Methods, systems, and graphical user interfaces for monitoring semiconductor manufacturing EAP systems

Abstract

The application discloses a method, system and graphical user interface for monitoring a semiconductor manufacturing EAP system, the method comprising: obtaining a predefined production process flow, the flow comprising a plurality of successive production steps; mapping each production step to a corresponding semiconductor production device and its corresponding EAP service instance; collecting a plurality of state data of each EAP service instance in real time; dynamically generating a visual flowchart according to the order of the production process flow, each node in the visual flowchart representing an EAP instance; dynamically updating the visual appearance of each node according to the real-time collected state data, in accordance with a predefined uniform coding rule, to simultaneously represent a plurality of states thereof; and displaying detailed information of the EAP instance in response to a user's interactive operation on the node. The application significantly improves the operation and maintenance efficiency and fault response speed of semiconductor production by deeply integrating EAP states with production processes.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of semiconductor manufacturing, and particularly relates to a method, system and graphical user interface for monitoring a semiconductor manufacturing EAP system. BACKGROUND

[0002] In a semiconductor manufacturing factory (Fab), the production of a Product Line usually involves dozens or even hundreds of equipment, each of which is controlled by an independent EAP (Equipment Automation Program) service program to achieve integration with the upper-layer manufacturing execution system (MES) and automated production. Currently, the management of EAPs in the factory is usually decentralized: the operation and maintenance perspective is isolated, and the operation and maintenance personnel can only check the status of a single EAP through its log or independent monitoring window, and cannot quickly grasp the health status of all EAPs from a global perspective; there is a lack of process association, and it is not intuitive to see the logical relationship of the equipment controlled by each EAP in the overall production flow (Flow), when an abnormality occurs in a certain process link, it is difficult to quickly locate the corresponding EAP and its upstream and downstream impact; the state perception is lagging, the connection state of the EAP and the equipment, the running state (such as Idle, Run, Down) of the equipment itself, and the control state (Online / Offline) of the EAP to the equipment are separate information, and the operator needs to cross-query multiple systems (such as MES, equipment interface) to make a comprehensive judgment, resulting in low efficiency and frequent misjudgment. Therefore, the prior art lacks a unified monitoring solution that deeply integrates EAP status with production flow, which leads to long problem troubleshooting time, low machine utilization rate and overall production efficiency, and becomes the main technical obstacle for the intelligent level improvement of semiconductor factory monitoring. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a method for monitoring a semiconductor manufacturing EAP system, which deeply integrates EAP status with production flow, realizes global visual monitoring, multi-dimensional state fusion and interactive drilling, and significantly improves the operation and maintenance efficiency and fault response speed of semiconductor production.

[0004] The first aspect of the present application discloses a method for monitoring a semiconductor manufacturing EAP system, comprising:

[0005] obtaining a predefined production process flow, the flow comprising a plurality of consecutive production steps;

[0006] mapping each production step to a corresponding semiconductor production equipment and its corresponding EAP service instance;

[0007] collecting multiple state data of each EAP service instance in real time;

[0008] generating a visual flowchart dynamically according to the sequence of the production process, each node in the visual flowchart representing an EAP instance;

[0009] updating the visual appearance of each node dynamically according to the state data collected in real time, to represent multiple states of the node simultaneously, according to a predefined unified coding rule;

[0010] displaying detailed information of the EAP instance in response to user interaction with the node.

[0011] The method, the multiple state data of each EAP service instance collected in real time is achieved by deploying a lightweight agent and an EAP interface, the lightweight agent continuously listens to the data stream of the EAP service instance through the EAP interface, and pushes the state change in the form of an event.

[0012] The method, the multiple state data includes EAP process state, EAP and device communication connection state, device running state, and EAP control state of the device.

[0013] The method, the predefined unified coding rule includes using the combination of the filling color, the border color, the internal icon and the text label of the node to represent the state, wherein the filling color is used to represent the device running state, the border color is used to represent the EAP process state, the internal icon is used to represent the communication connection state, and the text label is used to display the device identifier.

[0014] The method, the visual flowchart is generated dynamically based on a directed acyclic graph layout algorithm, to ensure that the nodes are arranged in the sequence of the production process.

[0015] The method, the visual appearance of each node is updated dynamically by using a graphical rendering engine, the graphical rendering engine listens to the state data change using a real-time communication protocol, and only updates the visual appearance of the node that has changed.

[0016] The method, the response to user interaction with the node includes clicking the node, and displaying the detailed information in a side or a pop-up window of the visual flowchart, the detailed information including running logs, performance indicators and historical alarm data.

[0017] The second aspect of the application discloses a system for monitoring a semiconductor manufacturing EAP system, comprising:

[0018] A process definition module is configured to configure the mapping relationship between the production process and the EAP instance.

[0019] a data collection module, configured to collect real-time state data of EAP processes, communication connection states, device running states and control states;

[0020] a graphical rendering engine, in communication connection with the process definition module and the data collection module, the engine being configured to generate a visual process diagram based on the mapping relationship and to drive the update of the visual appearance of nodes according to a unified coding rule based on the real-time state data;

[0021] a visual monitoring interface, configured to display the visual process diagram generated by the rendering engine and to provide a user interaction interface.

[0022] The above system, the data collection module realizes data collection through a lightweight agent or an EAP interface, and the lightweight agent is deployed on a server where an EAP instance is located.

[0023] The above system, the graphical rendering engine adopts an event-driven architecture and uses a WebSocket or SignalR protocol to receive real-time state updates of the data collection module.

[0024] The above system, the visual monitoring interface includes a main area for displaying nodes and connection lines and a detail area for displaying detailed data of a selected EAP instance when a user interacts.

[0025] The above system, the system further includes a storage module for caching production process flow mapping relationships and historical state data.

[0026] The third aspect of the present application discloses a graphical user interface of the visual process diagram in any aspect of the above method or system, displayed on a display device, the interface including:

[0027] a region displaying a plurality of nodes arranged in the order of a production process flow, the nodes being connected by connection lines representing the direction of the flow;

[0028] Each node simultaneously represents the survival state of the EAP service process corresponding thereto, the communication connection state with the device, the running state of the device and the control state of the EAP on the device through the combination of its fill color, border color, internal icon and text label.

[0029] In response to a user selecting a node, the side or pop-up window of the interface displays the detailed running log and historical data of the EAP instance corresponding to the node.

[0030] The graphical user interface, the fill color, the border color, the internal icon and the text label follow a predefined uniform coding rule, wherein the fill color in the first color represents that the device is running, the border color in the second color represents that the EAP process is running, and the internal icon in the first shape represents that the communication connection is normal.

[0031] The graphical user interface, the area supports zooming and panning operations to adapt to different scale production line visualizations.

[0032] The graphical user interface, the display of the detailed running log and the historical data includes a real-time performance index graph and an alarm list.

[0033] Compared with the prior art, the present application has the following advantages: the present application realizes deep binding of the EAP state and the physical world production process by obtaining a predefined production process and mapping each production step to a corresponding semiconductor production device and an EAP service instance, thereby breaking the limitation of isolated perspective in traditional operation and maintenance; since the method dynamically generates a visual process diagram according to the process sequence of the abstract EAP instance, the operator can intuitively master the logical relationship and health status of the EAP of the whole factory in one diagram without switching between multiple independent windows, which significantly improves the global monitoring efficiency and solves the positioning difficulty problem caused by the lack of process correlation in the background technology.

[0034] Further, the present application fuses the state information originally dispersed in multiple systems into one graph element by real-time collection of multiple state data (such as EAP process state, communication connection state, device running state and control state) of each EAP service instance and dynamic updating of the visual appearance of the node according to a predefined uniform coding rule; this multi-dimensional state fusion design enables the human eye to analyze complex information in seconds, greatly improving the accuracy and real-time performance of state perception and avoiding misjudgment and lag caused by cross-query in the background technology. At the same time, the method displays the detailed information of the EAP instance in response to the interactive operation of the user on the node, establishing a seamless drilling link from the global visual process diagram to the microscopic data, so that when an exception occurs, the operator can quickly locate the fault point and its upstream and downstream influence, realizing accurate tracing, which directly shortens the average fault repair time and improves the device utilization and overall production efficiency.

[0035] Overall, the method of the present application forms an intelligent monitoring closed loop through the synergistic effect of process mapping, state fusion and interactive drilling, effectively overcomes the defects of the prior art, and brings substantial improvement in operation and maintenance efficiency and production benefit.

[0036] The technical solutions of the present application will be further described in detail below with reference to the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 A flow chart of the method of embodiment 1 of the present application.

[0038] Figure 2 An effect diagram of the visualization flow chart of the present application.

[0039] Figure 3 A data flow diagram of collecting multiple status data of each EAP service instance.

[0040] Figure 4 A system module diagram of embodiment 2 of the present application. DETAILED DESCRIPTION

[0041] Embodiment 1

[0042] As shown in Figure 1 , a method for monitoring a semiconductor manufacturing EAP system, comprising:

[0043] acquiring a predefined production process flow, the flow comprising multiple successive production steps;

[0044] mapping each production step to a corresponding semiconductor production device and its corresponding EAP service instance;

[0045] collecting multiple status data of each EAP service instance in real time;

[0046] dynamically generating a visualization flow chart according to the order of the production process flow, each node in the visualization flow chart representing an EAP instance; the visualization flow chart is as shown in Figure 2 ;

[0047] dynamically updating the visual appearance of each node according to the real-time collected status data, in accordance with a predefined unified coding rule, to simultaneously represent multiple states thereof;

[0048] in response to user interaction with the node, displaying detailed information of the EAP instance.

[0049] In the implementation process, the system first acquires the predefined production process flow through the flow definition module. The flow is stored in a structured data format, such as an XML file, which clearly defines the sequence and dependency of the production steps. For example, the lithography step must be prior to the etching step, and the etching step must be prior to the thin film deposition step. The flow definition module parses the file, identifies each step as an independent unit, and verifies its continuity. In this way, the standardized loading of the production process flow is realized, providing an accurate basis for subsequent mapping and avoiding errors in manual configuration. For example, in a semiconductor workshop, the process engineer defines the flow as "lithography→etching→thin film deposition" in advance, and the system identifies three steps after acquisition and stores their sequence relationship.

[0050] Next, the system maps each production step to a corresponding semiconductor manufacturing device and its EAP service instance. The mapping process is based on queries to the device management database. For example, an SQL query might associate the lithography step with the lithography machine numbered "Tool_001" and its dedicated EAP service instance "EAP_Photo". The mapping relationships are stored in a relational database as key-value pairs, ensuring that each EAP instance uniquely corresponds to one production step. This approach binds abstract production steps to specific IT resources, creating a mirror relationship between the physical world and the digital system, thus improving the targeting of monitoring. For instance, for the etching step, the system maps it to device "Tool_002" and the EAP instance "EAP_Etch", and ensures mapping consistency through database transactions.

[0051] like Figure 3 As shown, real-time collection of multiple status data for each EAP service instance is achieved by deploying a lightweight agent. The lightweight agent runs on the server where the EAP instance resides, continuously monitoring the EAP service data stream through the EAP interface. For example, it checks the EAP process status (e.g., running or stopped), the communication connection status between EAP and the device (e.g., connected or disconnected), the device's operating status (e.g., processing, idle, or faulty), and the EAP's control status over the device (e.g., online or offline) every second via a heartbeat mechanism. The collected data is pushed to the central processor in the form of events, encapsulating status information in JSON format. This approach achieves low-latency collection and event-driven updates of status data, providing a reliable data source for real-time visualization and avoiding the overhead of polling. For instance, when the lithography machine changes from idle to processing status, the agent immediately captures the change and pushes event data, including timestamps and status values.

[0052] The dynamic generation of a visual flowchart based on the production process sequence is achieved through a graphical rendering engine. The engine uses a directed acyclic graph (DAG) layout algorithm to automatically arrange nodes, ensuring that lithography nodes come first, followed by etching nodes, and then thin film deposition nodes. Arrows connect the nodes to indicate the flow direction. The visual flowchart is rendered as vector graphics on the monitoring interface, using JavaScript libraries such as D3.js to process the graphics generation and supporting dynamic layout adjustments. This method achieves the logical and automatic generation of the visual flowchart, allowing maintenance personnel to intuitively understand the production relationships and reducing manual drawing time. For example, when the system generates the visual flowchart, the nodes are sequentially "EAP_Photo", "EAP_Etch", and "EAP_Film", with connecting lines showing the flow from lithography to etching, and then from etching to thin film deposition.

[0053] According to the real-time collected state data, the visual appearance of each node is dynamically updated according to the predefined unified coding rule. The unified coding rule defines the node attributes: the fill color represents the device running state (such as blue for processing, gray for idle), the border color represents the EAP process state (such as green for running, red for stopping), the internal icon represents the communication connection state (such as a chain icon for normal connection, a broken icon for disconnection), and the text label displays the device identifier (such as "Tool_001"). The rendering engine uses the WebSocket protocol to listen to state change events and only updates the appearance of the affected nodes, rather than refreshing the entire interface. In this way, the fusion display of multi-state information in a single graphic element is achieved, enabling users to analyze complex information within seconds and improving state awareness efficiency. For example, when the etching device running state changes to processing, the fill color of the node "EAP_Etch" is updated to blue; if the EAP process abnormally stops, the border becomes red, and the internal icon displays a broken chain.

[0054] In response to the user's interactive operation on the node, the detailed information of the EAP instance is displayed through the interactive interface of the monitoring interface. When the user clicks on the node, the system triggers the event handler to asynchronously load detailed information such as running logs, real-time performance indicators (such as CPU usage curve), and historical alarm data from the database. The detailed information is displayed in the form of a sidebar, supporting scrolling and filtering operations. In this way, seamless drilling from macroscopic visual flowchart to microscopic data is achieved, shortening the troubleshooting time and improving the efficiency of operation and decision-making. For example, after the user clicks on the "EAP_Film" node, the right side of the interface displays the recent log list, performance indicator curve, and historical alarms of the thin film deposition device, facilitating engineers to analyze abnormal reasons.

[0055] Through the cooperation of the above steps, the whole process monitoring of the semiconductor EAP system is realized, and the example based on the lithography-etching-thin film deposition scene demonstrates the complete chain from flow acquisition to interactive display. By deeply integrating the production process and EAP state, global perspective monitoring is achieved; by multi-state fusion display, information visualization is achieved; by real-time updating and interactive drilling, efficient fault response is achieved.

[0056] In one embodiment, the real-time collection of multiple state data of each EAP service instance is achieved by deploying a lightweight agent and an EAP interface, the lightweight agent continuously listens to the data stream of the EAP service instance through the EAP interface, and pushes the state changes in the form of events.

[0057] In implementation, a lightweight agent is deployed on the server where each EAP instance resides, for example, an agent program is installed on the server running EAP_Photo, which continuously monitors the process state, communication connection state, and other data streams of the EAP service using system APIs or custom scripts. The agent works in an event-driven manner, checking the EAP service heartbeat signal every second, and when the state changes, immediately encapsulates the data into a JSON format event message and pushes it to the central data processor through the message queue (such as RabbitMQ). This way avoids the traditional polling mechanism, reducing network bandwidth consumption and delay. Through the lightweight agent to realize data collection, it realizes efficient, low-cost real-time monitoring, and improves the response speed and reliability of the system. For example, in the lithography step, the agent monitors the EAP_Photo instance, and when the lithography equipment changes from idle state to processing state, the agent captures the device state change and pushes the event data, including timestamp and state value, to ensure timely data update.

[0058] In one embodiment, the plurality of state data includes EAP process state, EAP communication connection state with the device, device running state, and EAP control state on the device.

[0059] In implementation, the data collection module obtains these four states by querying the system log and device interface of the EAP service. For example, the EAP process state is determined by checking whether the PID file of the service process exists to determine running or stopping; the communication connection state is determined by testing the network port connectivity between EAP and device controller to determine connection or disconnection; the device running state is read from device sensors or MES system interface, such as processing, idle or failure; the control state is obtained by parsing the configuration parameters of EAP to get online or offline mode. The collected data is normalized into a standard format and stored in a time series database. By integrating four different sources of state data, comprehensive monitoring coverage is achieved, enabling operations personnel to comprehensively judge the health of the device and reducing the problem of information silos. For example, for the EAP_Etch instance of the etching step, the system simultaneously collects its process state (such as running), connection state (such as normal connection), device running state (such as processing), and control state (such as online), and displays these data in association, helping engineers quickly identify the root cause of the anomaly.

[0060] In one embodiment, the pre-defined unified coding rule includes using the combination of fill color, border color, internal icon, and text label of the node to represent the state, wherein the fill color is used to represent the device running state, the border color is used to represent the EAP process state, the internal icon is used to represent the communication connection state, and the text label is used to display the device identifier.

[0061] In implementation, the rendering engine predefines a set of visual coding rules: the fill color maps the device running status, e.g. blue means processing, gray means idle; the border color maps the EAP process status, e.g. green means running, red means stopped; the inner icon maps the communication connection status, e.g. the lock chain icon means normal connection, the broken icon means disconnection; the text label displays the device identifier, e.g. the device number "Tool_001". The rules are implemented by style sheets (e.g. CSS), and the engine dynamically applies these styles to the visual flowchart nodes according to real-time status data. Through such coding rules, the intuitive fusion of multi-state information in a single graphic element is achieved, so that users can analyze complex states in seconds without cross-querying, improving the readability and efficiency of the monitoring interface. For example, when the EAP_Film instance of the thin film deposition device is in normal running state, the node fill color is blue indicating that the device is processing, the border is green indicating that the EAP process is running, the inner display lock chain icon indicates normal connection, and the text label displays "EAP_Film". The overall node appearance clearly conveys all key states.

[0062] In one embodiment, the dynamically generated visual flowchart is implemented based on a directed acyclic graph layout algorithm to ensure that the nodes are automatically arranged in the order of the production process flow.

[0063] In implementation, the graphical rendering engine uses a directed acyclic graph algorithm, such as topological sorting, to handle the node order of the production process flow. The engine first parses the step sequence provided by the process definition module, treats each EAP instance as a graph node, and the dependency relationship between steps as a directed edge, ensuring no circular conflicts. Then, the algorithm automatically calculates the node positions, with the lithography node on the left, the etching node in the middle, and the thin film deposition node on the right, and uses arrow connection lines to represent the flow direction. In implementation, the engine calls the force-directed layout function of the graphics library (e.g. D3.js) to dynamically adjust the node spacing to avoid overlap. Through the directed acyclic graph layout algorithm, the logical automatic arrangement of the visual flowchart is achieved, making the visual interface truly reflect the order of the production process, reducing the workload of manual layout. For example, when the system generates the visual flowchart, the algorithm ensures that the EAP_Photo node comes first, the EAP_Etch node comes second, and the EAP_Film node comes last, with the connection lines pointing from lithography to etching and then to thin film deposition, intuitively displaying the production chain.

[0064] In one embodiment, the dynamic updating of the visual appearance of each node is achieved by a graphical rendering engine that listens to changes in status data using a real-time communication protocol and only updates the visual appearance of nodes that have changed.

[0065] In implementation, the rendering engine adopts an event-driven architecture, and uses a real-time communication protocol such as WebSocket to listen to state update events of the data collection module. When the state changes, for example, the device running state of the EAP_Etch instance changes from idle to processing, the engine only receives the changed data of the node, instead of refreshing the entire visual flowchart. The engine then queries the unified coding rules to update the visual attributes of the corresponding node, such as changing the fill color from gray to blue. In implementation, the engine uses a differential contrast algorithm to only redraw the changed part, reducing the interface rendering overhead. Through the local update mechanism of the graphical rendering engine, the efficiency and real-time performance of the monitoring interface are realized, avoiding the lag caused by full-screen refreshing and improving the user experience. For example, when the etching device suddenly fails, the device running state of the EAP_Etch node changes to failure, and the engine immediately updates the node fill color to red, while other nodes remain unchanged, ensuring quick response of the interface.

[0066] In one embodiment, the response to the user's interaction with the node includes clicking on the node, and displaying detailed information in a side or pop-up window of the visual flowchart, the detailed information including running logs, performance indicators, and historical alarm data.

[0067] In implementation, the visual monitoring interface provides a click event handler, and when the user clicks on a node, the system triggers an asynchronous request to load the detailed information of the EAP instance from the database. For example, after clicking on the EAP_Film node, the interface calls the backend API through AJAX to obtain running logs, performance indicators (such as CPU usage curve), and historical alarm data, and displays them in the form of a sidebar. The details area supports scrolling and filtering functions, such as sorting log entries by time. Through this interactive drilling design, seamless access from macro visual flowchart to micro data is realized, enabling operations and maintenance personnel to quickly and deeply troubleshoot problems, shortening the average fault repair time. For example, the user clicks on the EAP_Photo node of the photo step, and the interface displays the details on the right side, including a list of recent logs, real-time performance charts, and historical alarms, which engineers can use to analyze the causes of the photo machine anomaly.

[0068] Embodiment 2

[0069] As shown in Figure 4 A system for monitoring a semiconductor manufacturing EAP system, comprising:

[0070] A flow definition module for configuring the mapping relationship between the production process and the EAP instance;

[0071] A data collection module for real-time collection of EAP process state, communication connection state, device running state, and control state;

[0072] a graphical rendering engine, communicatively connected with the process definition module and the data collection module, the engine configured to generate a visual process graph based on the mapping relationship, and drive the update of node visual appearance according to a unified coding rule based on the real-time state data;

[0073] a visual monitoring interface for displaying the visual process graph generated by the rendering engine, and providing a user interaction interface.

[0074] In implementation, first, the process definition module is used to configure the mapping relationship between the production process and the EAP instance. In technical implementation details, the module is deployed on the central server in the form of a software component, and through a graphical user interface or a configuration file interface, it allows the user to define the production process. For example, the user can create a production step sequence on the interface by dragging and dropping, such as arranging the photolithography, etching, and thin film deposition steps in order, and specifying the corresponding device identifier and EAP instance name for each step. The module internally uses a database to store the mapping relationship, such as a table structure in a relational database, where each record contains the step name, device number, and EAP instance ID. In implementation, the module verifies the continuity and conflict-free of the process, ensuring that the photolithography step is prior to the etching step. Through the process definition module, standardized configuration and dynamic management of the production process are achieved, enabling the system to accurately map the physical production chain to the digital monitoring system, avoiding errors and inconsistencies in manual configuration. For example, in a semiconductor workshop, an engineer uses the module to configure the process as "photolithography → etching → thin film deposition", and maps the photolithography step to device Tool_001 and EAP instance EAP_Photo, the etching step to Tool_002 and EAP_Etch, and the thin film deposition step to Tool_003 and EAP_Film, and the module automatically saves these mappings for subsequent use.

[0075] The data collection module is used to collect EAP process status, communication connection status, device running status and control status in real time. In the technical implementation details, the module is deployed on each server where the EAP instance is located through a lightweight agent, and the agent program continuously monitors the EAP service using system calls or API interfaces. For example, the agent obtains the process status (such as running or stopping) by checking the PID file of the EAP process, obtains the communication connection status (such as connected or disconnected) by network ping test, reads the device running status (such as processing, idle or fault) from the device sensor or MES system interface, and parses the EAP configuration parameters to obtain the control status (such as online or offline). The collected data is pushed in the form of events, and a message queue such as Kafka is used to ensure reliable transmission. Through the data collection module, low-latency collection and centralized processing of multi-dimensional state data are realized, which enables the system to perceive the device health in real time and reduces the information island and delay problem. For example, in the photoetching step, the agent listens to the EAP_Photo instance, and when the photoetching device changes from idle to processing state, the agent immediately captures the change and pushes the event data containing the timestamp and state value for subsequent modules to use.

[0076] The graphical rendering engine is in communication connection with the flow definition module and the data collection module, generates a visual flowchart based on the mapping relationship, and drives the update of the visual appearance of the nodes based on real-time state data. In the technical implementation details, the engine uses an event-driven architecture and uses the WebSocket protocol to receive state updates from the data collection module and mapping information from the flow definition module. The engine first parses the mapping relationship to generate a directed acyclic graph representing the production process, where the nodes represent EAP instances and the edges represent the flow direction. Then, the engine applies a unified coding rule, such as filling the color of the node to represent the device running status (blue for processing), the border color to represent the EAP process status (green for running), and the internal icon to represent the communication connection status (lock chain icon for normal connection). The engine only updates the appearance of the nodes that have changed, rather than refreshing the entire screen, and uses a differential rendering algorithm to optimize performance. Through the graphical rendering engine, dynamic generation and real-time update of the visual flowchart are realized, making the monitoring interface efficiently respond to changes and improving the visual effect and user experience. For example, when the device running status of the EAP_Etch instance of the etching device changes to fault, the engine immediately updates the fill color of the corresponding node to red after receiving the event, while other nodes remain unchanged, ensuring a smooth interface.

[0077] The visual monitoring interface is used to display the visual flowchart generated by the rendering engine and provide a user interaction interface. In the technical implementation details, the interface is deployed in the form of a Web application, and HTML5 and JavaScript technologies are used to render the visual flowchart. The main area of the interface displays nodes and connection lines, and the nodes are arranged in the order of the production process, such as the photolithography node on the left, the etching node in the middle, and the thin film deposition node on the right. Users can trigger interactive events by clicking on the nodes, and the interface then loads detailed information such as running logs or performance indicators from the backend through AJAX requests and displays them in the sidebar. In implementation, the interface supports zooming and panning operations to adapt to production lines of different scales. Through the visual monitoring interface, intuitive global monitoring and interactive drilling are achieved, enabling operation and maintenance personnel to quickly access detailed data and improving fault troubleshooting efficiency. For example, after the user clicks on the EAP_Film node of the thin film deposition step, the interface displays the real-time log list and historical alarms of the device on the right side, and engineers can analyze abnormalities based on this.

[0078] The above modules work together to achieve comprehensive monitoring of the semiconductor EAP system. Through the process definition module configuration mapping, process standardization is achieved; through the data acquisition module real-time acquisition, data timeliness is achieved; through the graphical rendering engine dynamic update, interface efficiency is achieved; through the visual monitoring interface interactive display, user friendliness is achieved. These components collectively solve the problem of scattered monitoring in the prior art and improve the intelligent level of operation and maintenance of semiconductor factories.

[0079] In one embodiment, the data acquisition module implements data acquisition through a lightweight agent or an EAP interface, and the lightweight agent is deployed on the server where the EAP instance is located.

[0080] In implementation, the lightweight agent is installed as an independent process on each server running an EAP service instance, for example, deploying an agent program on the server running the EAP_Photo instance, which continuously monitors the status of the EAP service using system calls or custom scripts. The agent checks the EAP process status every second through a heartbeat mechanism, such as confirming running or stopping by querying the process ID file; at the same time, testing network port connectivity to obtain communication connection status; reading device running status from the device controller interface, such as processing or idle; and parsing EAP configuration parameters to obtain control status, such as online or offline mode. The collected data is packaged as structured events and pushed to the central data collection module through a message queue such as RabbitMQ. By deploying lightweight agents on local servers, distributed processing of data collection is achieved, reducing network latency and single point of failure risk, and improving system reliability and real-time performance. For example, in the lithography step, the agent is deployed on the server where EAP_Photo is located, and when the lithography device state changes, the agent immediately captures and pushes the event, ensuring timely data updates without the need for cross-server queries.

[0081] In one embodiment, the graphical rendering engine adopts an event-driven architecture and uses WebSocket or SignalR protocol to receive real-time state updates from the data collection module.

[0082] In implementation, the rendering engine is deployed as a microservice using an event-driven design, where the engine registers as an event listener to listen to state change events published by the data collection module. For example, the engine establishes a persistent connection through the WebSocket protocol, and when the data collection module pushes a state update for the EAP_Etch instance, the engine receives the event data in real time, and after parsing, only updates the visual appearance of the corresponding node. In implementation, the engine uses an asynchronous processing mechanism to avoid blocking the main thread, and handles network reconnection and message ordering through protocols such as SignalR. Through the event-driven architecture and real-time communication protocol, low-latency processing of state updates is achieved, enabling the monitoring interface to respond to changes in real time, improving user experience and system efficiency. For example, when the etching device changes from idle to processing state, the data collection module publishes an event, and the rendering engine receives it through WebSocket and immediately updates the fill color of the EAP_Etch node to blue, while the rest of the interface remains unchanged.

[0083] In one embodiment, the visual monitoring interface includes a main area for displaying nodes and connection lines, and a detail area for displaying detailed data of the selected EAP instance when user interaction occurs.

[0084] In implementation, the interface is realized in the form of a web application, the main area uses HTML5 Canvas or SVG technology to render the visual flowchart, and the nodes are arranged in the order of the production process flow, such as the photolithography node on the left, the etching node in the middle, and the thin film deposition node on the right, and the connection lines use arrows to represent the flow direction. The details area is designed in the form of a sidebar or a floating window, and when the user clicks on a node, the interface loads detailed information such as the running log, performance index curve, and historical alarm list from the backend through an AJAX request. In implementation, the interface uses responsive design, supports scaling and panning operations, and adapts to different screen sizes. Through the partition design of the main area and the details area, the spatial efficient use of the monitoring interface is realized, allowing users to simultaneously view the global process and detailed data, improving interaction efficiency and fault troubleshooting speed. For example, after the user clicks on the EAP_Film node of the thin film deposition step, the main area remains the visual flowchart display, and the details area displays the CPU usage curve and recent alarms of the device in real time, allowing engineers to quickly analyze abnormalities.

[0085] In one embodiment, the system further comprises a storage module for caching production process flow mapping relationships and historical state data.

[0086] In implementation, the storage module is integrated in the form of a database or a cache server, such as using Redis to cache real-time mapping relationships and using a time series database such as InfluxDB to store historical state data. The module interacts with the flow definition module and the data collection module through an API interface, caches the mapping relationships of the production process flow, such as the mapping of the photolithography step to the EAP_Photo instance, and periodically backs up historical state data for querying. In implementation, the module uses the LRU algorithm to manage the cache, ensuring fast access to high-frequency data. Through the caching function of the storage module, data persistence and fast retrieval are realized, reducing the system's dependence on real-time databases and improving response speed and robustness. For example, when the user queries historical state, the storage module directly returns the past 24 hours of running data of the EAP_Etch instance from the cache without real-time calculation, shortening the query time.

[0087] Embodiment 3

[0088] Referring to Figure 2 An example of a graphical user interface for a visual flowchart as described in embodiment 1 or embodiment 2, displayed on a display device, the interface comprising:

[0089] A region displaying a plurality of nodes arranged in the order of a production process flow, connected by connection lines representing the direction of the flow between nodes;

[0090] Each node uses a combination of its fill color, border color, internal icon, and text label to simultaneously represent the liveness status of its corresponding EAP service process, the communication connection status with the device, the running status of the device, and the control status of the EAP over the device.

[0091] In response to a user selecting a node, the interface or a pop-up window displays the detailed operation logs and historical data of the EAP instance corresponding to that node.

[0092] It should be noted that the implementation of the graphical user interface is first based on a display area, which is rendered using web technologies such as HTML5 and JavaScript, displaying multiple nodes arranged according to the production process flow sequence. Nodes represent EAP instances; for example, the photolithography step corresponds to the node EAP_Photo, the etching step corresponds to the node EAP_Etch, and the thin film deposition step corresponds to the node EAP_Film. The nodes are connected by arrow-shaped lines, indicating the flow direction from photolithography to etching and then to thin film deposition.

[0093] During implementation, the interface uses vector graphics libraries such as D3.js to dynamically generate nodes and connections, ensuring the layout automatically adapts to the screen size. This streamlined arrangement enables intuitive visualization of production relationships, allowing maintenance personnel to grasp the logic of all equipment in the plant at a glance, breaking the limitations of traditional isolated monitoring. For example, in the monitoring interface, nodes are arranged sequentially from left to right, and the connections clearly show the transition from photolithography to etching, and from etching to thin film deposition, helping users quickly understand the production chain.

[0094] Each node, through a combination of its fill color, border color, internal icon, and text label, simultaneously indicates the liveness status of the corresponding EAP service process, the communication connection status with the device, the running status of the device, and the control status of the EAP over the device.

[0095] During technical implementation, the interface predefines a set of unified coding rules: fill color maps the device's operating status, such as blue for processing and gray for idle; border color maps the EAP process status, such as green for running and red for stopped; internal icons map the communication connection status, such as a chain icon for a normal connection and a broken icon for a disconnection; and text labels display the device identifier, such as "EAP_Photo" or the device number.

[0096] In implementation, the interface updates the appearance of the nodes in real time through CSS styles and dynamic property bindings, such as modifying only the visual properties of the corresponding node when the state data changes. Through this multi-state fusion display, the complex information is expressed concisely within a single graph element, enabling users to analyze the health of the device within seconds and improving monitoring efficiency. For example, when the lithography device EAP_Photo instance is in a normal running state, the node is filled with a blue color indicating that the device is processing, the border is green indicating that the EAP process is running, the internal display chain icon indicates that the connection is normal, and the text label displays "Tool_001". The entire node appearance conveys all key states without the need for additional queries.

[0097] In response to the user selecting a certain node, the side or pop-up window of the interface displays the detailed running log and historical data of the EAP instance corresponding to the node. In the technical implementation details, the interface integrates an event handling mechanism, and a JavaScript event is triggered when the user clicks on the node. The detailed information such as running log (including timestamp and event description), performance indicator curve (such as CPU usage), and historical alarm list is loaded asynchronously from the backend database through AJAX request. The display area is presented in the form of a sidebar or a modal window, supporting scrolling and filtering functions. In implementation, the interface uses responsive design to ensure that the details display does not block the main visual flowchart. Through this interactive drilling, a seamless transition from a macro view to micro data is achieved, enabling operations and maintenance personnel to quickly and deeply troubleshoot problems and shorten the fault response time. For example, after the user clicks on the EAP_Etch node of the etching step, a detail window pops up on the right side of the interface, displaying the latest log entries, real-time performance charts, and alarm records in the past 24 hours for the device. Engineers can analyze the etching machine abnormality based on this.

[0098] In one embodiment, the fill color, border color, internal icon, and text label follow a predefined uniform coding rule, where the fill color is a first color indicating that the device is running, the first color here is blue, the border color is a second color indicating that the EAP process is running, the second color here is green, and the internal icon is a first shape indicating that the communication connection is normal, and the first shape here is a chain.

[0099] In implementation, the interface defines these rules through style sheets and scripting languages, such as setting the visual properties of a node using CSS classes: the fill color is achieved through the background-color property, blue corresponding to the running state of the device in processing; the border color is achieved through the border-color property, green corresponding to the EAP process state in running; the internal icon is achieved through SVG icons or font icons, and the lock chain shape represents that the communication connection is normal. The text label displays the device identifier through the HTML text element. In implementation, the interface engine dynamically binds state data to these styles, and only updates the corresponding properties when the state changes. Through such specific coding rules, the standardized and intuitive display of state information is realized, so that users can quickly interpret the node state without training, improving the accessibility and efficiency of the monitoring interface. For example, in the photo step, when the device corresponding to the EAP_Photo instance is in the running state, the node fill color is displayed in blue, the border is green indicating that the EAP process is normal, the internal lock chain icon indicates that the connection is stable, and the text label displays "Tool_001". The entire node appearance consistently conveys all key information, avoiding ambiguity.

[0100] In one embodiment, the region supports zooming and panning operations to adapt to different scale production line visualizations.

[0101] In implementation, the interface realizes zooming and panning functions through the event handling mechanism of the integrated graphics library, such as adding mouse and touch event listeners using JavaScript libraries such as D3.js or Leaflet. Zooming operation is triggered by mouse wheel or pinch gesture, and the interface dynamically adjusts the zoom scale of nodes and connection lines to ensure clear and readable content; panning operation is realized by dragging the interface region, allowing users to slide to view different parts of the visualization flowchart. In implementation, the interface uses transformation matrix to handle coordinate conversion, and sets zooming limits to avoid excessive zooming. By supporting zooming and panning operations, the adaptive display of the monitoring interface is realized, allowing operations and maintenance personnel to flexibly view the global or details of large-scale production lines, improving the practicality and user experience of the interface. For example, when viewing a production line containing dozens of EAP instances, users can zoom in on the photo area to carefully check the state of the EAP_Photo node, or slide to the etching area to view the EAP_Etch node, thereby efficiently managing complex processes.

[0102] In one embodiment, the display of detailed running logs and historical data includes real-time performance indicator graphs and alarm lists.

[0103] In implementation, the interface realizes these displays through chart libraries and list components, such as integrating Chart.js to draw real-time performance indicator graphs, display curves of CPU usage, memory occupation and other data changing over time; at the same time, HTML tables or list elements are used to show alarm lists, including alarm time, description and level. In implementation, the interface obtains data from the backend through AJAX requests, dynamically updates the curve graph and list, and supports filtering and sorting functions. By integrating the curve graph and the alarm list, the visual analysis of historical data is realized, which enables the operation and maintenance personnel to track trends and identify patterns, and improves the depth and accuracy of fault diagnosis. For example, after the user clicks the EAP_Film node of the thin film deposition step, the detail area displays the real-time performance indicator graph, showing the fluctuation of the device CPU usage in the past hour, and the alarm list lists the recent abnormal events, and the engineer can quickly locate the problem source in combination with the curve trend and alarm records.

[0104] The above is only a preferred embodiment of the present application, not any limitation on the present application, any simple modification, change and equivalent structure change of the above embodiment according to the technical essence of the present application are still within the protection scope of the technical solution of the present application.

Claims

1. A method for monitoring a semiconductor manufacturing EAP system, characterized by, Comprising: acquiring a predefined production process flow, the flow comprising a plurality of successive production steps; mapping each production step to a corresponding semiconductor production equipment and its corresponding EAP service instance; collecting a plurality of state data of each EAP service instance in real time; generating a visual flowchart dynamically according to the order of the production process flow, each node in the visual flowchart representing an EAP instance; updating the visual appearance of each node according to the real-time collected state data, to represent a plurality of states simultaneously, according to a predefined uniform coding rule, the predefined uniform coding rule comprising using a combination of fill color, border color, internal icon and text label of a node to represent states, wherein the fill color is used to represent the equipment running state, the border color is used to represent the EAP process state, the internal icon is used to represent the communication connection state, and the text label is used to display the equipment identifier; in response to user interaction operation on the node, displaying detailed information of the EAP instance.

2. The method of claim 1, wherein, The collecting of the plurality of state data of each EAP service instance in real time is achieved by deploying a lightweight agent and an EAP interface, the lightweight agent continuously listens to the data stream of the EAP service instance through the EAP interface, and pushes the state changes in the form of events.

3. The method of claim 1, wherein, The plurality of state data includes EAP process state, EAP and device communication connection state, device running state and EAP control state of the device.

4. The method of claim 1, wherein, The dynamic generation of the visual flowchart is based on a directed acyclic graph layout algorithm to ensure that the nodes are automatically arranged according to the order of the production process flow.

5. The method of claim 1, wherein, The dynamic updating of the visual appearance of each node is achieved by a graphical rendering engine, which uses a real-time communication protocol to listen to state data changes and only updates the visual appearance of the nodes that have changed.

6. The method of claim 1, wherein, The response to user interaction operation on the node includes clicking the node, and displaying detailed information in a side or pop-up window of the visual flowchart, the detailed information including running log, performance indicators and historical alarm data.

7. A system for monitoring a semiconductor manufacturing EAP system, characterized by, Comprising: a flow definition module for configuring the mapping relationship between the production process flow and the EAP instance; a data collection module for collecting EAP process state, communication connection state, device running state and control state in real time; a graphical rendering engine in communication connection with the flow definition module and the data collection module, the engine being configured to generate a visual flowchart based on the mapping relationship, and to drive the updating of the visual appearance of the nodes according to a uniform coding rule based on the real-time state data; the uniform coding rule comprising using a combination of fill color, border color, internal icon and text label of a node to represent states, wherein the fill color is used to represent the equipment running state, the border color is used to represent the EAP process state, the internal icon is used to represent the communication connection state, and the text label is used to display the equipment identifier; a visual monitoring interface for displaying the visual flowchart generated by the rendering engine, and providing a user interaction interface.

8. A graphical user interface for displaying a visualized flowchart as claimed in any one of the claims 1-7, displayed on a display device, characterized in that, The interface comprises: One area displays a plurality of nodes arranged in sequence based on a production process flow, and the nodes are connected by connection lines representing the flow direction; Each node simultaneously represents the survival state of the corresponding EAP service process, the communication connection state with the device, the running state of the device, and the control state of the EAP on the device through the combination of its fill color, border color, internal icon, and text label. In response to a user selecting a certain node, one side or a pop-up window of the interface displays detailed running logs and historical data of the EAP instance corresponding to the node.

9. The graphical user interface of claim 8, wherein, The fill color, border color, internal icon, and text label follow a predefined unified coding rule, wherein the fill color being a first color indicates that the device is running, the border color being a second color indicates that the EAP process is running, and the internal icon being a first shape indicates that the communication connection is normal.

Images