Virtual environment modeling method for OHT system of semiconductor factory

Abstract

The invention discloses a virtual environment modeling method for an OHT system of a semiconductor factory, and belongs to the technical field of semiconductor production, and the method comprises the following steps: data collection and preprocessing: collecting field facility data of a semiconductor factory production line, operation related data of an OHT vehicle, and product information data, and carrying out the preprocessing of the data, so as to obtain basic data; virtual environment modeling: constructing a virtual environment model according to the basic data; simulation and optimization: carrying out simulation operation based on the constructed virtual environment model, and carrying out performance evaluation; then, optimization is carried out according to a simulation result and a performance evaluation condition; and verifying and applying a result. According to the method, accurate modeling and simulation are carried out on the OHT system, the configuration and scheduling strategy of the OHT system is optimized, resource utilization can be optimized, the cost can be reduced, and the stability, reliability and flexibility of the system can be enhanced.

Description

Technical Field

[0001] The present invention mainly relates to the field of semiconductor production technology, and in particular to a virtual environment modeling method for an OHT system in a semiconductor factory. Background Art

[0002] In semiconductor factories, the Overhead Transport System (OHT system) is an important component for automated material handling, which can effectively transfer wafers and other materials from one workstation to another. The OHT system plays a key role in semiconductor factories. It is responsible for the fast and accurate transportation of wafers and other materials. The OHT system is usually composed of a track network, an automatic guided vehicle (AGV), a control system, etc., and can operate efficiently in a three-dimensional space, avoiding the congestion problem of ground traffic. An efficient OHT system can significantly reduce material handling time, thereby increasing the throughput of the entire production line.

[0003] Although existing OHT systems are able to meet basic material transportation needs, there are still some challenges and shortcomings in practical applications: Due to the dynamic nature of production tasks, how to plan the optimal path for OHT vehicles is a difficult problem. Unreasonable path planning will lead to increased material waiting time, which in turn affects production efficiency. Existing OHT systems are often designed based on fixed parameters. When production conditions change, it is difficult to quickly adjust the system configuration to adapt to new needs. Although many modern OHT systems are equipped with sensors and data acquisition devices, how to use this data to improve system performance remains a challenge.

[0004] Current semiconductor fab OHT systems face a number of technical challenges that limit their performance and efficiency. The main issues include:

[0005] Path planning and optimization: In the complex semiconductor manufacturing environment, the path planning of the OHT system directly affects the material handling efficiency. Traditional path planning methods are often based on fixed rules and static data, which cannot adapt to the ever-changing production needs and are prone to problems such as path conflicts and long material waiting times.

[0006] System performance evaluation and optimization: Existing OHT systems often have difficulty accurately evaluating their performance and optimizing them effectively. The lack of effective tools and methods to simulate different scenarios and configurations makes system optimization difficult.

[0007] Material flow analysis and prediction: Material flow changes frequently during semiconductor manufacturing, requiring real-time adjustment of the OHT system’s operating strategy. However, existing OHT systems often have difficulty responding to changes in material flow in a timely manner.

[0008] Resource allocation and scheduling: In the semiconductor manufacturing process, improper allocation of OHT system resources (such as the number and type of OHT vehicles) will lead to inefficiency. Existing systems often lack an effective resource scheduling mechanism.

[0009] Handling of abnormal situations: In actual production, abnormal situations such as equipment failure and material shortage may occur, which will affect the normal operation of the OHT system. Summary of the invention

[0010] The technical solution of the present invention aims at the technical problem that the existing technical solutions are too single, and provides a solution that is significantly different from the existing technologies. It mainly provides a virtual environment modeling method for the OHT system of a semiconductor factory, aiming to optimize the configuration and scheduling strategy of the OHT system by accurately modeling and simulating the OHT system, thereby improving the overall efficiency of the semiconductor production line.

[0011] The technical solution adopted by the present invention to solve the above technical problems is:

[0012] A method for modeling a virtual environment of an OHT system in a semiconductor factory comprises the following steps:

[0013] S1. Data collection and preprocessing: Collect on-site facility data of semiconductor factory production lines, operation-related data of OHT vehicles, and product information data, and preprocess them separately to obtain basic data;

[0014] S2, virtual environment modeling: constructing a virtual environment model according to the basic data in step S1;

[0015] S3, simulation and optimization: based on the virtual environment model constructed in step S2, simulate and run, and perform performance evaluation; then optimize according to the simulation results and performance evaluation;

[0016] S4. Result verification and application.

[0017] Specifically, in step S1, the field facility data of the semiconductor factory production line includes the position coordinates, track distances, locations of material storage areas, and connection relationships between various workstations on the production line;

[0018] and / or, the operation-related data of the OHT vehicle includes specification parameters of the OHT vehicle and load conditions, driving routes, and stop times of the OHT vehicle in different time periods;

[0019] And / or, the product information data includes product model, production batch, required material type and quantity.

[0020] Specifically, in step S2, constructing the virtual environment model includes physical environment modeling, OHT system model construction, workstation and material storage area models, and control logic settings.

[0021] Specifically, in step S2, constructing a virtual environment model includes three-dimensional modeling of physical objects, virtual space construction and physical engine settings. The three-dimensional modeling of physical objects includes aerial track modeling, transport vehicle modeling, and modeling of other related equipment in a semiconductor factory; the virtual space construction includes creating a virtual scene and importing digital objects.

[0022] Specifically, in step S3, the simulation operation includes intelligent scheduling, path planning, key indicator recording, parameter change experiment, and efficient parameter exploration;

[0023] and / or, the performance evaluation includes setting evaluation indicators, statistical analysis, multi-indicator comprehensive evaluation, sensitivity analysis, time series prediction and abnormal environment testing;

[0024] And / or, the optimization includes configuration parameter adjustment, machine learning model training, iterative optimization, implementation and optimization feedback.

[0025] Furthermore, the intelligent scheduling is as follows: when the entire map is designed in modules, the vehicle is first selected according to the starting point of the task; secondly, an idle vehicle in the area where the task is located is selected; thirdly, vehicles in adjacent areas are selected; then, vehicles whose task destination is in the area among non-idle vehicles in the map are selected; finally, an idle vehicle is selected in the entire map.

[0026] Furthermore, the path planning is: using Dijkstra to perform path planning, starting from the starting point, and gradually finding and determining the shortest path to other vertices in the graph.

[0027] Furthermore, the purpose of the sensitivity analysis is to determine the degree of influence of key indicators and their combinations on system performance;

[0028] And / or, the time series prediction is: using a clustering algorithm to process time series data to predict future performance.

[0029] Furthermore, the abnormal environment test is as follows: according to the actual operating environment of the system and the abnormal situations that may be encountered, a series of simulation scenarios are defined; these scenarios cover common error types, failure modes and extreme conditions; a test environment similar to the production environment is built to ensure that the test environment can truly reflect the actual situation of the production environment.

[0030] Specifically, step S4 includes:

[0031] Field testing: deploying the optimized OHT system configuration in the actual production environment and monitoring its performance, recording actual production data and comparing it with simulation results;

[0032] Continuous optimization: Based on the results of field tests, further adjust the model parameters in the virtual environment; regularly update the model to reflect the latest changes in production conditions.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] (1) The present invention collects the on-site facility data of the semiconductor factory production line, the operation-related data of the OHT vehicle, and the product information data, and pre-processes them as basic data, thereby ensuring the integrity and accuracy of the model input data and laying an excellent foundation for subsequent virtual environment modeling.

[0035] (2) The present invention constructs a virtual environment model based on highly accurate basic data collected in multiple modules, which can reduce the complexity of modeling and more easily capture system characteristics and behaviors. Through virtual environment modeling, problems caused by human operational errors can be reduced, the standardization of the production process is improved, and the risk of dependence on manual labor is reduced.

[0036] (3) In terms of path planning and optimization, the present invention is based on the constructed virtual environment model and can dynamically adjust the path of the OHT vehicle according to real-time production data through an adaptive path planning method, thereby effectively reducing waiting time and path conflicts; and through adaptive path planning, the need for human intervention is reduced, reducing the possibility of errors, which is ultimately conducive to improving optimization efficiency and effectiveness.

[0037] (4) In terms of system performance evaluation and optimization, the present invention can simulate the operation of the OHT system under various production conditions based on the constructed virtual environment model, thereby helping engineers evaluate the system performance under different configurations and optimize the configuration of the OHT system accordingly to obtain the optimal OHT system.

[0038] (5) In terms of material flow analysis and prediction, the present invention combines historical data and real-time data and uses machine learning technology to predict future material flow trends and adjust the scheduling strategy of the OHT system in advance.

[0039] (6) In terms of resource allocation and scheduling, the present invention uses advanced machine learning technology for intelligent scheduling, which improves the accuracy and automation level of scheduling, allowing the system to quickly respond to changes in production plans and emergencies, that is, it can dynamically adjust the path and scheduling of OHT vehicles according to real-time production conditions, thereby improving the flexibility of the system. Through detailed simulation and optimization of the OHT system, resources can be reasonably allocated to avoid over-investment or idle resources; thereby, unnecessary waste of resources can be reduced, and operating costs can be reduced; and through predictive scheduling strategies, production demand and resource supply can be better matched; the probability of OHT system failure can also be reduced, and the cost of repair and maintenance can be reduced.

[0040] (7) In terms of abnormal situation handling, the present invention can preview the system response under abnormal situations based on the constructed virtual environment model, and formulate corresponding emergency strategies to improve the robustness and flexibility of the system. Such an abnormality handling mechanism can effectively deal with emergencies and ensure the continuity of the production process. By regularly updating the model to reflect the latest changes in production conditions, the stable operation of the system is ensured.

[0041] (8) The present invention can be customized: According to the specific needs of different semiconductor factories, the model parameters can be flexibly adjusted to achieve personalized customization. The model can be adjusted according to the actual production situation of the factory to ensure the optimal effect.

[0042] In summary, the present invention can not only optimize resource utilization and reduce costs, but also enhance the stability, reliability and flexibility of the system through high-precision data acquisition, modular and parametric design, intelligent simulation and optimization technology, multi-source data fusion, and the combination of expert knowledge and machine learning.

[0043] The present invention will be explained in detail below with reference to the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The present invention provides a flow chart of a method for modeling a virtual environment of an OHT system in a semiconductor factory. DETAILED DESCRIPTION

[0045] To facilitate understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings, but the present invention can be implemented in different forms and is not limited to the embodiments described in the text. On the contrary, these embodiments are provided to make the content disclosed in the present invention more thorough and comprehensive.

[0046] Please refer to the attached Figure 1 , a semiconductor factory OHT system virtual environment modeling method, comprising the following steps:

[0047] S1, data collection and preprocessing;

[0048] S2, physical environment modeling, OHT system model construction;

[0049] S3. Determine whether the path planning and scheduling algorithm is completed? If completed, proceed directly to step S4; if not completed (the path planning and scheduling algorithm may cause OHT congestion during the actual execution of the task or the task completion time is longer than expected, it is considered incomplete), then adjust the algorithm appropriately (including task generation strategy, task distribution strategy and OHT scheduling strategy), and then proceed to step S4;

[0050] S4, simulation and optimization;

[0051] S5. Result verification and application.

[0052] Embodiment 1: A method for modeling a virtual environment of an OHT system in a semiconductor factory, comprising the following steps:

[0053] 1. Data collection and preprocessing

[0054] Data collection: Gain in-depth understanding of the production process, material handling requirements, and specific functions and performance indicators of the OHT system of the semiconductor factory. This includes collecting factory layout diagrams, equipment specification sheets, process flow diagrams, and existing control system documents. Obtain the location coordinates of all workstations on the production line, track distances, locations of material storage areas, and connections between each workstation; collect OHT vehicle specifications, including but not limited to maximum speed, acceleration, load capacity, etc.; record historical data such as load conditions, driving routes, and stop times of OHT vehicles in different time periods; collect production plans for semiconductor production lines, including product models, production batches, required material types and quantities, and other information.

[0055] Data preprocessing: Clean the collected data to remove invalid or erroneous data items; format the data in a unified manner to ensure data consistency for subsequent processing; standardize the data, such as converting timestamps into a unified time format, etc.

[0056] Specifically, the preprocessing of relevant data of workstations on the production line includes:

[0057] For location coordinates, check whether the coordinate data is within a reasonable geographical range, and delete or correct obviously unreasonable location data;

[0058] For the connection relationship between workstations, verify whether the connection relationship conforms to the physical layout and check whether there are duplicate or incorrect connections.

[0059] Standardize format: Make sure all location coordinates are in the same geographic coordinate system.

[0060] Specifically, the preprocessing of the operation-related data of the OHT vehicle includes:

[0061] For specification parameters, check whether the values ​​are within a reasonable range, such as the maximum speed should not exceed the theoretical limit;

[0062] For load conditions, verify whether the load data is consistent with the vehicle's maximum load capacity;

[0063] For driving routes, verify whether the driving routes are consistent with the actual layout of the workstation;

[0064] Timestamp conversion: Convert all timestamps to a unified time format (such as UTC time) and handle time zone differences.

[0065] 2. Virtual environment modeling

[0066] (1) Physical environment modeling: Use CAD software to draw the physical layout of the semiconductor factory, including the layout of the OHT track, the location of the workstations, the material storage area, etc. Create a three-dimensional model to more intuitively display the structure and operating environment of the OHT system.

[0067] (2) OHT system model construction: Create a model of the OHT vehicle in the simulation software, including its physical characteristics and control logic. Set the interaction logic between the OHT vehicle and the workstation and material storage area, such as how to load and unload materials, how to communicate with the workstation, etc. Set the maximum speed, acceleration and other parameters of the OHT vehicle according to actual conditions.

[0068] (3) Workstation and material storage area model: Establish a workstation model to define the conditions for receiving materials and the logic for processing materials. Establish a material storage area model to define the logic for materials entering and exiting.

[0069] (4) Control logic setting: In terms of path planning, design a path calculation module based on the Dijkstra algorithm or other advanced path planning algorithms. In terms of scheduling algorithms, design a priority-based scheduling algorithm that takes into account factors such as the urgency of materials and the urgency of workstation demand.

[0070] 3. Simulation and optimization

[0071] Ⅰ.Simulation operation:

[0072] (1) Intelligent Scheduling: When the entire map is designed in modules, the vehicle is first selected based on the starting point of the mission. Next, an idle vehicle in the mission area is selected. Next, vehicles in adjacent areas are selected. Then, vehicles whose mission destination is in the area are selected from non-idle vehicles on the map. Finally, idle vehicles are selected from the entire map.

[0073] (2) Path planning: Dijkstra is used for path planning. Starting from the starting point, the shortest path to other vertices in the graph is gradually found and determined. The core idea of ​​the algorithm is to continuously select the node closest to the starting point among the currently unvisited nodes, and update the distance value of the node's adjacent nodes until all nodes have been visited; during this period, elements including the working area, obstacles, starting point, and end point will all be factors to consider whether to bypass.

[0074] (3) Key indicator recording: During the simulation process, key indicators such as the OHT vehicle’s driving trajectory, stop time, and material handling time are recorded in real time. These data are the basis for subsequent performance evaluation and system optimization.

[0075] (4) Parameter variation experiment: Run multiple simulation scenarios by changing different parameter settings (such as the number of OHT vehicles, path planning algorithm, workstation processing capacity, etc.). Each scenario represents a possible system configuration or optimization solution.

[0076] (5) Efficient parameter exploration: Use techniques such as batch parameter sweeps, response surface methods, or parallel computing frameworks to more efficiently explore parameter space. These techniques can significantly reduce simulation time while maintaining the accuracy and reliability of the results.

[0077] II. Performance evaluation:

[0078] (1) Setting evaluation indicators: According to actual needs, set a series of performance evaluation indicators, such as average waiting time, OHT vehicle utilization rate, material turnover rate, system throughput, etc. These indicators should be able to fully reflect the operating efficiency and performance of the system.

[0079] (2) Statistical analysis: Use statistical methods and hypothesis testing to conduct in-depth analysis of simulation results. By comparing performance indicators under different scenarios, the bottleneck areas and optimization space in the system can be identified.

[0080] (3) Multi-index comprehensive evaluation: Combined with principal component analysis and multi-index comprehensive evaluation methods, the overall performance of the system is comprehensively evaluated. These methods can help determine the relative importance and mutual influence between various indicators.

[0081] (4) Sensitivity analysis: A sensitivity analysis is performed to determine the degree of impact of key indicators and their combinations on system performance. This helps identify which parameters or factors have the greatest impact on system performance, thereby guiding subsequent optimization work.

[0082] (5) Time series prediction: Use clustering algorithms to process time series data and predict future performance. This can help decision makers understand the problems and challenges that the system may face in advance and formulate corresponding countermeasures.

[0083] (6) Abnormal environment testing: Define a series of simulation scenarios based on the actual operating environment of the system and the abnormal situations that may be encountered. These scenarios cover common error types, failure modes, and extreme conditions. Build a test environment similar to the production environment, including hardware, software, network, and data. Ensure that the test environment can truly reflect the actual situation of the production environment.

[0084] III. System Optimization:

[0085] (1) Configuration parameter adjustment: According to the simulation results and performance evaluation, the configuration parameters of the OHT system are adjusted, for example, increasing or decreasing the number of OHT vehicles, optimizing the path planning algorithm, adjusting the workstation layout, etc.

[0086] (2) Machine learning model training: Machine learning methods are used to train predictive models to predict system performance under different configurations. These models can be built based on historical data and simulation results, and can be continuously updated and optimized as new data is added.

[0087] (3) Iterative optimization: The optimized system configuration is re-entered into the simulation engine for verification, and data collection and analysis continue. Through the iterative optimization process, the optimal system configuration and performance are gradually approached.

[0088] (4) Implementation and optimization feedback: Apply the optimized solution to the actual factory and continuously collect operating data to verify the optimization effect. At the same time, feedback and adjustments are made to the model based on the actual operating conditions to ensure that the system always maintains the best state.

[0089] 4. Result verification and application

[0090] Field testing: Deploy the optimized OHT system configuration in a real production environment and monitor its performance. Record actual production data and compare it with simulation results.

[0091] Continuous optimization: Based on the results of field tests, further adjust the model parameters in the virtual environment. Update the model regularly to reflect the latest changes in production conditions.

[0092] Embodiment 2: This embodiment differs from Embodiment 1 in that:

[0093] 1. Data collection and preprocessing

[0094] First, the project team needs to have an in-depth understanding of the production process, material handling requirements, and specific functions and performance indicators of the OHT system of the semiconductor factory. This includes collecting information such as the factory layout, equipment specification sheets, process flow charts, and existing control system documents. At the same time, communicate with the factory's process engineers, equipment maintenance personnel, and production line operators to clarify the specific goals and requirements of the modeling, such as improving transportation efficiency, reducing failure rates, and optimizing material flow paths.

[0095] 2. Virtual environment modeling

[0096] 3D modeling of solid objects

[0097] (1-1) Aerial track modeling: Based on the factory layout, the aerial track of the OHT system is accurately drawn using 3D modeling software. This includes the shape, size, bending radius, support structure, etc. of the track. At the same time, necessary materials and textures are added to the track, such as metal texture, anti-rust coating, etc., to enhance the realism of the virtual environment.

[0098] (1-2) Modeling of the transport trolley: Create a three-dimensional model of the transport trolley according to the equipment specification table. The model should include key components such as the trolley's body structure, drive system, suspension device, and gripping mechanism. At the same time, consider the changes in the trolley's posture under different working conditions (such as driving, accelerating, decelerating, stopping, grabbing materials, etc.) and reflect them in the model.

[0099] (1-3) Modeling of other equipment: Perform 3D modeling of other related equipment in the semiconductor factory (such as transmission equipment, storage equipment, purification equipment, etc.). Ensure that these models are coordinated with the aerial tracks and transport carts in terms of size, proportion, and position.

[0100] Virtual space construction and physics engine settings

[0101] (2-1) Create a virtual scene: Use a 3D simulation platform (such as Unity) to create a new virtual scene. Divide the virtual scene into different areas (such as processing area, storage area, buffer area, etc.) according to the actual layout and size ratio of the semiconductor factory.

[0102] (2-2) Import digital objects: Import the previously created 3D model and control program image file into the virtual scene. Adjust the position and angle of the model according to the actual layout to ensure that the layout in the virtual scene is consistent with the actual factory.

[0103] (2-3) Set up the physics engine: Configure the physics engine parameters (such as gravity, friction, collision detection, etc.) in the virtual scene. Ensure that the physics engine can accurately simulate the physical behavior of the OHT system in actual operation.

[0104] 3. Simulation and optimization

[0105] (1) Model driver writing and loading

[0106] Writing model drivers: Writing model drivers for movable objects in virtual space (such as transport carts). These drivers can receive instructions from the virtual control system (such as start, stop, accelerate, decelerate, turn, etc.) and control the objects to move accordingly in the virtual space.

[0107] Loading driver: Associate the written model driver with the corresponding digital object and load it into the virtual control system. Ensure that the driver can correctly interpret the control instructions and drive the object to move as expected.

[0108] (2) Virtual debugging and optimization

[0109] Start simulation operation: Start the simulation operation of the OHT system in a virtual environment. Simulate different production scenarios and operating states (such as full load operation, no-load operation, emergency braking, etc.) to observe the operation and performance of the system.

[0110] Performance evaluation: record key data and parameters (such as transportation time, energy consumption, failure rate, etc.) and compare and analyze them with expected goals. Evaluate whether the system performance meets the requirements and whether there is room for optimization.

[0111] Problem identification and optimization: Identify existing problems or deficiencies in the system (such as path conflicts, transportation bottlenecks, etc.) during the simulation process. Propose specific optimization solutions for these problems (such as adjusting path planning, optimizing control algorithms, etc.), and iterate and verify them in a virtual environment until satisfactory results are achieved.

[0112] 4. Output report and practical application

[0113] Output report: Write a detailed simulation report summarizing the modeling process, simulation results, performance evaluation, and optimization solutions.

[0114] Practical application: Apply the optimized solution to actual production to improve the operating efficiency and reliability of the OHT system in semiconductor factories.

[0115] It can be seen from the above content that the present invention has the following advantages:

[0116] In terms of path planning and optimization, the adaptive path planning method can dynamically adjust the path of the OHT vehicle according to real-time production data, thereby effectively reducing waiting time and path conflicts;

[0117] In terms of system performance evaluation and optimization, it can simulate the operation of the OHT system under various production conditions, thereby helping engineers evaluate the system performance under different configurations and optimize accordingly;

[0118] In terms of material flow analysis and prediction, by combining historical data with real-time data and using machine learning technology, it is possible to predict future material flow trends and adjust the OHT system scheduling strategy in advance;

[0119] In terms of resource allocation and scheduling: it can dynamically adjust resource allocation according to production needs to improve the overall performance of the system;

[0120] In terms of abnormal situation handling, it is possible to rehearse the system response under abnormal circumstances and formulate corresponding emergency strategies to improve the robustness and flexibility of the system.

[0121] By building a highly realistic virtual space, the OHT system and its related equipment can be simulated, thereby improving the debugging efficiency of the semiconductor production process and reducing the risk of equipment damage that may occur during physical debugging.

[0122] The above is an exemplary description of the present invention in combination with the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as such non-substantial improvements are made by adopting the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. A method for modeling a virtual environment of an OHT system in a semiconductor factory, characterized by: The steps include: S1. Data collection and preprocessing: Collect on-site facility data of semiconductor factory production lines, operation-related data of OHT vehicles, and product information data, and preprocess them separately to obtain basic data; S2, virtual environment modeling: constructing a virtual environment model according to the basic data in step S1; S3, simulation and optimization: based on the virtual environment model constructed in step S2, simulate and run, and perform performance evaluation; then optimize according to the simulation results and performance evaluation; S4. Result verification and application.

2. A semiconductor factory OHT system virtual environment modeling method according to claim 1, characterized in that: In step S1, the field facility data of the semiconductor factory production line includes the position coordinates, track distances, locations of material storage areas, and connection relationships between various workstations on the production line; and / or, the operation-related data of the OHT vehicle includes specification parameters of the OHT vehicle and load conditions, driving routes, and stop times of the OHT vehicle in different time periods; And / or, the product information data includes product model, production batch, required material type and quantity.

3. A semiconductor factory OHT system virtual environment modeling method according to claim 1, characterized in that: In step S2, constructing a virtual environment model includes physical environment modeling, OHT system model construction, workstation and material storage area model, and control logic setting.

4. A semiconductor factory OHT system virtual environment modeling method according to claim 1, characterized in that: In step S2, constructing a virtual environment model includes three-dimensional modeling of physical objects, virtual space construction and physical engine settings. The three-dimensional modeling of physical objects includes aerial track modeling, transport vehicle modeling, and modeling of other related equipment in a semiconductor factory; the virtual space construction includes creating a virtual scene and importing digital objects.

5. The method for modeling a virtual environment of an OHT system in a semiconductor factory according to claim 1, characterized in that: In step S3, the simulation operation includes intelligent scheduling, path planning, key indicator recording, parameter change experiment, and efficient parameter exploration; and / or, the performance evaluation includes setting evaluation indicators, statistical analysis, multi-indicator comprehensive evaluation, sensitivity analysis, time series prediction and abnormal environment testing; And / or, the optimization includes configuration parameter adjustment, machine learning model training, iterative optimization, implementation and optimization feedback.

6. A semiconductor factory OHT system virtual environment modeling method according to claim 5, characterized in that: The intelligent scheduling is as follows: when the entire map is designed in modules, the vehicle is first selected according to the starting point of the task; secondly, an idle vehicle in the area where the task is located is selected; thirdly, vehicles in adjacent areas are selected; then, vehicles whose task destination is in the area and are non-idle vehicles in the map are selected; finally, an idle vehicle is selected in the entire map.

7. A semiconductor factory OHT system virtual environment modeling method according to claim 5, characterized in that: The path planning is: using Dijkstra to perform path planning, starting from the starting point, and gradually finding and determining the shortest path to other vertices in the graph.

8. The method for modeling a virtual environment of an OHT system in a semiconductor factory according to claim 5, characterized in that: The purpose of the sensitivity analysis is to determine the degree of influence of key indicators and their combinations on system performance; And / or, the time series prediction is: using a clustering algorithm to process time series data to predict future performance.

9. A semiconductor factory OHT system virtual environment modeling method according to claim 5, characterized in that: The abnormal environment test is to define a series of simulation scenarios according to the actual operating environment of the system and the abnormal situations that may be encountered; These scenarios cover common error types, failure modes, and extreme conditions; build a test environment similar to the production environment to ensure that the test environment can truly reflect the actual situation of the production environment.

10. The semiconductor factory OHT system virtual environment modeling method according to claim 1, characterized in that: Step S4 includes: Field testing: deploying the optimized OHT system configuration in the actual production environment and monitoring its performance, recording actual production data and comparing it with simulation results; Continuous optimization: Based on the results of field tests, further adjust the model parameters in the virtual environment; regularly update the model to reflect the latest changes in production conditions.

Images