allocation processing on the production tasks to be scheduled and generate a pre-allocation list containing candidate intelligent agents and candidate devices.

[0101] The display panel production scheduling system extracts the task allocation priority description from the optimized collaboration strategy. Based on the rules prioritizing urgent orders and low-load agents, initial allocation of production tasks is performed. For urgent orders, suitable candidate agents and devices are selected for allocation. For other tasks, low-load agents and available devices are prioritized. These initial allocation results are then compiled to generate a pre-allocation list containing candidate agents and devices.

[0102] Step 143: Combine the current operating status of the devices in the real-time updated status data, exclude devices that are in a fault or maintenance state, and update the candidate device set in the pre-assigned list.

[0103] The display panel production scheduling system updates the candidate device set in the pre-allocation list by combining the current operating status of the devices with real-time status data. It checks the current status of each candidate device; if a device is faulty or under maintenance, it is removed from the candidate set. This ensures that subsequently allocated devices are available, improving production reliability.

[0104] Step 144: Based on the real-time length of the process queue in the real-time updated status data, adjust the priority of the candidate agents in the pre-allocation list and select agents whose queue length is less than the average queue length.

[0105] The display panel production scheduling system adjusts the priorities of candidate agents in the pre-allocation list by combining the real-time length of the process queue in the updated status data. First, the average real-time length of the process queue for all agents is calculated as the average queue length. Then, the real-time length of the process queue for each candidate agent is compared with the average queue length. Agents with queue lengths shorter than the average queue length are selected and their priorities are increased. These agents currently have lighter loads and more processing capacity to take on new tasks.

[0106] Optionally, step 144 includes:

[0107] Step 1441: Calculate the average real-time length of the process queue for all agents as the average queue length.

[0108] The display panel production scheduling system tracks the real-time queue lengths of all agents. These length values ​​are summed and then divided by the number of agents to obtain the average queue length. The average queue length reflects the average busy level of the process queues throughout the entire production system.

[0109] Step 1442: Extract the real-time queue length of each candidate agent in the pre-allocated list, and mark agents whose queue length is less than the average queue length as first priority agents, and agents whose queue length is greater than or equal to the average queue length as second priority agents.

[0110] The display panel production scheduling system extracts the real-time queue length of each candidate agent from the pre-allocation list. These lengths are then compared to the average queue length. Agents with queue lengths shorter than the average queue length are marked as first-priority agents; these agents currently have lighter loads and are given higher priority to accept new tasks. Agents with queue lengths greater than or equal to the average queue length are marked as second-priority agents; their loads are relatively heavier and they have lower priority.

[0111] Step 1443: Perform processing capability verification on the first priority agent, extract the number of processes that the first priority agent can process per unit time and the current remaining processing time, and calculate the number of additional tasks that the first priority agent can undertake.

[0112] The display panel production scheduling system verifies the processing capacity of the first-priority agents. It extracts the number of processes they can process per unit time and their current remaining processing time. Based on these parameters, it calculates the additional tasks that the first-priority agents can undertake. This number reflects the number of tasks the first-priority agents can handle without exceeding their processing capacity.

[0113] Step 1444: Perform a task backlog risk assessment on the second priority agent. Based on the queue length and processing capacity per unit time of the second priority agent, predict the task completion time. If the predicted completion time exceeds the order delivery time, mark it as an unavailable agent.

[0114] The display panel production scheduling system performs a task backlog risk assessment on second-priority agents. Based on the queue length and processing capacity per unit time of each agent, it predicts the time required for them to complete their current task. This predicted completion time is compared to the order delivery time. If the predicted completion time exceeds the order delivery time, it indicates that the agent may not be able to complete the task on time, posing a task backlog risk. Such agents are marked as unavailable and will not be considered in subsequent task allocation.

[0115] Step 1445: Select agents in the first priority agents that can undertake an additional number of tasks greater than zero as priority candidates, and sort them from high to low according to the number of additional tasks they can undertake.

[0116] The display panel production scheduling system filters out agents from the first-priority agents whose additional task capacity is greater than zero, designating them as priority candidates. These priority candidates are then sorted from highest to lowest based on their additional task capacity. The sorted list clearly defines the task capacity of each priority candidate agent, facilitating the rational allocation of tasks.

[0117] Step 1446: Select agents in the second priority agents that are not marked as unavailable as candidates, and sort them from low to high according to the difference between the predicted completion time and the order delivery time.

[0118] The display panel production scheduling system uses agents from the second-priority agents that are not marked as unavailable as backup candidates. These agents, although heavily loaded, still have some processing capacity. These backup candidates are then ranked from lowest to highest based on the difference between their predicted completion time and order delivery time. The smaller the difference, the more likely the agent is to complete the task on time. This ranking can be used to rationally select backup candidates when there are insufficient priority backup agents.

[0119] Step 1447: Generate an adjusted candidate agent priority list that includes the priority candidate ranking and the alternative candidate ranking.

[0120] The display panel production scheduling system organizes the priority candidate ranking and alternative candidate ranking to generate an adjusted priority list of candidate agents. This list can comprehensively represent the priority and undertaking capacity of each agent, providing an accurate reference for task allocation.

[0121] Step 145: Based on the adjusted candidate agent and the adjusted candidate device set, sort the production tasks according to the time dimension according to the order delivery time requirements, and generate a process execution time axis that includes the task start time and the task end time.

[0122] The display panel production scheduling system prioritizes production tasks based on an adjusted set of candidate agents and candidate equipment, prioritizing them according to their delivery time. It considers the delivery time requirements of each order, arranging tasks in chronological order. The system determines the start and end times of each task, creating a process execution timeline. This timeline visually displays the time arrangement of each task, facilitating efficient equipment allocation and agent coordination.

[0123] Step 146: Perform equipment allocation and matching processing based on the process execution time axis and the equipment time distribution matrix to obtain the equipment allocation and matching result.

[0124] The display panel production scheduling system performs equipment allocation matching based on the process execution timeline and the equipment time distribution matrix. For each task in the process execution timeline, the system checks the equipment time distribution matrix to determine which equipment is available during the task's execution time period. For each task time interval, the system counts the number of matching devices. If the number of matching devices is greater than zero, the device with the largest coverage between the idle time period and the task time interval is selected as the allocation device. If the number of matching devices is zero, the system extracts the end time of the busy period from the equipment's time distribution matrix, calculates the time difference between the start time of the task time interval and the end time of the busy period as the waiting time, selects the device with the smallest waiting time as the delayed allocation device, and adjusts the start time of that task in the process execution timeline to match the end time of the busy period. These matching results are then processed to obtain the equipment allocation matching result.

[0125] In one exemplary embodiment, step 146 includes:

[0126] Step 1461: Extract the start and end times of each task in the process execution time axis to generate a task time interval set; extract the idle time periods of each device in the device time distribution matrix to generate a device idle time interval set; perform intersection calculation on the task time interval set and the device idle time interval set to identify task-device matching pairs with overlapping times.

[0127] The display panel production scheduling system extracts the start and end times of each task in the process execution timeline and combines these times into a set of task time intervals. Simultaneously, it extracts the idle periods for each device from the device's time distribution matrix, forming a set of device idle time intervals.

[0128] The intersection of these two sets is calculated. This involves comparing the time interval of each task with the idle time interval of each device, identifying overlapping portions, and forming task-device matching pairs.

[0129] Step 1462: For each task time interval, count the number of matching devices. If the number of matching devices is greater than zero, select the device with the largest coverage between the idle time period and the task time interval as the allocation device. If the number of matching devices is zero, extract the end time of the busy period in the time distribution matrix of the device, and calculate the time difference between the start time of the task time interval and the end time of the busy period as the waiting time. Select the device with the smallest waiting time as the delayed allocation device, and adjust the start time of the task to the end time of the busy period in the process execution time axis. Generate the device allocation matching result containing directly matching devices and delayed allocation devices.

[0130] The display panel production scheduling system counts the number of matching devices for each task time interval. If the number of matching devices is greater than zero, it calculates the coverage of each device's idle time period with the task time interval, selecting the device with the largest coverage as the assigned device to maximize the utilization of idle time. If the number of matching devices is zero, it extracts the end time of the busy period from the device's time distribution matrix. It calculates the time difference between the start time of the task time interval and the end time of the busy period; this time difference is the waiting time. The device with the smallest waiting time is selected as the delayed assignment device, and the start time of this task is adjusted to match the end time of the busy period in the process execution timeline to ensure the task is processed as quickly as possible. The information of directly matched devices and delayed assignment devices is then compiled to generate the device allocation matching results.

[0131] Step 147: Integrate the process execution timeline and the equipment allocation matching results to generate a set of production scheduling instructions that includes the process execution order, task start and end time, and corresponding equipment identifiers.

[0132] The display panel production scheduling system integrates the process execution timeline and equipment allocation matching results. It combines the execution sequence, start and end times, and corresponding equipment identifiers of each task to form a production scheduling instruction set. This set contains detailed production scheduling information, guiding the actual production operations of the display panel production system and ensuring that each task is executed at the appropriate time and on the appropriate equipment.

[0133] In yet another embodiment, step 140 further includes:

[0134] Step 148: During the scheduling process, monitor the state changes of each agent in real time. When a sudden equipment failure is detected, extract the target task identifier affected by the failed equipment. Reassign the affected tasks corresponding to the target task identifier, call the task allocation priority description in the optimized collaboration strategy, and select the remaining available agents and devices as replacement recipients. Calculate the difference between the task processing time of the replacement recipient and the original planned processing time as the scheduling adjustment delay. If the scheduling adjustment delay exceeds the preset delay, trigger the order priority re-evaluation process.

[0135] The display panel production scheduling system monitors the status changes of each agent in real time during the scheduling process. It continuously collects agent status information through sensors and monitoring equipment. When a sudden equipment failure is detected, it quickly extracts the identifiers of the target tasks affected by the failure, clearly identifying which tasks are impacted. Affected tasks are then reassigned. The system invokes the task allocation priority description in the optimized collaboration strategy, selecting suitable replacement agents from the remaining available agents and devices. The difference between the processing time of the replacement agent and the original planned processing time is calculated to obtain the scheduling adjustment delay. If the scheduling adjustment delay exceeds a preset delay, it indicates a significant impact of the failure on production progress, triggering an order priority reassessment process to readjust order priorities and ensure timely completion of critical orders.

[0136] Step 149: When an order requirement change is detected, extract the product type and new delivery time requirement of the changed order; insert the changed order into the specified position of the process execution sequence, and adjust the execution order of related tasks according to the emergency order priority allocation rule in the optimized collaboration strategy; perform time conflict detection processing on the adjusted process execution sequence, and if there is time overlap, call the negotiation mechanism in the optimized collaboration strategy for secondary conflict resolution processing; generate a set of production scheduling instructions containing faulty equipment replacement allocation information and order change adjustment information.

[0137] When the display panel production scheduling system detects a change in order requirements, it extracts the product type and new delivery time requirement of the changed order, clarifying the specific details of the order change. The changed order is then inserted into a designated position in the process execution sequence. This designated position is determined based on the new delivery time requirement and product type. Next, according to the urgent order priority allocation rule in the optimized collaboration strategy, the execution order of relevant tasks is adjusted to ensure the changed order is processed promptly. Time conflict detection and handling are performed on the adjusted process execution sequence. It checks for any overlapping tasks; if time overlap exists, it indicates a potential conflict. When a time conflict is detected, the negotiation mechanism in the optimized collaboration strategy is invoked for secondary conflict resolution. Relevant agents negotiate to find the best solution to the conflict. Finally, a set of production scheduling instructions is generated, containing information on faulty equipment replacement allocation and order change adjustment information. This set reflects the latest production scheduling status, ensuring the smooth operation of the production process.

[0138] In a non-limiting embodiment, after step 140, the method further includes: collecting real-time execution feedback data of each agent during the execution of the production scheduling instruction set; performing deviation feature extraction processing on the real-time execution feedback data to generate a deviation feature vector containing process time deviation rate, load fluctuation amplitude, and material consumption deviation coefficient; inputting the deviation feature vector into a pre-trained adaptive optimization model, adjusting the time parameters of the process execution sequence and the load parameters of the equipment allocation scheme through the adaptive optimization model to generate optimized adjustment parameters containing time parameter correction values ​​and load parameter correction values; iteratively updating the process execution sequence and equipment allocation scheme in the production scheduling instruction set based on the optimized adjustment parameters to generate an iteratively optimized scheduling instruction set after deviation correction.

[0139] The display panel production scheduling system collects real-time execution feedback data from each agent during the execution of the production scheduling instruction set. Through sensors and data acquisition devices, it gathers the actual execution status of each agent, such as the actual execution time of processes, equipment load, and material consumption. Deviation features are extracted from the real-time execution feedback data to calculate the process time deviation rate (the ratio of actual execution time to planned execution time), load fluctuation amplitude (reflecting changes in equipment load over different time periods), and material consumption deviation coefficient (measuring the degree of deviation between actual and planned material consumption). These deviation features are combined into a deviation feature vector, which is then input into a pre-trained adaptive optimization model. This model is a neural network model trained on a large amount of data, capable of adjusting the time parameters of the process execution sequence and the load parameters of the equipment allocation scheme based on the input deviation feature vector. The adaptive optimization model outputs optimized adjustment parameters containing time and load parameter correction values. Based on these optimization adjustment parameters, the process execution sequence and equipment allocation scheme in the production scheduling instruction set are iteratively updated. Through continuous iterative updates, deviations are gradually eliminated, generating an iteratively optimized scheduling instruction set with deviation correction.

[0140] In a non-limiting embodiment, after step 140, the method further includes: collecting real-time data of multi-dimensional optimization indicators of the display panel production system; performing feature fusion processing on the real-time data of the multi-dimensional optimization indicators to generate a multi-objective feature matrix including an energy efficiency-loss-cooperative correlation matrix; inputting the multi-objective feature matrix into a pre-configured multi-objective cooperative optimization model, and using the multi-objective cooperative optimization model to perform multi-objective trade-off adjustments on the process execution sequence priority and equipment allocation scheme resource configuration parameters in the production scheduling instruction set, generating multi-objective optimization parameters including priority adjustment coefficients and resource configuration correction parameters; and coordinating the process execution sequence and equipment allocation scheme in the production scheduling instruction set based on the multi-objective optimization parameters to generate a multi-objective balanced cooperative optimization scheduling instruction set.

[0141] The display panel production scheduling system collects real-time data on multi-dimensional optimization indicators of the display panel production system. These indicators include data on energy efficiency, losses, and collaboration. For example, energy efficiency reflects the energy utilization efficiency of equipment, losses represent material and equipment losses during the production process, and collaboration measures the efficiency of cooperation between agents. Feature fusion processing is performed on the real-time data of multi-dimensional optimization indicators. The system integrates and analyzes the indicator data from different dimensions to generate a multi-objective feature matrix containing an energy efficiency-loss-collaboration correlation matrix. This matrix reflects the interrelationships between different optimization indicators. The multi-objective feature matrix is ​​input into a pre-configured multi-objective collaborative optimization model. This model is a neural network model specifically designed to handle multi-objective optimization problems, capable of multi-objective trade-off adjustments to the priority of process execution sequences and resource allocation parameters of equipment allocation schemes in the production scheduling instruction set. The multi-objective collaborative optimization model comprehensively considers multiple objectives such as energy efficiency, losses, and collaboration to find an optimal solution. The output includes multi-objective optimization parameters containing priority adjustment coefficients and resource allocation correction parameters. Based on the multi-objective optimization parameters, the process execution sequences and equipment allocation schemes in the production scheduling instruction set are updated collaboratively. Through collaborative updates, a balance among multiple objectives is achieved, generating a collaboratively optimized scheduling instruction set after the balance of multiple objectives.

[0142] In a non-limiting embodiment, step 140 is followed by: collecting real-time data of disturbance events from each agent during the production process; performing feature quantization on the real-time data of the disturbance events to generate a disturbance feature vector containing an event type identifier, an impact range vector, and a duration parameter; inputting the disturbance feature vector into a pre-trained disturbance impact assessment model, and calculating the impact assessment result of each disturbance event on the production scheduling instruction set through the disturbance impact assessment model, wherein the impact assessment result includes the affected scheduling instruction identifier, the impact severity coefficient, and the deviation from the production target; calling a preset disturbance response optimization model based on the impact assessment result to generate an adaptive adjustment strategy including a process execution sequence adjustment strategy and an equipment reallocation strategy; and modifying the process execution sequence and equipment allocation scheme in the production scheduling instruction set according to the adaptive adjustment strategy to generate an optimized scheduling instruction set adapted to the disturbance.

[0143] The display panel production scheduling system collects real-time data on disturbance events from various agents during the production process. Disturbance events may include equipment failures, raw material supply interruptions, and order changes. Various monitoring methods are used to collect relevant information about these events in real time. The real-time disturbance event data undergoes feature quantification to determine the event type, such as equipment failure events or order change events; an impact range vector is calculated to reflect which agents and processes are affected; and the duration parameter of the event is calculated, i.e., the length of time from occurrence to end. These quantified features are combined into a disturbance feature vector, which is then input into a pre-trained disturbance impact assessment model. This model is a neural network model trained on a large amount of historical data, capable of calculating the impact assessment results of each disturbance event on the production scheduling instruction set based on the input disturbance feature vector. The impact assessment results include the affected scheduling instruction identifier, clarifying which scheduling instructions are affected by the disturbance event; an impact severity coefficient, measuring the degree of impact of the event on the production process; and a production target deviation, reflecting the degree to which the event causes deviation from the production target. Based on the impact assessment results, a pre-defined disturbance response optimization model is invoked. This model generates an adaptive adjustment strategy, including a process execution sequence adjustment strategy and an equipment reallocation strategy, according to the impact assessment results. Based on the adaptive adjustment strategy, the process execution sequence and equipment allocation scheme in the production scheduling instruction set are modified. Through modification, the production scheduling can adapt to the impact of the disturbance event, generating an optimized scheduling instruction set after disturbance adaptation.

[0144] In practical applications, the training and application of neural network models can be achieved based on existing industrial IoT data acquisition frameworks and open-source deep learning libraries (such as TensorFlow or PyTorch).

[0145] Specifically, for the conflict classification model, a multilayer perceptron structure can be constructed using supervised learning methods. The model can be trained using a conflict sample dataset from the production history (including features such as the frequency of state data lag and the severity of equipment competition). The weight parameters can be optimized through the backpropagation algorithm so that the model can accurately output the conflict type identifier and the scope of its impact.

[0146] For adaptive optimization models, reinforcement learning mechanisms can be used to process real-time execution feedback data. The process time deviation rate and load fluctuation amplitude can be used as state inputs, scheduling efficiency can be defined as the reward function, and the agent can be trained to generate time parameter correction values ​​to optimize the process sequence.

[0147] To address the issue of inconsistent dimensions, we can refer to industry standardization protocols (such as OPC UA) to clearly define time parameters as minutes (such as average waiting time and task processing time), and introduce dimensionless processing (such as the ratio of normalized overlap allocation frequency to the maximum processing capacity of the equipment) when calculating parameters of the severity of equipment contention conflicts to ensure data comparability.

[0148] For multi-objective collaborative optimization models, the input features of the energy efficiency-loss-collaboration correlation matrix can be integrated based on evolutionary algorithms (such as NSGA-II). Resource allocation correction parameters can be generated through multi-objective Pareto optimal search, so that the output simultaneously satisfies energy consumption minimization and production collaboration maximization.

[0149] For disturbance impact assessment models, time-series prediction networks (such as LSTM) can be used to analyze the correlation between the duration of disturbance events and the impact range vector. During training, historical fault datasets are used to simulate deviations from production targets, and quantitative impact severity coefficients are output for adaptive adjustment. In addition, the model training process can use cross-validation techniques to divide the dataset, employ early stopping strategies to prevent overfitting, and deploy on edge computing nodes to achieve real-time inference, ultimately forming a complete and reproducible intelligent scheduling closed loop.

[0150] In this embodiment of the invention, the inputs and outputs of the relevant neural network model constitute a hierarchical decision chain. The base layer input originates from real-time production data: the conflict classification model receives raw conflict data generated by state collaborative updates (such as lagging data due to mismatched interaction frequencies, and overlapping allocation signals for urgent orders), extracts nonlinear correlation features between conflicts through a hidden layer, and outputs high-level semantic conflict analysis results (including type identification and impact range). This result directly serves the task adjustment of the auction negotiation sub-mechanism. In the advanced layer, the adaptive optimization model takes the deviation feature vector from real-time execution feedback as input (such as process time deviation rate and material consumption deviation coefficient), quantifies the degree of production deviation through a feature fusion layer, and outputs time parameter correction values ​​and load parameter correction values ​​to drive iterative optimization of the process sequence. In the objective optimization layer, the multi-objective collaborative optimization model receives a correlation matrix (energy efficiency-loss-collaboration data) fused from multi-dimensional indicators, performs multi-objective trade-off calculations at the output layer, and generates priority adjustment coefficients to reconfigure equipment resources. In the disturbance response layer, the disturbance impact assessment model inputs the disturbance feature vector (event type identifier, duration) into the time series analysis module and outputs the impact assessment results (including the deviation of production target), thereby triggering the disturbance response optimization model to generate a process sequence adjustment strategy.

[0151] It can be understood that all the above models form a closed loop: low-level inputs (state data, execution deviations) are transformed into optimization parameters (time correction values, priority coefficients) through feature abstraction. These outputs are then input as high-level instructions to the scheduling layer (such as adjusting the process execution time axis), ultimately achieving hierarchical decision-making collaboration from data acquisition to strategy generation. This ensures that the entire logic strictly depends on the production context characteristics of the input data (such as equipment idle periods, order delivery times), and ensures that the output decisions maintain a causal relationship with production goals (load balancing, urgent order delivery).

[0152] In summary, the embodiments of the present invention significantly improve the efficiency and quality of display panel production scheduling, and realize the intelligent, dynamic and efficient production process.

[0153] First, real-time status collaborative update operations are performed on multiple intelligent agents in the display panel production system, which comprehensively and timely grasps the production-related status of each intelligent agent. This provides an accurate and real-time data foundation for subsequent scheduling decisions, enabling the scheduling system to formulate strategies based on accurate information and avoiding production chaos caused by information lag or inaccuracy.

[0154] Secondly, the set of collaborative strategies among agents, generated based on the set of cooperative states, clarifies the rules for information interaction and the priority of task allocation among agents, effectively coordinating the work among them. Reasonable information interaction rules ensure that agents can share key information in a timely and accurate manner, improving information flow efficiency; scientific task allocation priorities ensure that tasks are reasonably assigned to appropriate agents, avoiding unreasonable task accumulation and resource waste.

[0155] Furthermore, by invoking a pre-configured negotiation mechanism to detect and resolve conflicts in the set of collaboration strategies, potential conflicts that may arise during agent collaboration are identified and resolved in a timely manner. This not only avoids conflicts hindering production progress but also improves the collaboration efficiency between agents and the overall stability of production by optimizing collaboration strategies.

[0156] Finally, based on the optimized collaboration strategy, dynamic scheduling operations for display panel production are executed, generating a set of production scheduling instructions containing process execution sequences and equipment allocation schemes. This instruction set can be dynamically adjusted according to the actual production situation, ensuring the reasonable execution sequence of processes and the effective allocation of equipment. Dynamic scheduling enables the production process to respond quickly to various changes, such as equipment failures and order changes, improving production flexibility and adaptability. Ultimately, optimal scheduling for display panel production is achieved, improving production efficiency and product quality while reducing production costs.

[0157] See Figure 2 As shown in the figure, this is a schematic diagram of the basic structure of a display panel production scheduling system 200 provided in an embodiment of the present invention. The display panel production scheduling system 200 includes:

[0158] Processor 201;

[0159] Storage device 202, on which computer program 2020 is stored;

[0160] When the computer program 2020 is executed by the processor 201, the processor 201 implements any of the aforementioned agent-based display panel production scheduling methods.

[0161] Based on the above, a readable storage medium is provided, on which a program or instructions are stored, and when the program or instructions are executed by a processor, the steps of the above method are implemented.

[0162] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

Claims

1. A method for scheduling the production of display panels based on agent collaboration, characterized in that, include: A real-time state collaborative update operation is performed on multiple intelligent agents in the display panel production system to obtain a collaborative state set for each intelligent agent. The collaborative state set includes intelligent agent identification information and corresponding production-related state descriptions. Based on the set of cooperative states, a set of cooperative strategies among agents is generated. The set of cooperative strategies includes information interaction rules and task allocation priority descriptions among agents. The pre-configured negotiation mechanism is invoked to perform conflict detection and negotiation on the set of cooperation strategies, and an optimized cooperation strategy after conflict resolution is generated. Based on the optimized collaboration strategy, dynamic scheduling operations for display panel production are executed to generate a set of production scheduling instructions that includes process execution sequences and equipment allocation schemes. The invocation of the pre-configured negotiation mechanism to perform conflict detection and negotiation processing on the set of cooperation strategies, generating an optimized cooperation strategy after conflict resolution, includes: The information interaction rules in the collaboration strategy set are processed to identify conflict types, and conflicts caused by mismatched interaction frequencies and information comprehension deviations caused by inconsistent shared data types are extracted. The task allocation priority description in the collaboration strategy set is processed to identify conflict scenarios, and to identify equipment competition conflicts caused by overlapping allocation of urgent orders and task backlog conflicts caused by insufficient capacity of low-load intelligent agents. The status data lag conflict, the information comprehension bias conflict, the device competition conflict, and the task backlog conflict are input into a pre-configured conflict classification model to generate conflict analysis results that include conflict type identifiers and scope of impact. Based on the conflict analysis results, the auction negotiation sub-mechanism is invoked, and the agent involved in the conflict analysis results submits bidding parameters including equipment availability time, process efficiency and resource consumption cost. The task allocation priority is adjusted according to the comprehensive evaluation results of the bidding parameters. The contract network negotiation sub-mechanism is invoked to negotiate and modify the information interaction rules. The leading agent publishes the interaction rules to adjust the bidding information, and the non-leading agents respond and provide feedback on constraints including data processing capabilities and communication bandwidth limitations. The information interaction rules are optimized based on the compatibility analysis results of the constraints. By integrating the processing results of the auction negotiation sub-mechanism and the contract network negotiation sub-mechanism, an optimized collaboration strategy is generated after conflict resolution.

2. The method according to claim 1, characterized in that, The process of generating a set of cooperation strategies among agents based on the set of cooperative states, wherein the set of cooperation strategies includes information interaction rules and task allocation priority descriptions among agents, including: Analyze the production-related state descriptions of each agent in the collaborative state set, and extract a subset of state features including equipment operating status, order completion progress, and process waiting queue length; Availability assessment processing is performed on the device operating status in the aforementioned state feature subset to generate a time distribution matrix of idle and busy periods. The order completion progress in the subset of state features is prioritized and sorted, and an order priority sequence is generated based on the order delivery time requirements and product type complexity. Load balancing analysis is performed on the process waiting queue length in the state feature subset to calculate the average waiting time and queue backlog rate for different agents corresponding to the processes. Information interaction rules are constructed based on the time distribution matrix, the order priority sequence, and the average waiting time. The information interaction rules include real-time shared state data types and interaction frequency parameters among intelligent agents. A task allocation priority description is generated by combining the queue backlog rate and the order priority sequence. The task allocation priority description includes rules for prioritizing urgent orders and rules for prioritizing low-load agents.

3. The method according to claim 1, characterized in that, The process of performing conflict scenario localization processing on the task allocation priority descriptions in the collaboration strategy set, identifying equipment competition conflicts caused by overlapping allocation of urgent orders and task backlog conflicts caused by insufficient handling capacity of low-load intelligent agents, includes: Extract the target agent allocation list of emergency orders from the task allocation priority description, and count the number of times the same device is simultaneously allocated by multiple emergency orders as the device overlap allocation frequency. The severity parameter of equipment competition conflict is calculated based on the frequency of equipment overlap allocation and the maximum number of orders processed by a single equipment. Extract the process processing capability parameters of the low-load agent from the task allocation priority description. The process processing capability parameters include the number of processes that can be processed per unit time and the maximum continuous working time. The number of tasks assigned to the low-load intelligent agent and the total processing time of the tasks are counted, and the difference between the total processing time of the tasks and the maximum continuous working time is calculated as the remaining capacity. When the remaining capacity is less than the preset capacity value, it is determined that there is a task backlog conflict caused by insufficient capacity of the low-load agent, and the agent identifier and task identifier involved in the task backlog conflict are recorded. By combining the severity parameters of the device competition conflict and the scope of the task backlog conflict, a conflict scene location result including conflict location coordinates and impact degree is generated.

4. The method according to claim 1, characterized in that, The contract network negotiation sub-mechanism is invoked to negotiate and modify the information interaction rules. The leading agent publishes adjustments to the bidding information, and non-leading agents respond and provide feedback on constraints including data processing capabilities and communication bandwidth limitations. Based on the compatibility analysis results of these constraints, the information interaction rules are optimized, including: The agent responsible for global coordination in the current production stage is selected as the leading agent, and the leading agent generates bidding information for adjusting interaction rules, including the requirements for adjusting the data type and frequency of interaction. The bidding information for adjusting the interaction rules is broadcast to the non-dominant intelligent agent, which then generates constraint feedback information based on its own data processing capabilities and communication bandwidth limitations. The constraint feedback information is subjected to compatibility analysis and processing to calculate the matching degree between the interaction data type adjustment requirements and the data processing capabilities of each agent, as well as the adaptability between the interaction frequency adjustment requirements and the communication bandwidth limitations of each agent. Agents whose matching degree and adaptability both exceed the preset adaptability are selected as rule-accepting agents, and the average data processing capability and average communication bandwidth limit of the rule-accepting agents are statistically analyzed. The interaction data types are adjusted based on the average data processing capability, retaining data types that can be processed by each agent and eliminating data types that do not match the data processing capability. The interaction frequency is adjusted based on the average of the communication bandwidth limit, and the interaction frequency is set to the maximum supportable frequency that does not exceed the communication bandwidth limit of each agent. Generate optimized information interaction rules that include the adjusted data types and frequencies of interaction.

5. The method according to claim 1, characterized in that, The dynamic scheduling operation for display panel production, executed according to the optimized collaboration strategy, generates a set of production scheduling instructions containing process execution sequences and equipment allocation schemes, including: Extract the information interaction rules from the optimized collaboration strategy, and obtain the real-time updated status data of each intelligent agent based on the information interaction rules. The real-time updated status data includes the current operating status of the equipment and the real-time length of the process queue. Extract the task allocation priority description from the optimized collaboration strategy, and perform preliminary allocation processing on the current production tasks to be scheduled according to the emergency order priority allocation rule and the low-load intelligent agent priority undertaking rule in the task allocation priority description, and generate a pre-allocation list containing candidate intelligent agents and candidate devices. Based on the current operating status of the devices in the real-time updated status data, devices that are in a fault or maintenance state are excluded, and the candidate device set in the pre-assigned list is updated. Based on the real-time length of the process queue in the real-time updated status data, the priority of the candidate agents in the pre-allocation list is adjusted, and agents with queue lengths less than the average queue length are selected. Based on the adjusted candidate agent and candidate device set, the production tasks are sorted according to the order delivery time requirements to generate a process execution timeline that includes the task start time and task end time. Based on the process execution time axis and the equipment time distribution matrix, equipment allocation and matching processing is performed to obtain the equipment allocation and matching result; By integrating the process execution timeline and the equipment allocation matching results, a set of production scheduling instructions is generated, which includes the process execution order, task start and end time, and corresponding equipment identifiers.

6. The method according to claim 5, characterized in that, The step of adjusting the priority of candidate agents in the pre-allocation list by combining the real-time length of the process queue in the real-time updated status data, and selecting agents with queue lengths less than the average queue length, includes: Calculate the average real-time queue length of all agents as the average queue length; Extract the real-time length of the process queue for each candidate agent in the pre-allocated list, and mark agents whose queue length is less than the average queue length as first priority agents, and agents whose queue length is greater than or equal to the average queue length as second priority agents; The processing capability of the first priority agent is reviewed, the number of processes that the first priority agent can process per unit time and the current remaining processing time are extracted, and the number of additional tasks that the first priority agent can undertake is calculated. The second priority agent is subjected to a task backlog risk assessment. The task completion time is predicted based on the queue length and unit time processing capacity of the second priority agent. If the predicted completion time exceeds the order delivery time, the agent is marked as an unavailable agent. Among the first priority agents, those agents that can undertake an additional number of tasks greater than zero are selected as priority candidates and sorted from high to low according to the number of additional tasks they can undertake. Agents not marked as unavailable in the second priority agents are selected as candidates and sorted from low to high according to the difference between the predicted completion time and the order delivery time. Generate an adjusted candidate agent priority list that includes a priority ranking of first candidates and a ranking of alternative candidates.

7. The method according to claim 5, characterized in that, The step of performing equipment allocation and matching processing based on the process execution time axis and the equipment time distribution matrix to obtain equipment allocation and matching results includes: Extract the start and end times of each task in the process execution timeline to generate a set of task time intervals; Extract the idle time periods of each device from the time distribution matrix of the devices, and generate a set of device idle time intervals; The intersection of the task time interval set and the device idle time interval set is calculated to identify task-device matching pairs with overlapping times. For each task time interval, count the number of matching devices. If the number of matching devices is greater than zero, select the device with the largest coverage between the idle time period and the task time interval as the assigned device. If the number of matching devices is zero, extract the end time of the busy period from the time distribution matrix of the device, and calculate the time difference between the start time of the task time interval and the end time of the busy period as the waiting time. Select the device with the shortest waiting time as the delay allocation device, and adjust the start time of the task in the process execution time axis to the end time of the busy period; Generate device allocation matching results that include directly matched devices and delayed allocation devices.

8. The method according to claim 1, characterized in that, The step of executing dynamic scheduling operations for display panel production according to the optimized collaboration strategy, generating a set of production scheduling instructions including process execution sequences and equipment allocation schemes, further includes: During the scheduling process, the state changes of each intelligent agent are monitored in real time. When a sudden equipment failure is detected, the target task identifier affected by the failure equipment is extracted. The affected tasks corresponding to the target task identifier are reallocated. The task allocation priority description in the optimized collaboration strategy is invoked, and the remaining available agents and devices are selected as alternative recipients. The difference between the task processing time of the alternative receiving object and the original planned processing time is calculated as the scheduling adjustment delay amount. If the scheduling adjustment delay amount exceeds the preset delay amount, the order priority re-evaluation process is triggered. When a change in order requirements is detected, extract the product type and new delivery time requirements of the changed order; Insert the change order into the specified position of the process execution sequence, and adjust the execution order of related tasks according to the emergency order priority allocation rule in the optimized collaboration strategy; The adjusted process execution sequence is subjected to time conflict detection and processing. If there is time overlap, the negotiation mechanism in the optimized collaboration strategy is invoked for secondary conflict resolution. Generate a set of production scheduling instructions that include information on replacement allocation of faulty equipment and information on order changes and adjustments.

9. A display panel production scheduling system, characterized in that, include: processor; A storage device having a computer program stored thereon, which, when executed by the processor, causes the processor to implement the display panel production scheduling method based on intelligent agent cooperation as described in any one of claims 1-8.

Images