Chip processing visual management system and method based on artificial intelligence

Abstract

The invention discloses a chip processing visual management system and method based on artificial intelligence, and relates to the related technical field of chip processing management, and the system comprises a visual path construction module which is used for collecting a plurality of visual projects of a chip processing workshop and constructing a plurality of visual paths; the dynamic response relation analysis module is used for constructing a visual update trigger threshold constraint of the source data; the risk accompanying analysis module is used for executing visual updating processing and carrying out risk accompanying analysis on the basis of the source data set with the concurrent updating requirement; and the configuration updating processing module is used for performing updating processing after priority configuration based on the risk accompanying index set. The technical problems that in the prior art, the dynamic response capability of visual management is insufficient, and the priority scheduling efficiency is low in a high-concurrency updating scene, so that the management efficiency and reliability are poor are solved, and the technical effects that intelligent dynamic updating is achieved, risk-controllable efficient visual scheduling is achieved, and the visual management efficiency and reliability are improved are achieved.

Description

Technical Field

[0001] This application relates to the technical field of chip processing management, specifically to an artificial intelligence-based chip processing visualization management system and method. Background Technology

[0002] With the rapid development of semiconductor manufacturing, the production processes in chip processing workshops are becoming increasingly complex, involving the collaborative management of a large number of equipment, process parameters, and real-time data. Traditional chip workshop management relies on manual monitoring and static data display, which is difficult to adapt to the demands of a highly dynamic and high-concurrency production environment. Achieving efficient and accurate visual management has become crucial for improving chip processing efficiency and yield. Current visual management systems typically use fixed thresholds or periodic updates, lacking the ability to intelligently respond to dynamic data changes. For example, when multiple key process parameters fluctuate simultaneously, they may not be able to effectively distinguish priorities, leading to delays in key information updates or visual interface lag, thus affecting the real-time nature of production decisions. Furthermore, because chip processing involves multi-source heterogeneous data (such as equipment status, environmental parameters, and process indicators), the data dependencies between different visualization items are complex. Existing methods struggle to meet the demands of high-concurrency updates, potentially causing data inconsistencies or visualization conflicts, thereby impacting the management efficiency and reliability of chip processing workshops.

[0003] Therefore, current technologies suffer from insufficient dynamic response capabilities for visual management and low priority scheduling efficiency in high-concurrency update scenarios, resulting in poor management efficiency and reliability. Summary of the Invention

[0004] This application provides an AI-based chip manufacturing visualization management system and method, which solves the technical problems of insufficient dynamic response capability of visualization management and low priority scheduling efficiency in high-concurrency update scenarios, resulting in poor management efficiency and reliability. It achieves intelligent dynamic updates, realizes efficient visualization scheduling with controllable risks, and improves the technical effects of visualization management efficiency and reliability.

[0005] This application provides an AI-based chip manufacturing visualization management system, comprising: a visualization path construction module for collecting multiple visualization projects from visualization terminals corresponding to the chip manufacturing workshop and constructing multiple visualization paths corresponding to the multiple visualization projects; a dynamic response relationship analysis module for performing dynamic response relationship analysis between source data and visualization projects on the multiple visualization paths and constructing visualization update trigger threshold constraints for the source data; a risk-accompanied analysis module for executing visualization update processing of the multiple visualization paths based on the visualization update trigger threshold constraints, and when the concurrent update demand exceeds a preset demand threshold, performing risk-accompanied analysis based on the source dataset of the concurrent update demand to generate a risk-accompanied indicator set corresponding to the concurrent update project set; and a configuration update processing module for performing priority configuration and update processing on the concurrent update project set based on the risk-accompanied indicator set.

[0006] In a possible implementation, the AI-based chip manufacturing visualization management system also performs the following processing: the multiple visualization items refer to preset display types displayed on the electronic screen of the visualization terminal, and any one of the multiple visualization paths includes several processing nodes executed from the source data acquisition end until the characteristic value of the visualization item is obtained.

[0007] In a possible implementation, the AI-based chip manufacturing visualization management system further performs the following processes: historical data collection for the multiple visualization paths, generating multiple historical mapping datasets of source data and visualization projects; reading multiple minimum update dynamic boundaries of multiple visualization projects input by the management-authorized user of the visualization terminal; performing dynamic response relationship analysis between source data and visualization projects based on the multiple historical mapping datasets, with visualization projects as dependent variables and source data as independent variables, to determine multiple source data change dynamic thresholds that satisfy the multiple minimum update dynamic boundaries; and adding the multiple source data change dynamic thresholds to the visualization update trigger threshold constraint.

[0008] In a possible implementation, the AI-based chip manufacturing visualization management system further performs the following processing: determining whether there are multiple source data acquisition terminals corresponding to any visualization project; if so, performing single-item response relationship analysis and multiple joint response relationship analysis between multiple source data acquisition terminals and visualization projects based on historical mapping datasets, determining the single-item source data change dynamic threshold and the multiple source data joint change dynamic threshold that satisfy the multiple minimum update dynamic boundaries, and using them as the source data change dynamic threshold corresponding to any visualization project; and constructing the multiple source data change dynamic thresholds using the source data change dynamic threshold corresponding to any visualization project.

[0009] In a possible implementation, the AI-based chip manufacturing visualization management system also performs the following processing: a single response relationship is the independent influence relationship of any one of the multiple source data acquisition terminals on the visualization project; a multiple joint response relationship is the joint influence relationship of two or more source data acquisition terminals on the visualization project.

[0010] In a possible implementation, the AI-based chip manufacturing visualization management system further performs the following processes: extracting each source dataset corresponding to each concurrent update project in the concurrent update project set; determining the project type of each concurrent update project, performing chip manufacturing risk identification contribution analysis based on the project type, and constructing risk contribution weights for each concurrent update project; comparing the differences between each source dataset and the source data of the previous update node to determine each difference risk indicator; weighting each difference risk indicator with the risk contribution weights of each concurrent update project to generate each risk-accompanying indicator for each concurrent update project; and constructing the risk-accompanying indicator set using the risk-accompanying indicators.

[0011] In a possible implementation, the AI-based chip manufacturing visualization management system further performs the following processes: retrieving historical chip manufacturing risk data by the project type; performing a detection importance analysis of manufacturing anomaly events based on the historical chip manufacturing risk data, and generating risk contribution weights for each concurrent update project.

[0012] In a possible implementation, the AI-based chip manufacturing visualization management system further performs the following processing: based on the risk-accompanying indicator set, extracting a first type of concurrent update items and a second type of concurrent update items whose risk-accompanying indicators are greater than a preset indicator threshold; configuring the update priority of the first type of concurrent update items to be higher than that of the second type of concurrent update items; and performing update processing on the first type of concurrent update items and the second type of concurrent update items under the constraint of the priority configuration result.

[0013] In a possible implementation, the AI-based chip manufacturing visualization management system also performs the following processing: the preset demand threshold is a data batch concurrent processing threshold determined based on the real-time network status of the chip manufacturing workshop.

[0014] This application also provides an AI-based chip manufacturing visualization management method, the method comprising: collecting multiple visualization projects from a visualization terminal corresponding to a chip manufacturing workshop, and constructing multiple visualization paths corresponding to the multiple visualization projects; performing dynamic response relationship analysis between source data and visualization projects on the multiple visualization paths, and constructing visualization update trigger threshold constraints for the source data; executing visualization update processing on the multiple visualization paths based on the visualization update trigger threshold constraints; when concurrent update demand exceeds a preset demand threshold, performing risk-related analysis based on the source dataset of concurrent update demand, and generating a risk-related indicator set corresponding to the concurrent update project set; and performing priority configuration and update processing on the concurrent update project set based on the risk-related indicator set.

[0015] This application proposes an AI-based chip manufacturing visualization management system and method, comprising: a visualization path construction module for collecting multiple visualization projects in a chip manufacturing workshop and constructing multiple visualization paths; a dynamic response relationship analysis module for constructing visualization update trigger threshold constraints for source data; a risk-associated analysis module for performing visualization update processing and conducting risk-associated analysis based on the source dataset with concurrent update requirements; and a configuration update processing module for priority configuration and update processing based on a set of risk-associated indicators. This addresses the technical problems of insufficient dynamic response capabilities and low priority scheduling efficiency in high-concurrency update scenarios in existing visualization management technologies, leading to poor management efficiency and reliability. It achieves intelligent dynamic updates, efficient visualization scheduling with controllable risks, and improves the efficiency and reliability of visualization management. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a schematic diagram of the structure of an AI-based chip manufacturing visualization management system provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the chip manufacturing visualization management method based on artificial intelligence provided in the embodiments of this application.

[0019] Figure labeling: Visual path construction module 10, dynamic response relationship analysis module 20, risk accompaniment analysis module 30, configuration update processing module 40. Detailed Implementation

[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0023] This application provides an AI-based chip manufacturing visualization management system, such as... Figure 1 As shown, the system includes: The visualization path construction module 10 is used to collect multiple visualization projects from the visualization terminal corresponding to the chip processing workshop and construct multiple visualization paths corresponding to the multiple visualization projects.

[0024] Furthermore, the specific configuration of the visualization path construction module 10 also includes that the multiple visualization items refer to the preset display types displayed on the electronic screen of the visualization terminal, and any one of the multiple visualization paths includes several processing nodes executed from the source data acquisition end until the characteristic value of the visualization item is obtained.

[0025] Preferably, in a chip manufacturing workshop, a visualization terminal is a device or system interface used to display various production data, equipment status, process flow, and other information. These terminals are typically distributed throughout the workshop, facilitating real-time viewing and monitoring of production by staff. A visualization project refers to a preset display type shown on the electronic screen of the visualization terminal, which may include charts, graphs, images, text, data tables, etc., used to intuitively present key information in the chip manufacturing process, such as equipment operation status monitoring projects, process parameter monitoring projects, production progress tracking projects, and quality inspection result display projects. Multiple visualization projects from the corresponding visualization terminals in the chip manufacturing workshop are collected, and the source data acquisition end corresponding to each visualization project is determined. This source data may come from various sensors, equipment control systems, production management systems, etc., within the workshop. For example, the data source for equipment operation status monitoring projects may be temperature sensors, pressure sensors, vibration sensors, etc., on the equipment, used to collect real-time data during equipment operation.

[0026] Preferably, multiple visualization paths are constructed for each visualization project to comprehensively and accurately display key production information in the chip manufacturing workshop. Each visualization path starts from the source data acquisition end, runs through multiple processing nodes, and finally obtains the feature values ​​of the visualization project. These multiple processing nodes include several processing steps from source data to visualization feature values, potentially including data preprocessing nodes, data analysis and transformation nodes, data extraction and calculation nodes, and data integration and association nodes. Specifically, the data preprocessing node cleans the raw data, removes noise, fills in missing values, and detects and handles outliers to ensure reliable data quality. For example, temperature data collected by sensors may have instantaneous abnormal peak values ​​due to equipment interference, which are smoothed using filtering algorithms; missing production count data is reasonably filled using interpolation. The data analysis and transformation node converts the preprocessed data into a format suitable for visualization. The data extraction and calculation node calculates key feature values ​​according to business needs. For example, it calculates statistical indicators (such as average, standard deviation, maximum, and minimum values) from time-series data collected by equipment sensors to display equipment performance fluctuation characteristics, and calculates overall production efficiency and bottleneck station identification indicators from the production cycle data of each workstation on the production line.

[0027] Preferably, the data integration and association node integrates and associates data from different sources. For example, it associates and integrates photoresist coating thickness data and subsequent etching depth data during chip manufacturing to analyze the impact of the matching of process parameters between the two on chip quality. Finally, according to the different visualization project requirements, the source data is transformed into project feature values ​​that can be intuitively displayed on the visualization terminal. For example, for equipment operation status monitoring projects, charts, graphs, animations, and other forms can be used to display the real-time operation status and historical trends of the equipment. Changes in visual elements such as color, shape, and position can intuitively reflect different states of the equipment, such as normal, warning, and alarm.

[0028] The dynamic response relationship analysis module 20 is used to perform dynamic response relationship analysis between source data and visualization projects on the multiple visualization paths, and to construct visualization update trigger threshold constraints for source data.

[0029] Preferably, data mining and machine learning are used to analyze the dynamic response relationship between source data and visualization projects. This involves studying the real-time correlation between changes in source data and updates to visualization projects. In a chip manufacturing workshop, source data is constantly generated and changing, and visualization projects need to be updated promptly to reflect the actual state of the workshop. For example, correlation analysis can be used to determine strong correlation indicators between source data and visualization projects, or time series analysis can be used to determine the delay time and impact of changes in source data on visualization project updates. This helps identify source data that significantly affects visualization projects, reducing unnecessary visualization updates and improving system efficiency and performance. Then, based on the results of the dynamic response relationship analysis, a source data change threshold is set for each visualization project as a visualization update triggering threshold constraint. When the source data change exceeds the set threshold, the visualization project is triggered to update. The visualization update triggering threshold constraint may include the magnitude, frequency, and duration of data changes. For example, if the equipment temperature changes by a certain magnitude (e.g., 5 degrees Celsius) or changes continuously by a certain frequency (e.g., 0.5 degrees Celsius per second) within a certain time (e.g., 1 minute), the visualization project is triggered to update. By constructing a visual update trigger threshold constraint, precise control over visual updates can be achieved, avoiding frequent updates due to slight fluctuations in source data, reducing the computational and transmission burden on the system, and thus enabling real-time monitoring and control of product quality.

[0030] Furthermore, the specific configuration of the dynamic response relationship analysis module 20 also includes: collecting historical data for the multiple visualization paths to generate multiple historical mapping datasets of source data and visualization projects; reading multiple minimum update dynamic boundaries of multiple visualization projects input by the management permission user of the visualization terminal; performing dynamic response relationship analysis between source data and visualization projects based on the multiple historical mapping datasets, with visualization projects as dependent variables and source data as independent variables, to determine multiple source data change dynamic thresholds that satisfy the multiple minimum update dynamic boundaries; and adding the multiple source data change dynamic thresholds to the visualization update trigger threshold constraint.

[0031] Preferably, source data and corresponding visualization project data over a period of time are collected from various visualization paths in the chip processing workshop. This may include equipment operating parameters (such as temperature, pressure, rotation speed, etc.), process parameters (such as lithography time, etching depth, etc.), quality inspection data (such as chip size, electrical performance indicators, etc.), and the content of the source data displayed on the visualization terminal in a specific form, such as equipment status charts, production progress bar charts, quality pass rate curves, etc. The collected source data is associated and integrated with the corresponding visualization project data to generate multiple historical mapping datasets for the source data and visualization projects. Each dataset records the specific display content of the visualization project when the source data takes a specific value at a specific time point or time period. For example, a certain historical mapping dataset records the display content of the equipment status chart corresponding to different temperature and pressure parameters of a certain lithography machine in a certain shift on a certain date, including whether the equipment is operating normally and the alarm information that occurs.

[0032] Preferably, the user with management authority refers to the manager of the chip processing workshop, who has the authority to configure and manage the visualization terminal. The user with management authority reads multiple minimum update dynamic boundaries of multiple visualization items input by the user with management authority. That is, the minimum update change range set by the user with management authority according to the importance and actual needs of different visualization items. For example, for the visualization item of equipment temperature, the minimum update dynamic boundary may be set to 2 degrees Celsius, which means that the visualization item needs to be updated only when the equipment temperature changes to or exceeds 2 degrees Celsius; while for the visualization item of production progress, the minimum update dynamic boundary may be set to the production quantity of 5 chips.

[0033] Preferably, using the visualization project as the dependent variable and the source data as the independent variable, statistical analysis and data mining are employed to perform dynamic response relationship analysis on the source data and visualization project data in each historical mapping dataset. A dynamic response relationship model is established to quantify the impact of changes in source data on the visualization project, i.e., to determine which changes in source data will lead to significant changes in the visualization project. Then, based on the dynamic response relationship model and the minimum dynamic update boundary set by the user with management permissions, a dynamic threshold for source data changes that meets this boundary is calculated. This threshold indicates the magnitude of change in the source data required to cause the visualization project to exceed its minimum dynamic update boundary. Finally, multiple dynamic thresholds for source data changes are added to the visualization update trigger threshold constraint to control the updating of the visualization project. When the real-time monitored changes in source data reach or exceed the corresponding dynamic threshold, the updating operation of the visualization project is triggered, ensuring that the visualization terminal can reflect the actual situation of the chip processing workshop in a timely and accurate manner.

[0034] Furthermore, the specific configuration of the dynamic response relationship analysis module 20 also includes determining whether there are multiple source data acquisition terminals corresponding to any visualization project; if so, performing single-item response relationship analysis and multiple joint response relationship analysis between multiple source data acquisition terminals and visualization projects based on the historical mapping dataset, determining the single-item source data change dynamic threshold and the multiple source data joint change dynamic threshold that satisfy the multiple minimum update dynamic boundaries, and using them as the source data change dynamic threshold corresponding to any visualization project; and constructing the multiple source data change dynamic thresholds using the source data change dynamic threshold corresponding to any visualization project.

[0035] Furthermore, the specific configuration of the dynamic response relationship analysis module 20 also includes: a single response relationship is the independent influence relationship of any one of the multiple source data acquisition terminals on the visualization project; and a multiple joint response relationship is the joint influence relationship of two or more source data acquisition terminals on the visualization project.

[0036] Preferably, it determines whether a visualization project depends on one or more source data acquisition terminals. If a visualization project depends on multiple source data acquisition terminals, it performs single-item response relationship analysis (the independent impact of each source data acquisition terminal on the visualization project) and multi-item joint response relationship analysis (the combined impact of multiple source data acquisition terminals on the visualization project). The single-item response relationship analysis assesses the impact of individual changes in each source data acquisition terminal on the visualization project. For example, changing only the equipment temperature and observing the changes in the visualization project (such as an equipment status chart) to determine the degree of independent impact of each source data acquisition terminal on the visualization project. The multi-item joint response relationship analysis assesses the impact of simultaneous changes in multiple source data acquisition terminals on the visualization project. For example, simultaneously changing the equipment temperature and pressure and observing the changes in the visualization project to determine the interaction between different source data and their combined impact on the visualization project.

[0037] Preferably, based on the analysis results of the historical mapping dataset and combined with the minimum update dynamic boundary set by the management user, the dynamic threshold for single-item change at each source data acquisition terminal and the dynamic threshold for joint change at multiple source data acquisition terminals are determined. Specifically, the dynamic threshold for single-item source data change refers to the magnitude of change required for each source data acquisition terminal to trigger an update of the visualization project when it changes alone; the dynamic threshold for joint change at multiple source data acquisition terminals refers to the combined magnitude of change required for multiple source data acquisition terminals to trigger an update of the visualization project when they change simultaneously. Finally, the dynamic thresholds for single-item source data change and the dynamic thresholds for joint change at multiple source data acquisition terminals corresponding to each visualization project are integrated to construct a dynamic threshold for joint change at multiple source data acquisition terminals. This is used to control the update triggering conditions of all visualization projects, ensuring that each visualization project can be updated while meeting its minimum update dynamic boundary, intelligently identifying and responding to significant changes in source data, avoiding unnecessary frequent updates, and ensuring that important changes are reflected on the visualization terminal in a timely manner, thereby improving the efficiency and reliability of the system.

[0038] The risk-accompanied analysis module 30 is used to perform visualization update processing of the multiple visualization paths based on the visualization update trigger threshold constraint. When the concurrent update demand is greater than the preset demand threshold, risk-accompanied analysis is performed based on the source dataset of the concurrent update demand to generate a risk-accompanied indicator set corresponding to the concurrent update item set.

[0039] Furthermore, the specific configuration of the risk-accompanied analysis module 30 also includes that the preset demand threshold is a data batch concurrent processing threshold determined based on the real-time network status of the chip processing workshop.

[0040] Preferably, multiple visualization paths are monitored and updated according to the visualization update trigger threshold constraint. That is, when the change of the source data reaches the corresponding dynamic threshold, the update of the corresponding visualization project is triggered. If multiple visualization projects trigger update requests at the same time within the same time period, and the number of concurrent update requests exceeds the system's preset demand threshold, risk-related analysis is performed based on the source dataset corresponding to the concurrent update requests. The preset demand threshold is the maximum number of concurrent data batch processing limits determined based on the real-time network conditions of the chip processing workshop to prevent system overload and ensure update efficiency and data transmission stability. The real-time network conditions include network bandwidth (calculated based on the actual available bandwidth of the current network to determine the amount of data that can be transmitted per unit time) and network latency (when the network latency is high, too many concurrent update requests may lead to cumulative delays in data transmission, resulting in untimely updates). Risk-associated analysis primarily assesses the potential risks arising from the simultaneous updating of multiple visualization items. For example, simultaneous updates to visualization items for multiple critical equipment indicate significant fluctuations or anomalies in the workshop production process, potentially impacting product quality and equipment safety. Through risk-associated analysis, a set of risk-associated indicators is generated corresponding to the current set of concurrently updated items. These indicators include various quantitative risk assessment metrics, such as risk level (high, medium, low), risk probability, and risk impact scope. For instance, a risk level indicator might suggest a high-risk equipment malfunction; a risk probability indicator might show the probability of a quality problem occurring at a specific value; and a risk impact scope indicator might indicate how many production lines or equipment will be affected by the current situation.

[0041] Preferably, assuming the chip manufacturing workshop's visualization management system has multiple visualization projects, such as temperature monitoring of equipment A, pressure monitoring of equipment B, and progress monitoring of production line processes, the dynamic thresholds for source data changes corresponding to each project are 3 degrees Celsius for temperature change, 0.5 MPa for pressure change, and 10 chips for production progress change. When it is detected that the temperature of equipment A has risen by 4 degrees Celsius (exceeding its dynamic threshold), the pressure of equipment B has risen by 0.6 MPa (exceeding its dynamic threshold), and the progress of that process on the production line has increased by 20 chips (exceeding its dynamic threshold), this constitutes a concurrent update request. If the total number of concurrent update requests exceeds the preset request threshold (for example, if the maximum is set to handle two concurrent updates simultaneously, but there are now three), a risk-related analysis is performed based on the source datasets corresponding to these three projects. It may be found that simultaneous abnormal changes in equipment A and B may lead to malfunctions in the coordinated operation between the equipment, thereby affecting the overall production progress. This generates a corresponding set of risk-related indicators, such as high risk level, high failure probability, and large impact range, thus helping managers to take measures in advance to avoid potential production accidents and losses.

[0042] Furthermore, the specific configuration of the risk-associated analysis module 30 also includes: extracting each source dataset corresponding to each concurrent update project in the concurrent update project set; determining the project type of each concurrent update project; performing chip processing risk identification contribution analysis based on the project type; constructing risk contribution weights for each concurrent update project; comparing the differences between each source dataset and the source data of the previous update node to determine each difference risk indicator; weighting each difference risk indicator with the risk contribution weights of each concurrent update project to generate each risk-associated indicator for each concurrent update project; and constructing the risk-associated indicator set using the risk-associated indicators.

[0043] Preferably, when multiple concurrent update projects occur (i.e., multiple visualization projects need to be updated simultaneously), the source datasets corresponding to each concurrent update project are extracted. Each concurrent update project's source dataset includes all source data information related to that visualization project, such as data collected from multiple sensors for equipment temperature, pressure, vibration frequency, etc. Each concurrent update project is then categorized to determine its project type, such as equipment monitoring, production progress, or quality inspection. Then, based on different project types, the potential risk contribution of each concurrent update project during chip manufacturing is analyzed. Specifically, abnormal source data in equipment monitoring projects may lead to equipment failure risks, thus affecting the continuity of chip manufacturing; abnormal source data in quality inspection projects may indicate chip quality defects. Based on the risk contribution level of each concurrent update project, a corresponding weight value is assigned to it, reflecting the risk contribution level of each concurrent update project. For example, monitoring projects for key process parameters that directly affect the core performance of the chip have a higher weight, while monitoring projects for auxiliary equipment status have a relatively lower weight. Example concurrent update project risk contribution weight data is shown in Table 1. Table 1. Concurrent Update Project Risk Contribution Weight Data Table

[0044] Preferably, the current source dataset for each concurrent update project is compared with the source data from the previous update to identify differences in data changes. For example, if the device temperature rises from 80 degrees Celsius in the previous update to 85 degrees Celsius, a change of 5 degrees Celsius is considered a difference. Based on these differences in source data, corresponding difference risk indicators are determined to quantify and assess the potential risks brought about by changes in source data. For example, for differences in device temperature, the difference risk indicator might be the temperature change rate. If the temperature change rate is too high, it may mean an increased risk of device overheating. Then, the risk contribution weight of each concurrent update project is multiplied by the corresponding difference risk indicator to obtain a weighted risk value. This weighted risk value is used as the risk-accompanying indicator for that concurrent update project. Each concurrent update project may have multiple risk-accompanying indicators, each corresponding to a different risk brought about by changes in source data. Finally, all risk-accompanying indicators for all concurrent update projects are summarized and integrated to form a risk-accompanying indicator set. This set comprehensively reflects the potential risk status of various aspects of the chip manufacturing process under the current concurrent update situation, helping managers to quickly and accurately identify and respond to potential risks.

[0045] Furthermore, the specific configuration of the risk-associated analysis module 30 also includes retrieving historical chip processing risk data by the project type; performing a detection importance analysis of processing anomaly events based on the historical chip processing risk data; and generating risk contribution weights for each concurrent update project.

[0046] Preferably, based on the determined project type, corresponding chip processing risk data is matched in the historical chip processing risk database. This may include past equipment failures due to excessive equipment temperature, order delays caused by production schedule delays, and quality problems caused by chip size deviations. By analyzing the historical chip processing risk data, various processing anomalies are identified, which may include equipment overheating, sudden pressure increases, abnormal vibration, chip surface scratches, and deviations in electrical performance parameters from standards. Then, a detection importance analysis of processing anomalies is performed. This involves assessing the detection importance of processing anomalies based on their impact on the chip processing process and the quality of the final product. For example, equipment failures caused by overheating may halt the entire production line, affecting the production of a large number of chips, and thus have a high detection importance. Minor surface scratches may have a small impact on chip performance, and thus have a relatively low detection importance. Furthermore, based on the results of the detection importance analysis of processing anomalies, a corresponding risk contribution weight is assigned to each concurrent update project to reflect the project's contribution to the overall risk during chip processing, ensuring a more accurate quantification of the relative importance of each concurrent update project in chip processing risk.

[0047] The configuration update processing module 40 is used to perform priority configuration and update processing on the concurrent update item set based on the risk-accompanying indicator set.

[0048] Furthermore, the specific configuration of the configuration update processing module 40 also includes: based on the risk-accompanying indicator set, extracting a first type of concurrent update items and a second type of concurrent update items whose risk-accompanying indicators are greater than a preset indicator threshold; configuring the update priority of the first type of concurrent update items to be higher than that of the second type of concurrent update items; and performing update processing on the first type of concurrent update items and the second type of concurrent update items under the constraint of the priority configuration result.

[0049] Preferably, the preset indicator threshold is a set benchmark value used to distinguish the severity of risks. By comparing the risk-related indicators with the preset indicator threshold, and based on different risk levels or processing priorities, a first category of concurrent update projects and a second category of concurrent update projects with risk-related indicators greater than the preset indicator threshold are extracted. Then, the first category of concurrent update projects is assigned a higher update priority than the second category of concurrent update projects. That is, when performing update processing, the first category of concurrent update projects is given priority to ensure that high-risk projects can be updated and processed in a timely manner, thereby responding to potential risks more quickly. Under the constraints of the priority configuration results, the first and second categories of concurrent update projects are updated and processed in order of priority. Specifically, for the high-priority first category of concurrent update projects, more system resources and attention are allocated to complete the update as soon as possible to reflect the true status and risk situation of the project in a timely manner. For the second category of concurrent update projects, after processing the first category of projects, update processing is performed according to resource availability and priority order to rationally allocate system resources, ensure that high-risk projects are processed first, improve overall risk response capabilities and operational efficiency, and effectively cope with various risks in the chip processing workshop.

[0050] In the above text, refer to Figure 1 This paper describes in detail an artificial intelligence-based chip manufacturing visualization management system according to embodiments of the present invention. Next, reference will be made to... Figure 2 This paper describes an artificial intelligence-based chip manufacturing visualization management method according to embodiments of the present invention. The artificial intelligence-based chip manufacturing visualization management method, such as... Figure 2 As shown, the method includes: collecting multiple visualization projects from the visualization terminal corresponding to the chip processing workshop, and constructing multiple visualization paths corresponding to the multiple visualization projects; performing dynamic response relationship analysis between source data and visualization projects on the multiple visualization paths, and constructing visualization update trigger threshold constraints for the source data; executing visualization update processing for the multiple visualization paths based on the visualization update trigger threshold constraints; when the concurrent update demand exceeds a preset demand threshold, performing risk-related analysis based on the source dataset of the concurrent update demand, and generating a risk-related indicator set corresponding to the concurrent update project set; and performing priority configuration and update processing on the concurrent update project set based on the risk-related indicator set.

[0051] In one possible implementation, the AI-based chip manufacturing visualization management method further includes: the multiple visualization items refer to preset display types displayed on the electronic screen of the visualization terminal, and any one of the multiple visualization paths includes several processing nodes executed from the source data acquisition end until the characteristic value of the visualization item is obtained.

[0052] In one possible implementation, the AI-based chip manufacturing visualization management method further includes: collecting historical data from the multiple visualization paths to generate multiple historical mapping datasets of source data and visualization projects; reading multiple minimum update dynamic boundaries of multiple visualization projects input by the management-authorized user of the visualization terminal; performing dynamic response relationship analysis between source data and visualization projects based on the multiple historical mapping datasets, with visualization projects as dependent variables and source data as independent variables, to determine multiple source data change dynamic thresholds that satisfy the multiple minimum update dynamic boundaries; and adding the multiple source data change dynamic thresholds to the visualization update trigger threshold constraint.

[0053] In one possible implementation, the AI-based chip manufacturing visualization management method further includes: determining whether there are multiple source data acquisition terminals corresponding to any visualization project; if so, performing single-item response relationship analysis and multiple joint response relationship analysis between multiple source data acquisition terminals and visualization projects based on historical mapping datasets, determining single-item source data change dynamic thresholds and multiple source data joint change dynamic thresholds that satisfy the multiple minimum update dynamic boundaries, and using them as source data change dynamic thresholds corresponding to any visualization project; and constructing the multiple source data change dynamic thresholds using the source data change dynamic thresholds corresponding to any visualization project.

[0054] In one possible implementation, the AI-based chip manufacturing visualization management method further includes: a single-response relationship is the independent influence relationship of any one of the multiple source data acquisition terminals on the visualization project; and a multiple joint response relationship is the joint influence relationship of two or more source data acquisition terminals on the visualization project.

[0055] In one possible implementation, the AI-based chip manufacturing visualization management method further includes: extracting each source dataset corresponding to each concurrent update project in the concurrent update project set; determining the project type of each concurrent update project, performing chip manufacturing risk identification contribution analysis based on the project type, and constructing risk contribution weights for each concurrent update project; comparing the differences between each source dataset and the source data of the previous update node to determine each difference risk indicator; weighting each difference risk indicator with the risk contribution weights of each concurrent update project to generate each risk-accompanying indicator for each concurrent update project; and constructing the risk-accompanying indicator set using the risk-accompanying indicators.

[0056] In one possible implementation, the AI-based chip manufacturing visualization management method further includes: retrieving historical chip manufacturing risk data by the project type; performing a detection importance analysis of manufacturing anomalies based on the historical chip manufacturing risk data; and generating risk contribution weights for each concurrent update project.

[0057] In one possible implementation, the AI-based chip manufacturing visualization management method further includes: extracting a first type of concurrent update items and a second type of concurrent update items based on the risk-accompanying indicator set, where the risk-accompanying indicators are greater than a preset indicator threshold; configuring the update priority of the first type of concurrent update items to be higher than that of the second type of concurrent update items; and performing update processing on the first type of concurrent update items and the second type of concurrent update items under the constraint of the priority configuration result.

[0058] In one possible implementation, the AI-based chip manufacturing visualization management method further includes: the preset demand threshold is a data batch concurrent processing threshold determined based on the real-time network status of the chip manufacturing workshop.

[0059] The AI-based chip manufacturing visualization management system provided in this embodiment of the invention can execute the AI-based chip manufacturing visualization management method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0060] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0061] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A chip manufacturing visualization management system based on artificial intelligence, characterized in that, include: The visualization path construction module is used to collect multiple visualization projects from the visualization terminal corresponding to the chip processing workshop and construct multiple visualization paths corresponding to the multiple visualization projects. The dynamic response relationship analysis module is used to perform dynamic response relationship analysis between source data and visualization projects on the multiple visualization paths, and to construct visualization update trigger threshold constraints for source data. The risk-accompanied analysis module is used to execute the visualization update processing of the multiple visualization paths based on the visualization update trigger threshold constraint. When the concurrent update demand is greater than the preset demand threshold, risk-accompanied analysis is performed based on the source dataset of the concurrent update demand to generate a risk-accompanied indicator set corresponding to the concurrent update item set. The configuration update processing module is used to perform priority configuration and update processing on the concurrent update item set based on the risk-accompanying indicator set.

2. The AI-based chip manufacturing visualization management system as described in claim 1, characterized in that, The multiple visualization projects refer to the preset display types displayed on the electronic screen of the visualization terminal. Any one of the multiple visualization paths includes several processing nodes executed from the source data acquisition end until the characteristic value of the visualization project is obtained.

3. The AI-based chip manufacturing visualization management system as described in claim 2, characterized in that, The steps performed by the dynamic response relationship analysis module include: Historical data is collected from the multiple visualization paths to generate multiple historical mapping datasets about the source data and the visualization projects. Read the minimum dynamic boundaries of multiple visualization items input by the management user of the visualization terminal; Using the visualization project as the dependent variable and the source data as the independent variable, the dynamic response relationship between the source data and the visualization project is analyzed based on the multiple historical mapping datasets to determine multiple source data change dynamic thresholds that satisfy the multiple minimum update dynamic boundaries. The dynamic thresholds for changes in the multiple source data are added to the visualization update trigger threshold constraint.

4. The AI-based chip manufacturing visualization management system as described in claim 3, characterized in that, The steps performed by the dynamic response relationship analysis module include: Determine whether there are multiple source data collection terminals for any given visualization project; If so, perform single-response relationship analysis and multi-joint response relationship analysis between multiple source data collection terminals and visualization projects based on the historical mapping dataset, determine the single-source data change dynamic threshold and multi-source data joint change dynamic threshold that satisfy the multiple minimum update dynamic boundaries, and use them as the source data change dynamic threshold for any visualization project. The multiple source data change dynamic thresholds are constructed using the source data change dynamic threshold corresponding to any visualization project.

5. The AI-based chip manufacturing visualization management system as described in claim 4, characterized in that, A single-response relationship is the independent influence of any one of the multiple source data acquisition terminals on the visualization project; a multi-joint response relationship is the joint influence of two or more source data acquisition terminals on the visualization project.

6. The AI-based chip manufacturing visualization management system as described in claim 1, characterized in that, The steps performed by the risk-associated analysis module include: Extract the source datasets corresponding to each concurrent update item in the concurrent update item set; The project type of each concurrent update project is determined, and a risk contribution analysis of chip processing risk is performed based on the project type to construct the risk contribution weight of each concurrent update project; The differences between each source dataset and the source data of the previous update node are compared to determine the risk indicators of each difference. The risk contribution weights of each concurrent update project are used to weight the various differential risk indicators to generate various risk-related indicators for each concurrent update project. The risk-associated indicator set is constructed using the aforementioned risk-associated indicators.

7. The AI-based chip manufacturing visualization management system as described in claim 6, characterized in that, The steps performed by the risk-associated analysis module include: Retrieve historical chip manufacturing risk data by the project type; Based on the historical chip processing risk data, an importance analysis of the detection of processing anomalies is performed to generate the risk contribution weight of each concurrent update item.

8. The AI-based chip manufacturing visualization management system as described in claim 1, characterized in that, The steps performed by the configuration update processing module include: Based on the risk-accompanying indicator set, extract the first type of concurrent update items and the second type of concurrent update items whose risk-accompanying indicators are greater than the preset indicator threshold. The update priority of the first type of concurrent update projects is configured to be higher than that of the second type of concurrent update projects; Under the constraints of the priority configuration results, update processing is performed on the first type of concurrent update items and the second type of concurrent update items.

9. The AI-based chip manufacturing visualization management system as described in claim 1, characterized in that, The preset demand threshold is a data batch concurrent processing threshold determined based on the real-time network status of the chip processing workshop.

10. A chip manufacturing visualization management method based on artificial intelligence, characterized in that, The method is applied to the AI-based chip manufacturing visualization management system according to any one of claims 1-9, the method comprising: Collect multiple visualization projects from the visualization terminal corresponding to the chip processing workshop, and construct multiple visualization paths corresponding to the multiple visualization projects; The dynamic response relationship between source data and visualization projects is analyzed for the multiple visualization paths, and a visualization update trigger threshold constraint for source data is constructed. Based on the visualization update trigger threshold constraint, the visualization update processing of the multiple visualization paths is executed. When the concurrent update demand is greater than the preset demand threshold, risk accompaniment analysis is performed based on the source dataset of the concurrent update demand to generate a risk accompaniment indicator set corresponding to the concurrent update item set. The concurrent update item set is then updated after priority configuration based on the aforementioned risk-related indicator set.

Images