Semiconductor production line anomaly detection methods, devices, storage media, and equipment

Abstract

A method, apparatus, storage medium, and device for anomaly detection in a semiconductor production line are disclosed. The method includes: acquiring information about an anomaly object; reading corresponding raw material information from a preset raw material database based on the anomaly object information; and analyzing the read raw material information to determine the cause of the semiconductor production line anomaly. This approach can improve the speed of anomaly detection from the raw material perspective.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and specifically to a method, apparatus, storage medium, and equipment for detecting anomalies in a semiconductor production line. Background Technology

[0002] The manufacturing process of semiconductor devices involves a wide variety of equipment, complex processes, and high-specification requirements for raw materials, which makes semiconductor production very difficult.

[0003] In actual manufacturing processes, production line anomalies frequently occur. When such anomalies occur, it is necessary to investigate the cause from a quality perspective. Typically, the root cause of the production line anomaly needs to be investigated from six dimensions: human factors, equipment, raw materials, parameters, environment, and testing.

[0004] However, the current rate of anomaly detection at the raw material level is low. Summary of the Invention

[0005] The problem this invention aims to solve is: how to improve the rate of anomaly detection from the perspective of raw materials.

[0006] To address the above problems, embodiments of the present invention provide a method for detecting anomalies in a semiconductor production line, the method comprising:

[0007] Obtain information about the exception object;

[0008] Based on the abnormal object information, the corresponding raw material information is read from the preset raw material database;

[0009] The analysis is based on the raw material information obtained to determine the cause of the semiconductor production line abnormality.

[0010] In one possible embodiment, the abnormal object is at least one of the following: abnormal process, abnormal process step, abnormal process equipment, and abnormal plant equipment.

[0011] In one possible embodiment, the preset raw material database includes at least one of the following raw material information:

[0012] Warehouse entry information for each raw material;

[0013] Outbound information for each raw material;

[0014] Usage records for each raw material;

[0015] Outgoing inspection information for each raw material;

[0016] Change records for each raw material;

[0017] Abnormal raw material information.

[0018] In one possible embodiment, the raw material warehousing information includes at least one of the following: the name, quantity, batch number, and supplier information of the raw material.

[0019] In one possible embodiment, the raw material outbound information includes at least one of the following: the name, quantity, and batch number of the raw material.

[0020] In one possible embodiment, the raw material usage record information includes at least one of the following: raw material start-up time information and raw material end-up time information.

[0021] In one possible embodiment, the shipment inspection information of the raw materials includes: inspection results of various performance parameters of the raw materials.

[0022] In one possible embodiment, the raw material change record information includes at least one of the following: raw material model change information and manufacturer change information.

[0023] In one possible embodiment, the abnormal raw material information includes: identification information of expired raw materials, identification information of raw materials with abnormal shipment inspection, and identification information of raw materials with abnormal storage environment.

[0024] In one possible embodiment, the preset raw material database further includes: correspondence information between each abnormal object and each raw material information.

[0025] In one possible embodiment, the analysis based on the read raw material information includes:

[0026] A commonality analysis was performed on the measured values ​​of the raw material information read at different times on the wafers, and a commonality analysis chart was generated.

[0027] In one possible embodiment, the analysis based on the read raw material information further includes:

[0028] The measured values ​​of wafers using abnormal raw materials or wafers using modified raw materials are identified in the commonality analysis chart.

[0029] This invention also provides a semiconductor production line anomaly detection device, the device comprising:

[0030] The acquisition unit is suitable for acquiring information about abnormal objects.

[0031] The reading unit is adapted to read corresponding raw material information from a preset raw material database based on the abnormal object information;

[0032] The analysis unit is suitable for performing analysis based on the read raw material information to determine the cause of abnormalities in the semiconductor production line.

[0033] This invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of any of the methods described above.

[0034] This invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the steps of any of the methods described above when running the computer program.

[0035] Compared with the prior art, the technical solution of the embodiments of the present invention has the following advantages:

[0036] By applying the solution of this invention, a raw material database is pre-set. When abnormal object information is obtained, the corresponding raw material information can be read from the pre-set raw material database. Based on the read raw material information, analysis can be performed to determine the cause of the semiconductor production line abnormality. Compared with the method of manually determining the raw material information corresponding to the abnormal object information and analyzing it, the computer can quickly lock the corresponding raw material information and perform analysis automatically. This can quickly determine whether the raw material caused the semiconductor production line abnormality, effectively improving the speed of abnormality detection from the raw material dimension. Attached Figure Description

[0037] Figure 1 This is a schematic diagram illustrating the factors that can lead to abnormalities in a semiconductor production line.

[0038] Figure 2 This is a flowchart of a semiconductor production line anomaly detection method according to an embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of the architecture of a raw material database in an embodiment of the present invention;

[0040] Figure 4 This is a schematic diagram of a commonality analysis chart in an embodiment of the present invention;

[0041] Figure 5 This is a schematic diagram of another commonality analysis chart in an embodiment of the present invention;

[0042] Figure 6 This is a schematic diagram of another commonality analysis chart in an embodiment of the present invention;

[0043] Figure 7 This is a schematic diagram of the structure of a semiconductor production line anomaly detection device according to an embodiment of the present invention. Detailed Implementation

[0044] Figure 1 This is a schematic diagram illustrating a factor contributing to abnormalities in a semiconductor production line. (Refer to...) Figure 1 When production line abnormalities occur, it is necessary to investigate the cause of the abnormality from a quality perspective. Generally, the root cause of the production line abnormality needs to be investigated from six dimensions: human factors (people), equipment (machines), raw materials (materials), parameters (methods), environment (environment), and testing (measurement).

[0045] Among these, the detection of equipment, parameters, testing, and environment are supported by mature online systems in major manufacturing companies, facilitating rapid detection and analysis when production line anomalies occur. Human factors, however, have a smaller impact in automated production companies and cannot be controlled by the system. Only the raw material dimension requires manual retrieval and analysis of relevant raw material information from multiple systems, resulting in low detection efficiency.

[0046] To address this problem, the present invention provides a method for detecting anomalies in semiconductor production lines. Using this method, when information about an anomaly is obtained, the corresponding raw material information is directly read from a preset raw material database and analyzed, thereby achieving rapid anomaly detection at the raw material level.

[0047] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0048] Reference Figure 2 This invention provides a method for detecting anomalies in a semiconductor production line, the method comprising the following steps:

[0049] Step 21: Obtain information about the abnormal object.

[0050] In practical implementation, when an anomaly occurs on a semiconductor production line, it may be due to an anomaly in a specific process, a specific step in the process, or a specific piece of equipment or facility (Fac) equipment. Therefore, in one embodiment of the present invention, the anomaly object includes at least one of the following: an abnormal process, an abnormal step in the process, an abnormal piece of equipment, or an abnormal facility equipment. The facility equipment refers to equipment that provides special gases.

[0051] Step 22: Based on the abnormal object information, read the corresponding raw material information from the preset raw material database.

[0052] In practical implementation, after identifying the abnormal object, the corresponding raw material information can be read from a preset raw material database. This preset raw material database includes each abnormal object, raw material information, and the correspondence between the abnormal objects and the raw material information. That is to say, as... Figure 3As shown, in the preset raw material database, each abnormal object is bound to the raw material information.

[0053] In practice, a pre-set raw material database can store information on various raw materials involved in the semiconductor production line. These raw materials may include: chemicals, photoresists, slurries, special gases, target materials, etc.

[0054] Specifically, the information about the raw materials may include at least one of the following:

[0055] Warehouse entry information for each raw material;

[0056] Outbound information for each raw material;

[0057] Usage records for each raw material;

[0058] Outgoing inspection information for each raw material;

[0059] Change records for each raw material;

[0060] Abnormal raw material information.

[0061] In specific implementation, the raw material warehousing information refers to the information recorded when the raw materials are put into storage, that is, the information recorded when the raw materials are stored in the warehouse. Specifically, it may include at least one of the following: the name, quantity, batch number, and supplier information of the raw materials.

[0062] In specific implementation, the raw material outbound information refers to the information recorded when the raw materials are released from the warehouse, that is, the information recorded when the raw materials stored in the warehouse are provided to the production line. Specifically, it may include at least one of the following: the name, quantity and batch number information of the raw materials.

[0063] In practical implementation, raw material usage record information refers to the time information when the raw materials are used, which may include at least one of the raw material's online and offline time information. This online and offline time information can be automatically collected using sensors, scanning devices, etc.

[0064] In specific implementation, the shipment inspection information of the raw materials includes: the test results of various performance parameters of the raw materials. This shipment inspection information can be obtained through the raw material shipment inspection report (CoA). The shipment inspection report is a report provided by the raw material supplier to the semiconductor manufacturer when the raw materials are supplied, detailing the testing of the raw material's performance parameters. The shipment inspection report contains the test results of various performance parameters of the raw materials; for example, the shipment inspection report for substrate wafers may include the measured values ​​of key parameters such as the diameter, thickness, warpage, and resistivity of the substrate wafer.

[0065] In practical implementation, the raw material change record information refers to the information recorded when a change occurs to the raw material, which may include at least one of the following: raw material model change information, manufacturer change information, production location change information, and other change information affecting the performance of the raw material. Specifically, the raw material model change information may include: change time, model information before the change, and model information after the change. The raw material manufacturer change information may include: change time, manufacturer identification information before the change, and manufacturer identification information after the change.

[0066] In practice, the abnormal raw material information refers to the identification information of raw materials that have experienced various abnormal situations. Specifically, this can include various abnormal situations and the corresponding identification information of the raw materials. Abnormal situations can include: expiration, abnormal shipment inspection, and abnormal storage environment. Expiration refers to raw materials exceeding their safe usage period. Abnormal shipment inspection refers to anomalies found in the raw material's shipment inspection report. Abnormal storage environment refers to abnormal storage conditions before the raw materials were stored in the warehouse, such as being stored under high temperature or high pressure for a period of time.

[0067] In one embodiment, the abnormal raw material information may include at least one of the following: identification information of expired raw materials, identification information of raw materials with abnormal shipment inspection, and identification information of raw materials with abnormal storage environment.

[0068] In specific implementation, the preset raw material database may include only one of the following: the warehousing information of each raw material, the outbound information of each raw material, the usage record information of each raw material, the shipment inspection information of each raw material, the change record information of each raw material, and the abnormal raw material information; it may also include two or more of these; or it may include all of the above information.

[0069] It is understandable that the more raw material information in the preset raw material database, the better it is for semiconductor production line anomaly detection, and the higher the efficiency of semiconductor production line anomaly detection.

[0070] In practice, the preset raw material database can be dynamically maintained. For example, the raw material information in the preset raw material database can be updated in real time. That is, once the raw material information changes, the raw material information in the preset raw material database can be updated at a certain period (such as once a week), thereby ensuring the accuracy of the raw material information in the preset raw material database.

[0071] Step 23: Analyze the raw material information obtained to determine the cause of the semiconductor production line abnormality.

[0072] In practice, various methods can be used to analyze the raw material information obtained, and there are no restrictions here, as long as the cause of the semiconductor production line abnormality can be determined.

[0073] In one embodiment, commonality analysis can be performed on the wafer measured values ​​of raw material information read at different times to form a commonality analysis chart, thereby providing a basis for the analysis.

[0074] Specifically, it can receive externally input analysis target information, classify the read raw material information based on the analysis target information, and then obtain the wafer measurement values ​​of various raw material information at different times, and form a common analysis chart. The common analysis chart can be displayed to the user, and the user can more intuitively determine whether various raw material information is abnormal.

[0075] In other words, the system can receive externally input analysis target information via an input device, i.e., specify the analysis target, and then graphically display the raw material information according to the specified analysis target. The analysis target can be any attribute of the raw material information in a preset raw material database, such as the name, batch number, model, or manufacturer of the raw material.

[0076] For example, taking the photolithography process as an anomaly, the corresponding raw material for the photolithography process is photoresist. After reading the photoresist information from a preset raw material database, the photoresist can be classified according to batch number, and the measured linewidth values ​​of the photolithographic patterns on wafers used at different times for each batch number can be obtained, forming a commonality analysis chart, such as... Figure 4 As shown.

[0077] Reference Figure 4 The horizontal axis represents time, and the vertical axis represents the measured linewidth of the photolithographic pattern on the wafer. 2KA028, 2KA030, and 2KA035 are photoresist batch numbers. Line 41 represents the target upper limit of the photolithographic pattern linewidth, line 42 represents the target lower limit of the photolithographic pattern linewidth, and line 43 represents the ideal value of the photolithographic pattern linewidth. From Figure 4It can be seen that the linewidth of the photolithographic pattern on the wafer with batch number 2KA035 basically exceeds the target upper limit of the photolithographic pattern linewidth. Therefore, it can be determined that the photoresist with batch number 2KA035 is one of the reasons for the production line abnormality.

[0078] In some embodiments, after forming a commonality analysis chart, the measured values ​​of wafers using abnormal raw materials or wafers using modified raw materials can be identified in the chart. Specifically, the measured values ​​of wafers using abnormal raw materials can be identified based on abnormal raw material information in a preset raw material database. For example, the measured values ​​of wafers using expired raw materials or wafers using modified raw materials can be identified in the commonality analysis chart. This allows for a more intuitive determination of the cause of raw material abnormalities, further improving the efficiency of anomaly detection on semiconductor production lines.

[0079] For example, refer to Figure 5 Assuming batch number 2KA035 is used as expired raw material in some wafers, the commonality analysis chart can show which wafers used expired photoresist in their measured values. The measured values ​​of wafers using expired photoresist are distributed within region 51.

[0080] For example, refer to Figure 6 Assuming the raw material batch numbers corresponding to the photolithography process are 2KA028, 2KA030, and WCA012, where the photoresist with batch numbers 2KA028 and 2KA030 is supplied by manufacturer A, and the photoresist with batch number WCA012 is supplied by manufacturer B, and the manufacturer of the photoresist with batch number WCA012 has changed compared to batch numbers 2KA028 and 2KA030, then the measured values ​​of the wafers with the changed manufacturer can be marked on the commonality analysis chart, as shown in area 61. From Figure 6 It can be seen that the measured values ​​of the wafers corresponding to the photoresist provided by Manufacturer B all exceeded the target upper limit, which is likely the main reason for the abnormality of the raw materials.

[0081] In practice, semiconductor production line anomalies can exist not only in the raw material dimension, but also in other dimensions, such as parameters (methods) and environment (environment). The causes of semiconductor production line anomalies can be obtained by combining other dimensions.

[0082] As can be seen from the above, the semiconductor production line anomaly detection method in this embodiment of the invention can improve the efficiency of semiconductor production line anomaly detection at the raw material level by setting up a raw material database, thereby avoiding the impact of production line anomalies on semiconductor production.

[0083] To enable those skilled in the art to better understand and implement the present invention, the apparatus, testing system, electronic device and computer-readable storage medium corresponding to the above method are described in detail below.

[0084] Reference Figure 7 This invention also provides a semiconductor production line anomaly detection device, which may include: an acquisition unit 71, a reading unit 72, and an analysis unit 73. Wherein:

[0085] The acquisition unit 71 is adapted to acquire abnormal object information;

[0086] The reading unit 72 is adapted to read corresponding raw material information from a preset raw material database based on the abnormal object information;

[0087] The analysis unit 73 is adapted to perform analysis based on the read raw material information to determine the cause of the semiconductor production line abnormality.

[0088] In some embodiments, the semiconductor production line anomaly detection device may further include: a raw material database, which can store information on various raw materials involved in the semiconductor production line. The reading unit 72 can be connected to the raw material database, thereby enabling it to read the raw material information from the database for anomaly analysis.

[0089] For details regarding the acquisition unit 71, the reading unit 72, and the analysis unit 73, please refer to the above description of steps 21 to 23, which will not be repeated here.

[0090] This invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of any of the above methods.

[0091] In specific implementations, the computer-readable storage medium may include ROM, RAM, disk, or optical disk, etc.

[0092] This invention also provides an electronic device, which includes a memory and a processor. The memory stores a computer program that can run on the processor, and when the processor runs the computer program, it performs the steps of any of the methods described above.

[0093] Regarding the modules / units included in the various devices and products described in the above embodiments, they can be software modules / units, hardware modules / units, or a combination of both. For example, for various devices and products applied to or integrated into a chip, all of their modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits; for various devices and products applied to or integrated into a chip module, all of their modules / units can be implemented using hardware methods such as circuits, and different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using hardware methods such as circuits. The components can be implemented using software programs that run on the processor integrated within the chip module. The remaining (if any) modules / units can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into the terminal, each of its components / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or in different components within the terminal. Alternatively, at least some modules / units can be implemented using software programs that run on the processor integrated within the terminal, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits.

[0094] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A method for detecting anomalies in a semiconductor production line, characterized in that, The method is executed by a computer; the method includes: Obtain information about the exception object; Based on the abnormal object information, the corresponding raw material information is read from the preset raw material database; The analysis is based on the raw material information obtained to determine the cause of the semiconductor production line abnormality.

2. The semiconductor production line anomaly detection method as described in claim 1, characterized in that, The abnormal object is at least one of the following: abnormal process, abnormal process step, abnormal process equipment, and abnormal plant equipment.

3. The semiconductor production line anomaly detection method as described in claim 1, characterized in that, The preset raw material database includes information on at least one of the following raw materials: Warehouse entry information for each raw material; Outbound information for each raw material; Usage records for each raw material; Outgoing inspection information for each raw material; Change records for each raw material; Abnormal raw material information.

4. The semiconductor production line anomaly detection method as described in claim 3, characterized in that, The raw material warehousing information includes at least one of the following: the name, quantity, batch number, and supplier information of the raw material.

5. The semiconductor production line anomaly detection method as described in claim 3, characterized in that, The outbound information of the raw materials includes at least one of the following: the name, quantity, and batch number of the raw materials.

6. The semiconductor production line anomaly detection method as described in claim 3, characterized in that, The raw material usage record information includes at least one of the following: raw material start-up time information and raw material end-up time information.

7. The semiconductor production line anomaly detection method as described in claim 3, characterized in that, The shipment inspection information of the raw materials includes: the test results of various performance parameters of the raw materials.

8. The semiconductor production line anomaly detection method as described in claim 3, characterized in that, The change record information for the raw materials includes at least one of the following: change information for the model of the raw materials and change information for the manufacturer.

9. The semiconductor production line anomaly detection method as described in claim 3, characterized in that, The abnormal raw material information includes: identification information of expired raw materials, identification information of raw materials with abnormal shipment inspection, and identification information of raw materials with abnormal storage environment.

10. The semiconductor production line anomaly detection method as described in claim 1, characterized in that, The preset raw material database also includes: the correspondence information between each abnormal object and each raw material information.

11. The semiconductor production line anomaly detection method as described in claim 1, characterized in that, The analysis based on the read raw material information includes: A commonality analysis was performed on the measured values ​​of the raw material information read at different times on the wafers, and a commonality analysis chart was generated.

12. The semiconductor production line anomaly detection method as described in claim 11, characterized in that, The analysis based on the read raw material information also includes: The measured values ​​of wafers using abnormal raw materials or wafers using modified raw materials are identified in the commonality analysis chart.

13. A semiconductor production line anomaly detection device, characterized in that, include: The acquisition unit is suitable for acquiring information about abnormal objects. The reading unit is adapted to read corresponding raw material information from a preset raw material database based on the abnormal object information; The analysis unit is suitable for performing analysis based on the read raw material information to determine the cause of abnormalities in the semiconductor production line.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by a processor to implement the steps of the method according to any one of claims 1 to 12.

15. An electronic device comprising a memory and a processor, wherein the memory stores a computer program capable of running on the processor, characterized in that, When the processor runs the computer program, it performs the steps of the method according to any one of claims 1 to 12.

Images