TWM618950U - Product process abnormality analysis platform using artificial intelligence - Google Patents

Abstract

This application proposes a product process anomaly analysis platform using artificial intelligence. The product process anomaly analysis platform includes a huge data storage device and a product process anomaly analysis system. The huge data storage device is used to store multiple manufacturing primitives. Data, and the manufacturing raw data comes from multiple manufacturing equipment. The product process anomaly analysis system is used to generate multiple graphical analysis results corresponding to multiple manufacturing equipment and a graphical interface, and selectively use the graphical interface At least one of the multiple graphical analysis results is included. This improves the convenience and efficiency of analyzing huge amounts of data, and quickly eliminates process abnormalities.

Description

Product process anomaly analysis platform using artificial intelligence

This application relates to a data analysis platform, especially a product process anomaly analysis platform using artificial intelligence.

Generally speaking, the production of semiconductor components requires more than a thousand manufacturing and testing procedures, and the yield of the final product is maintained by sophisticated manufacturing equipment and process design. Therefore, in order to effectively control the manufacturing process of semiconductor components, the manufacturing process All information and data in the process must be collected and analyzed in real time.

With the advancement of the manufacturing process, multiple manufacturing equipment on the production line will generate tens of thousands of real-time monitoring data, nearly 10,000 on-line sampling and inspection measurements (metrology), and hundreds of electrical tests measured at different positions on semiconductor components At the same time, coupled with various integrated circuit production modes, the production line can generate more than billions of huge amounts of data in one month.

Therefore, in order to analyze the huge amount of data generated on the production line, commercial statistical analysis methods are often used in order to quickly find out the cause of process abnormalities. However, the statistical correlation analysis method can only present possible correlation problems, and cannot accurately locate the cause of the abnormal process. Moreover, because the manufacturing equipment generates a huge amount of data at any time, the statistical correlation analysis method cannot complete the corresponding analysis in real time, and the user needs additional It takes time to find related problems in the results of statistical correlation analysis, and analyzes the causes of abnormalities in its manufacturing equipment on its own. As a result, the current statistical analysis methods cannot quickly eliminate the abnormalities.

Based on the above-mentioned problems in the prior art, it is indeed necessary to propose a better solution.

In view of the above-mentioned shortcomings of the prior art, the main purpose of this application is to provide a product process anomaly analysis platform using artificial intelligence, which uses artificial intelligence to analyze the process parameters and measurement parameters of the manufacturing equipment on the production line, and generate multiple graphics Analyze the results, and optionally provide one of multiple graphical analysis results for viewing. Users can quickly obtain the prediction results corresponding to different manufacturing equipment through the graphical analysis results, thereby improving the convenience of analyzing huge amounts of data Performance and efficiency to quickly eliminate process abnormalities and optimize the overall product manufacturing process.

In order to achieve the above objective, this application proposes a product process anomaly analysis platform using artificial intelligence, which includes a huge data storage device and a product process anomaly analysis system, wherein the huge data storage device is used to store multiple manufacturing primitives. The manufacturing raw data comes from multiple manufacturing equipment and includes multiple process parameters and multiple measurement data. The product process anomaly analysis system is used to generate graphical analysis results corresponding to multiple manufacturing equipment and generate a A graphical interface, and the graphical interface includes the graphical analysis result.

Further, the product process anomaly analysis system further includes a data analysis engine and an interface generation module, and the data analysis engine includes: a pre-processor, an artificial intelligence analyzer, and a post-processor, wherein the pre-processing The device is used to obtain at least a part of multiple manufacturing raw data from the huge data storage device, and convert at least a portion of the multiple manufacturing raw data into multiple processed data, and the artificial intelligence analyzer is used to receive and analyze the multiple After processing the data, corresponding analysis result data is generated. The analysis result data is a process abnormal analysis result. The post-processor is used to receive the analysis result data and generate the graphical analysis result. The interface generating module is used for receiving the graphical analysis result and used for generating the graphical interface, and the graphical interface includes the graphical analysis result.

With the above structure, the product process anomaly analysis platform using artificial intelligence can use the huge amount of data it collects to provide the required analysis results simply and clearly with graphical analysis results, and users do not need to analyze the results by themselves. Or search, you can quickly obtain the required process-related information. In addition, by providing the graphical interface of the operation information of the multiple manufacturing equipment, the user can quickly grasp the overall status of the production line, thereby improving the analysis of huge amounts of data Convenience and efficiency.

Please refer to Figure 1. Figure 1 is a schematic diagram of the implementation environment of the product process anomaly analysis platform using artificial intelligence in this application. It includes a product process anomaly analysis platform 100, a production line system 200, and an input and output module 300. The product The process abnormality analysis platform 100 is in communication connection with the production line system 200 and the input/output module 300. The production line system 200 is used to monitor multiple manufacturing equipment on the production line, and collect the manufacturing raw data related to the production line and the multiple manufacturing equipment in real time, and the manufacturing raw data has a time series relationship before and after, so as to control the production line in real time. All states of the multiple manufacturing equipment on the production line, where the multiple manufacturing raw data include multiple process parameters and multiple measurement data corresponding to different manufacturing processes. The product process anomaly analysis platform 100 is used to receive the multiple manufacturing raw data from the production line system 200 and analyze the multiple manufacturing raw data to generate graphical analysis results and a graphical interface. The graphical interface includes the graphical analysis result, in other words, the graphical interface may not include all the graphical analysis results at the same time. The input and output module 300 is used for receiving and displaying the graphical interface, and for receiving user selection information input by the user.

In this way, when the product is abnormal, a user can quickly confirm the abnormal state or abnormal state of multiple manufacturing equipment and multiple processes on the production line through the graphical analysis result of the graphical interface displayed by the input and output module 300 Relevant information can quickly determine the cause of the abnormal state without finding the required information among the numerous analysis results to analyze the possible cause of the abnormal state, thereby achieving the purpose of improving the convenience and efficiency of the analysis of huge amounts of data.

In one embodiment, the product process anomaly analysis platform 100 is a yield quality engineering analysis decision system.

In one embodiment, the production line system 200 is, for example, a production execution system, an equipment engineering system, or a precision manufacturing engineering system, and the application is not limited thereto.

In one embodiment, the input and output module 300 is, for example, a touch display screen or a display with an input device, such as a keyboard and a mouse, and the application is not limited thereto.

In one embodiment, the production line is a semiconductor device production line, and this application is not limited thereto.

Please refer to FIG. 2, the product process abnormality analysis platform 100 at least includes a central control system 110, a massive data storage device 130, a product process abnormality analysis system 150, a data input interface 170, and an input/output interface 190. Among them, The central control system 110 is electrically connected to the massive data storage device 130, the product process abnormality analysis system 150, the data input interface 170, and the input/output interface 190.

The central control system 110 is used to manage the product process anomaly analysis system 150, and monitor the data exchange among the huge data storage device 130, the product process anomaly analysis system 150, the data input interface 170, and the input/output interface 190 , And dispatch the overall computing, storage, and network resources of the product process anomaly analysis platform 100, so as to improve the resource utilization efficiency and operation efficiency of the product process anomaly analysis platform 100.

In one embodiment, the central control system 110 is, for example, a Composer module, and the application is not limited thereto.

The data input interface 170 is used to communicate with the production line system 200 to receive the multiple pieces of manufacturing raw data.

In one embodiment, the data input interface 170 can be a serial data (RS232) communication interface, a universal serial bus (USB) standard connection port, and this application is not limited thereto.

The massive data storage device 130 is used to store multiple pieces of manufacturing raw data and their corresponding time. The manufacturing raw data are provided by the production line system 200 and come from multiple manufacturing equipment, where each manufacturing equipment is The production line generates corresponding manufacturing raw data over time and process.

In one embodiment, the massive data storage device 130 can be implemented by multiple hard disk storage devices, and the application is not limited thereto.

The product process anomaly analysis system 150 is used for reading at least a part of the multiple pieces of manufacturing raw data from the huge data storage device 130, and analyzing and predicting at least a portion of the multiple pieces of manufacturing raw data. To generate the plurality of graphical analysis results and the graphical interface, wherein the graphical interface may selectively include at least one of the plurality of graphical analysis results, in other words, the graphical interface may not include all of the graphical analysis results at the same time Graphical analysis results.

The I/O interface 190 is in communication with the I/O module 300 to provide the graphical interface to the I/O module 300 and receive user selection information input by the user.

In one embodiment, the I/O interface 190 includes ports that comply with Universal Serial Bus (USB), High-Definition Multimedia Interface (HDMI), and Video Graphics Array (VGA) specifications, and the application is not limited thereto.

In one embodiment, the product process anomaly analysis platform 100 can be implemented by a Blade Server, and this application is not limited thereto.

In one embodiment, the multiple pieces of manufacturing raw data may include multiple process parameters and multiple measurement data. The multiple process parameters are, for example, process steps, process recipe guidance parameters, target parameters, reference calibration, field fine calibration, crystal Circular grid correction, focus correction, exposure dose correction, etching time, deposition, air flow rate and sputtering voltage, etc. The multiple measurement data include overlay measurement, critical dimension measurement, alignment measurement, leveling measurement, exposure dose, Measurement data such as etching measurement or deposition measurement, but the plurality of process parameters and the plurality of measurement data vary depending on the process, and this application is not limited thereto.

Further, the product process abnormality analysis system 150 further includes a data analysis engine 151 and an interface generation module 153, as shown in FIG. 3, and the data analysis engine 151 is in communication connection with the interface generation module 153. The data analysis engine 151 is used to perform analysis and prediction based on the received manufacturing raw data to generate corresponding graphical analysis results, for example, a graphical analysis result corresponding to a related manufacturing process of a product. The interface generating module 153 is used for receiving the plurality of graphical analysis results, and for correspondingly generating the graphical interface, and selectively making the graphical interface only include one of the plurality of graphical analysis results, For example, only include the graphical analysis results of abnormal conditions, or only include the graphical analysis results corresponding to the information selected by the user.

In one embodiment, the abnormal state may be a state in which the process step is judged to be abnormal based on the analysis and prediction result of the data analysis engine, and this application is not limited thereto. For example, according to the graphical analysis result, the analysis and prediction result shows that a process parameter of a process step is lower than a preset value. The interface generation module 153 determines that it is an abnormal state and makes the graphical interface include the process parameter Numerical graphical analysis results, whereby users can quickly identify abnormal process parameters to eliminate process abnormalities.

In one embodiment, the user selection information is the information that the user wants to actively view, and it includes a graphical analysis result generated by the production process abnormality analysis platform 100 from the manufacturing raw data obtained by the production line system 200. For example, the user selection information is the selection information corresponding to the process steps of the manufacturing equipment on the production line, and the interface generation module 153 makes the graphical interface include the graphical analysis results corresponding to the process steps according to the user selection information To provide users with a view.

In one embodiment, the data analysis engine 151 can be implemented by a single board computer executing a corresponding analysis preset program, and this application is not limited thereto.

In one embodiment, the interface generation module 153 can be implemented by a single-board computer executing a corresponding interface generation program, and this application is not limited thereto.

In one embodiment, the aforementioned program is used to store a computer-readable memory device of the single-board computer, such as a hard disk device, and this application is not limited thereto.

Please refer to FIG. 4, which is a platform architecture diagram of the data analysis engine 151. The data analysis engine 151 includes at least a pre-processor 1511, an artificial intelligence analyzer 1513, and a post-processor 1515.

The pre-processor 1511 is used for reading the required manufacturing raw data from the massive data storage device 130 according to the analysis and prediction to be performed by the artificial intelligence analyzer 1513. For example, the artificial intelligence analyzer 1513 can be used to analyze and predict product yields based on manufacturing parameters and measurement data. Therefore, in this embodiment, the pre-processor 1511 is used to send data to the massive data storage device 130 Read the required manufacturing raw materials for analysis and prediction.

Further, the pre-processor 1511 is used for pre-processing the received manufacturing raw data, that is, screening, cleaning (Clean), unified format (Format), missing value (Missing value) processing, and transforming (Transform) the manufacturing raw data. ) And so on to remove abnormal data and ensure data quality and high prediction accuracy to generate processed data to be provided to the artificial intelligence analyzer 1512.

In one embodiment, the pre-processor 1511 can be implemented in a structured query language (SQL), a data warehouse (Hive), an R language, or a Python language, and this application is not limited thereto. .

The artificial intelligence analyzer 1513 is used to receive the multiple processed data, and use its analysis and prediction model to perform second-level operations to calculate and predict the corresponding prediction value (Conjecture value), confidence index (Reliance Index), and overall similarity of process parameters Process abnormal analysis results such as the Global Similarity Index (Global Similarity Index).

In this embodiment, the analysis and prediction model is trained on the multiple pieces of manufacturing raw data such as manufacturing parameters and measurement data generated by the manufacturing equipment along the manufacturing process on the production line to build the model.

In this embodiment, the artificial intelligence analyzer 1513 is used to analyze and predict the manufacturing parameters of the manufacturing equipment and the predicted value of the measurement data, and the application is not limited thereto.

Further, the confidence index indicates the credibility of the accuracy of the predicted value. The purpose of the confidence index is to calculate a confidence value between 0 and 1 by analyzing the manufacturing parameters of the manufacturing equipment to determine whether the analysis result can be trusted. And use the maximum tolerable error upper limit (EL) to correspond to the confidence index to obtain the confidence index threshold (RIT). When the confidence indicator value is greater than the confidence indicator threshold, it means that the analysis result can be trusted; on the contrary, when the confidence indicator value is lower than the confidence indicator threshold, a warning is issued. Therefore, users as equipment engineers can use this to inspect manufacturing equipment, or users as process engineers can adjust parameters to confirm whether the process is stable.

The artificial intelligence analyzer 1513 is also used to calculate the similarity index. The main purpose of the similarity index is to compare the similarity of the manufacturing parameters between the prediction segment and the modeling segment. The similarity index includes two parts, one is the overall similarity index of the process parameters, and the second is the individual process parameter similarity index (ISI). The overall similarity index of the process parameters is the degree of similarity between the process parameters in the prediction section and all the parameters in the modeling section. The individual similarity index of the process parameter is the standardized absolute similarity between any process parameter in the prediction section and all samples of the parameter in the modeling section. The similarity index is also used as an auxiliary confidence index for judgment. Therefore, a user as a process engineer can determine whether to perform parameter adjustment based on the analysis and prediction result of the similarity index to confirm whether the process is stable.

For example, if the confidence index of the prediction point is higher than the confidence index threshold, and the overall similarity index of the process parameter at the prediction point is higher than the overall similarity index threshold of the process parameter, check the individual similarity index display of the process parameter Whether the process parameters are abnormal.

For example, if the confidence index of the prediction point is lower than the threshold value of the confidence index, and the overall similarity index of the process parameters of the prediction point is lower than the overall similarity index threshold of the process parameters, the predicted value may be inaccurate, but due to the prediction The overall similarity index of the process parameters of the point is low, indicating that the new component (such as wafer) has a high similarity with the modeling parameter data. At this time, there may be a situation where the prediction value of the analysis and prediction model is not good.

For example, if the confidence index of the prediction point is lower than the confidence index threshold, and the overall similarity index of the process parameters of the prediction point is higher than the overall similarity index threshold of the process parameters, it means that the prediction value of the analysis and prediction model is not good, and Since the overall similarity index value of the process parameters of the prediction point is high, it means that the similarity of the new component and the modeled manufacturing parameters is low, so the predicted value can be regarded as inaccurate.

For example, when the overall similarity index of the process parameters at each prediction point is higher than the threshold value of the overall similarity index of the process parameters, it indicates that the process parameters may be abnormal.

In addition, because all the manufacturing raw data have a time series relationship before and after, the artificial intelligence analyzer 1513 can accurately and effectively locate the problem. For example, when the time points of the abnormality in the analysis result data are denser, the problem can be located.

In one embodiment, the artificial intelligence analyzer 1513 can be implemented by Simple Recurrent Neural Networks (SRNN) and Multiple Regression Analysis (MRA), and this application is not limited thereto .

In one embodiment, the manufacturing raw data received by the production line system 200 in real time is further used to update and adjust or retrain the analysis and prediction model, so that the artificial intelligence analyzer 1513 can accurately grasp the operating status of the production line to improve The artificial intelligence analyzer 1513 analyzes and predicts accuracy.

The post-processor 1515 is used to receive the analysis result of the artificial intelligence analyzer 1513, and perform data graphing accordingly to generate the corresponding graphical analysis result, for example, graphs realized by charts, graphs, and distribution graphs The results of chemical analysis, for example, show the yield rate of multiple manufacturing equipment on the production line as a graph. In this way, the user can quickly understand the state or trend of the analysis result of the artificial intelligence analyzer 1513 through the graphical analysis result.

In an embodiment, the post-processor 1515 can be implemented by a data statistical analysis program, and this application is not limited thereto.

According to the above-mentioned embodiments and application methods of the present application, the operation method of a product process abnormality analysis platform can be summarized, as shown in Fig. 5, the steps include:

Step S100: Collect multiple pieces of manufacturing original data. A product process anomaly analysis platform 100 is used to receive multiple pieces of manufacturing raw data sent by a production line system 200 and store them in the massive data storage device 130 of the product process anomaly analysis platform 100.

Step S200: perform data analysis. The product process anomaly analysis system 150 of the product process anomaly analysis platform 100 generates a plurality of graphical analysis results according to at least a part of the multiple pieces of manufacturing raw data.

Step S300: Generate a graphical interface, the graphical interface including at least one graphical analysis result. The product process anomaly analysis system 150 is used to generate the graphical interface, and selectively make the graphical interface include at least one of the graphical analysis results, and the graphical interface is used to display the information of the multiple manufacturing equipment Operational information.

Further, step 200 further includes the following steps:

Step S210: Obtain the required manufacturing original data. The product process anomaly analysis system 150 includes a data analysis engine 151. The preprocessor 1511 of the data analysis engine 151 is used to predict what is needed for reading from the massive data storage device 130 based on the analysis to be performed by an artificial intelligence analyzer 1513 The manufacturing source material.

Step S230: Perform pre-processing on multiple pieces of manufacturing original data. The pre-processor 1511 further performs pre-processing on the received multiple pieces of manufacturing raw data, and generates corresponding processed data.

Step S250: Perform data analysis on the processed data. The artificial intelligence analyzer 1513 performs analysis and prediction on the processed data and generates corresponding analysis results.

Step S270: Perform post-processing on the analysis result. The post-processor 1515 of the data analysis engine 151 performs post-processing on the analysis result to generate a corresponding graphical analysis result.

In summary, the embodiment of the product process anomaly analysis platform using artificial intelligence in this application can use the huge amount of data it collects to use artificial intelligence to provide the required analysis and prediction results in a simple and clear graphical analysis result. Users do not need to analyze or search among the many analysis results to quickly obtain the required process-related information. In addition, by providing the graphical interface of the operation information of the multiple manufacturing equipment, the user can quickly grasp the overall production line Status, thereby improving the convenience and efficiency of massive data analysis.

100: Product process abnormal analysis platform 110: Central Control System 130: Massive data storage device 150: Product Process Abnormal Analysis System 151: Data Analysis Engine 1511: preprocessor 1513: Artificial Intelligence Analyzer 1515: post processor 153: Interface generation module 170: data input interface 190: I/O interface 200: Production line system 300: Input and output modules S100, S200, S210, S230, S250, S270, S300: steps

FIG. 1 is a schematic diagram of the implementation environment of an embodiment of a product process anomaly analysis platform using artificial intelligence according to this application; FIG. 2 is a schematic diagram of the platform architecture of an embodiment of a product process anomaly analysis platform using artificial intelligence according to the present application; FIG. 3 is a schematic diagram of the platform architecture of an embodiment of the product process anomaly analysis system using artificial intelligence according to the present application; Figure 4 is a platform architecture diagram of an embodiment of the data analysis engine according to the present application; FIG. 5 is a schematic diagram of a step flow diagram of an embodiment of an operating method according to the present application; and FIG. 6 is a schematic diagram of another step flow diagram of the embodiment of the operating method according to the present application.

150: Product Process Abnormal Analysis System

151: Data Analysis Engine

153: Interface generation module

Claims (8)

A product process anomaly analysis platform using artificial intelligence, which includes: A huge data storage device for storing multiple pieces of manufacturing raw data, the manufacturing raw data including multiple process parameters and multiple measurement data, and the manufacturing raw data comes from multiple manufacturing equipment; and A product process anomaly analysis system, which includes: A data analysis engine, which includes: A pre-processor that obtains at least a part of the multiple manufacturing raw data from the huge data storage device, and converts at least a part of the multiple manufacturing raw data into multiple processed data; An artificial intelligence analyzer for receiving and analyzing the multiple processed data, and generating corresponding analysis result data, the analysis result data being a process abnormality analysis result; and A post-processor receives the analysis result data and generates graphical analysis results corresponding to multiple process steps; and An interface generation module receives the graphical analysis result and is used to generate a graphical interface, and the graphical interface includes the graphical analysis result. The product process abnormality analysis platform according to claim 1, wherein the product process abnormality analysis platform further includes a central control system, which is in communication connection with the product process abnormality analysis system and the huge data storage device. The product process anomaly analysis platform according to claim 1, wherein the graphical interface only includes the graphical analysis result of the abnormal state. The product process anomaly analysis platform according to claim 1, wherein the preprocessor is a structured query language, a data warehouse, an R language or a Python language. The product process abnormality analysis platform according to claim 1, wherein the artificial intelligence analyzer includes a simple loop-like neural network and a multiple regression analysis. The product process abnormality analysis platform according to claim 1, wherein the data analysis engine is a single board computer. The product process anomaly analysis platform according to claim 1, wherein the product process anomaly analysis platform is a blade server. The product process abnormality analysis platform according to claim 1, wherein the multiple manufacturing equipment is used in a semiconductor device production line.

Images