Wafer yield influencing factor analysis method and device, electronic equipment and storage medium

Abstract

The present disclosure relates to a wafer yield influencing factor analysis method and device, an electronic device and a storage medium. The method comprises: obtaining yield data and equipment data of batch wafers manufactured by a wafer manufacturing equipment; determining a key chamber causing low-yield wafers and a wafer set manufactured in the key chamber according to the yield data and the equipment data of the batch wafers; and determining an influencing factor analysis result of the wafer manufacturing equipment on wafer yield according to the yield of each wafer in the wafer set manufactured in the key chamber and parameter values of a plurality of operation parameters when each wafer is manufactured, wherein the influencing factor analysis result comprises a key operation parameter causing low-yield wafers in the key chamber and a parameter threshold corresponding to the key operation parameter, and the parameter threshold represents a parameter value limit causing low-yield wafers. Thus, the key chamber and the key operation parameter affecting the yield can be quickly and accurately located, thereby facilitating the development of corresponding improvement measures.

Description

Technical Field

[0001] This disclosure relates to the field of semiconductor manufacturing, and more particularly to a method and apparatus for analyzing factors affecting wafer yield, electronic equipment, and storage media. Background Technology

[0002] In wafer manufacturing, wafer yield is a crucial indicator of production quality and efficiency. Even minor deviations in each process step can lead to low yield, high defect rates, or abnormal wafer testing parameters. Failure analysis is an important tool for engineers to address these issues. Through various correlation analysis methods, such as tool commonality analysis and applied analysis of variance (ANOVA), specific process steps and equipment (machines) that may cause anomalies are identified.

[0003] However, existing analytical methods are still not precise enough to pinpoint which equipment parameters are abnormal and causing a decline in overall wafer yield. Fluctuations in equipment parameters often directly impact wafer yield. Therefore, how to analyze and optimize the factors affecting wafer yield in more detail to improve wafer yield has become a key focus of the industry. Summary of the Invention

[0004] In view of this, this disclosure proposes a method and apparatus, electronic equipment and storage medium for analyzing factors affecting wafer yield, which can combine yield data and equipment data of batch wafers to quickly and accurately locate key chambers and key operating parameters that affect yield, thereby facilitating the formulation of corresponding improvement measures.

[0005] According to one aspect of this disclosure, a method for analyzing factors affecting wafer yield is provided, comprising: acquiring yield data and equipment data of a batch of wafers manufactured by a wafer manufacturing equipment, wherein the wafer manufacturing equipment includes multiple chambers for manufacturing wafers, the yield data includes the yield of each wafer in the batch, and the equipment data includes the chamber in which each wafer in the batch is manufactured and a set of operating parameters for the chamber, the set of operating parameters including parameter values ​​of multiple operating parameters during wafer manufacturing within the chamber; and based on the yield data and equipment data of the batch of wafers... The critical chamber for producing low-yield wafers and the set of wafers manufactured within the critical chamber are identified, wherein the low-yield wafers are wafers with a yield below a first threshold. Based on the yield of each wafer in the set of wafers manufactured within the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield are determined. The analysis results of the factors affecting the equipment include the critical operating parameters in the critical chamber that lead to low-yield wafers and the parameter thresholds corresponding to the critical operating parameters. The parameter thresholds characterize the parameter value limits that produce low-yield wafers.

[0006] In one possible implementation, determining the analysis results of the factors affecting wafer yield of the wafer manufacturing equipment based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: using a first analysis algorithm to determine a first analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, wherein the first analysis result includes a first operating parameter in the critical chamber that leads to low yield wafers and a parameter threshold corresponding to the first operating parameter; and using a second analysis algorithm to determine the factors affecting wafer yield based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer. The yield of the wafer and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer are used to determine a second analysis result. The second analysis result includes the second operating parameter in the critical chamber that leads to low-yield wafers and the degree of influence of the second operating parameter on the wafer yield. Based on the first analysis result and the second analysis result, the analysis result of the influencing factors of the wafer manufacturing equipment on the wafer yield is determined. The second operating parameter is the same as or partially the same as the first operating parameter. The analysis result of the influencing factors also includes the degree of influence of the critical operating parameter in the critical chamber that leads to low-yield wafers on the wafer yield and / or the result of ranking the critical operating parameters according to the degree of influence.

[0007] In one possible implementation, determining the second analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: determining the degree of influence of each of the multiple operating parameters on the wafer yield based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer; and determining a second operating parameter in the critical chamber that leads to low-yield wafers based on the degree of influence of each of the multiple operating parameters on the wafer yield.

[0008] In one possible implementation, determining the degree of influence of each of the multiple operating parameters on the wafer yield based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: determining a first parameter group corresponding to low-yield wafers and a second parameter group corresponding to high-yield wafers based on the yield of each wafer in the wafer set manufactured in the critical chamber, wherein the high-yield wafers are wafers with a yield higher than a second threshold, the second threshold being greater than the first threshold, and the first parameter group including the yield of each low-yield wafer manufactured in the critical chamber. The first parameter group contains the parameter values ​​of multiple operating parameters during high-yield wafer fabrication, and the second parameter group contains the parameter values ​​of multiple operating parameters during the fabrication of each high-yield wafer in the critical chamber. By comparing and analyzing the first parameter group and the second parameter group, the parameter difference corresponding to each of the multiple operating parameters is obtained. The parameter difference characterizes the degree of difference between the parameter values ​​of the operating parameters corresponding to high-yield wafers and the parameter values ​​of the operating parameters corresponding to low-yield wafers. The parameter difference corresponding to each of the multiple operating parameters is determined as the degree of influence of each of the multiple operating parameters on the wafer yield.

[0009] In one possible implementation, the first analysis algorithm includes a decision tree algorithm, and the second analysis algorithm includes a T-detection algorithm.

[0010] In one possible implementation, the method further uses a decision tree algorithm to determine the key chamber that produces low-yield wafers and the set of wafers manufactured in the key chamber based on the yield data and equipment data of the batch of wafers, and to determine the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield based on the yield of each wafer in the set of wafers manufactured in the key chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer.

[0011] In one possible implementation, the method further includes: generating optimization suggestions for key chambers and / or key operating parameters within key chambers of the wafer manufacturing equipment based on the analysis results of the factors affecting wafer yield of the wafer manufacturing equipment; generating an analysis report based on the analysis results of the factors affecting wafer yield and the optimization suggestions; and pushing the analysis report to relevant personnel so that the relevant personnel can optimize the key chambers and / or key operating parameters within key chambers of the wafer manufacturing equipment with the goal of improving wafer yield based on the analysis report.

[0012] According to another aspect of this disclosure, a wafer yield influencing factor analysis device is provided, comprising: an acquisition module, configured to acquire yield data of a batch of wafers manufactured by a wafer manufacturing equipment and equipment data, wherein the wafer manufacturing equipment includes a plurality of chambers for manufacturing wafers, the yield data includes the yield of each wafer in the batch, and the equipment data includes the chamber in which each wafer in the batch is manufactured and a set of operating parameters for the chamber, the set of operating parameters including parameter values ​​of a plurality of operating parameters when manufacturing wafers in the chamber; and a first determination module, configured to determine the yield data of the batch of wafers and the equipment data based on the yield data of the batch of wafers and the equipment data. The system prepares data to identify the critical chamber that produces low-yield wafers and the set of wafers manufactured within the critical chamber, wherein the low-yield wafers are wafers with a yield lower than a first threshold. A second determining module is used to determine the analysis results of the factors influencing the wafer manufacturing equipment's impact on wafer yield based on the yield of each wafer in the set of wafers manufactured within the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer. The analysis results include the key operating parameters in the critical chamber that lead to low-yield wafers and the corresponding parameter thresholds, wherein the parameter thresholds characterize the parameter value limits that produce low-yield wafers.

[0013] According to another aspect of this disclosure, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to implement the above-described method when executing instructions stored in the memory.

[0014] According to another aspect of this disclosure, a non-volatile computer-readable storage medium is provided that stores computer program instructions thereon, wherein the computer program instructions, when executed by a processor, implement the above-described method.

[0015] According to another aspect of this disclosure, a computer program product is provided, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above-described method.

[0016] According to various aspects of this disclosure, by acquiring wafer yield data and equipment data from the equipment, the critical chambers that produce low-yield wafers can be identified. Furthermore, based on the yield of the wafers manufactured in the critical chambers and the parameter values ​​of the operating parameters, the key operating parameters and corresponding parameter thresholds that cause low-yield wafers in the critical chambers can be determined. This enables the rapid and accurate location of the critical chambers, key operating parameters, and corresponding parameter thresholds that affect yield, thereby helping relevant personnel to promptly locate equipment problems and formulate corresponding improvement measures to effectively improve the yield of wafers manufactured on the production line.

[0017] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0018] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this disclosure together with the specification and serve to explain the principles of this disclosure.

[0019] Figure 1 A flowchart is shown for a method for analyzing factors affecting wafer yield according to an embodiment of the present disclosure.

[0020] Figure 2 This diagram illustrates a tree-like visualization of the influencing factor analysis results output by a decision tree algorithm according to an embodiment of the present disclosure.

[0021] Figure 3 A block diagram of a wafer yield influencing factor analysis apparatus according to an embodiment of the present disclosure is shown.

[0022] Figure 4 A block diagram of an electronic device 1900 according to an embodiment of the present disclosure is shown. Detailed Implementation

[0023] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0024] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0025] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more of a plurality of elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C. In the description of this disclosure, "multiple" means two or more, unless otherwise explicitly specified.

[0026] It should be understood that the terms "first," "second," etc., in the claims, specification, and drawings of this disclosure are used to distinguish different objects, rather than to describe a specific order. The terms "comprising" and "including" as used in the specification and claims of this disclosure indicate the presence of the described features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0027] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0028] In practical applications, the wafer yield influencing factor analysis method of this disclosure can be deployed on various terminal devices through software or hardware modifications. These terminal devices may be equipped with a rendering engine. The terminal devices involved in this disclosure can refer to devices with wireless and / or wired connection functions. Wireless connection means that they can connect to other devices via Wi-Fi, Bluetooth, or other wireless connection methods. The terminal devices involved in this disclosure can also communicate with other devices via wired connection functions. The terminal devices involved in this disclosure can be touchscreen, non-touchscreen, or screenless. Touchscreen devices can be controlled by clicking or swiping on the display screen using fingers, styluses, etc. Non-touchscreen devices can connect to input devices such as mice, keyboards, and touch panels to control the terminal device. Screenless devices can be, for example, screenless Bluetooth speakers. For example, the terminal devices in this application may include, but are not limited to, user equipment (UE), mobile devices, mobile terminals, handheld devices, tablet computers, laptops, PDAs, computing devices, etc.

[0029] The wafer yield influencing factor analysis method of this disclosure can also be deployed on a server. This server may have a rendering engine deployed on it. The server can be located in the cloud or locally, and can be a physical device or a virtual device, such as a virtual machine or container. It has wireless communication capabilities, which can be configured in the server's chip (system) or other components. This can refer to a device with wireless connectivity, meaning it can connect to other servers or terminal devices via Wi-Fi, Bluetooth, or other wireless connection methods. The server involved in this disclosure can also have wired communication capabilities. For example, the server in this disclosure can be located in the cloud, communicating with terminal devices, receiving yield data and device data sent by the terminal devices, and using the wafer yield influencing factor analysis method deployed on the server, determining the wafer yield influencing factor analysis results based on the yield data and device data, and returning them to the terminal devices so that the wafer yield influencing factor analysis results can be displayed to the user on the terminal devices.

[0030] Figure 1 A flowchart illustrating a method for analyzing factors affecting wafer yield according to an embodiment of this disclosure is shown. Figure 1 As shown, the method includes steps S11 to S13.

[0031] In step S11, the yield data of the batch of wafers manufactured by the wafer manufacturing equipment and the equipment data are obtained. The wafer manufacturing equipment includes multiple chambers for manufacturing wafers. The yield data includes the yield of each wafer in the batch. The equipment data includes the chamber in which each wafer in the batch is manufactured and the set of operating parameters of the chamber. The set of operating parameters includes the parameter values ​​of multiple operating parameters when manufacturing wafers in the chamber.

[0032] The wafer manufacturing equipment can be a device used to manufacture wafers. A wafer manufacturing equipment can include multiple chambers, each of which can be used to manufacture at least one wafer. The operating parameters can be parameters used to manufacture wafers, such as temperature, pressure, content of various chemical reagents, density, etc. The types and quantities of operating parameters in the embodiments of this disclosure are not limited.

[0033] In practical applications, yield data can come from yield test results at CP (Chip Probing) testing stations, including classification information for high-yield and low-yield wafers (i.e., which wafers are high-yield and which are low-yield). CP testing stations can test the yield of each wafer. Equipment data can be synchronously acquired through an FDC (Fault Detection and Classification) system, including the operating parameters and values ​​of each chamber. The FDC system can record the parameter values ​​of the operating parameters within the chambers in real time during wafer manufacturing. It should be understood that this disclosure does not limit the method of acquiring yield data and equipment data.

[0034] This method can periodically acquire yield data and equipment data of batches of wafers manufactured by the wafer manufacturing equipment within a cycle. Alternatively, it can define custom trigger conditions (such as when the number of low-yield wafers manufactured by the equipment exceeds a certain threshold) to trigger the acquisition of yield data and equipment data of batches of wafers manufactured by the equipment within a certain cycle, in order to analyze the influencing factors of wafer yield. This disclosure does not limit the scope of this method. It should be understood that this disclosure also does not limit the number of wafers in a batch.

[0035] Optionally, yield data and equipment data of all wafers manufactured by the equipment within a certain period can be obtained. Alternatively, yield data and equipment data of a portion of the wafers manufactured by the equipment within a certain period can be obtained. This portion of data may include low-yield wafers with yields below a first threshold and high-yield wafers with yields above a second threshold. That is, the batch of wafers used in step S11 may include low-yield wafers and high-yield wafers. The first threshold is less than the second threshold. Those skilled in the art can customize the first and second thresholds according to actual needs. For example, the first threshold may be 70% and the second threshold may be 90%. That is, yield data and equipment data of wafers with yields below 70% and wafers with yields above 90% manufactured by the equipment can be obtained. This disclosure embodiment does not limit this.

[0036] In practical applications, after obtaining the aforementioned yield data and equipment data, the yield data and equipment data can be correlated. That is, the wafer yield can be correlated with the chamber where the wafer was manufactured and the set of operating parameters of the chamber to obtain a unified dataset, which facilitates subsequent analysis. Considering that there may be missing data in the yield data and equipment data (such as missing yield of a certain wafer or missing parameter value of a certain operating parameter), the missing data can be filled (e.g., filled with the average value) or other data associated with the missing data can be directly removed. Furthermore, considering that there are many types of operating parameters in the equipment data and that the values ​​may vary greatly, in order to facilitate subsequent processing of the equipment data, preprocessing operations such as normalization can be performed on the parameter values ​​of various operating parameters in each set of operating parameters in the equipment data. This disclosure does not limit this aspect.

[0037] In step S12, based on the yield data of batch wafers and equipment data, the key chamber for generating low-yield wafers and the set of wafers manufactured in the key chamber are determined. Low-yield wafers are wafers with a yield lower than a first threshold.

[0038] It should be understood that, given the known yield of each wafer in a batch and the chamber in which each wafer is manufactured, and considering that different chambers of the equipment can typically be numbered, the batch of wafers can be classified according to the chamber numbers. This allows us to obtain the wafers manufactured in each chamber and their corresponding yields. Furthermore, based on a first threshold for identifying low-yield wafers and a second threshold for identifying high-yield wafers, we can determine whether each wafer manufactured in each chamber is a high-yield or low-yield wafer. This also reveals the critical chamber that produces low-yield wafers and the set of wafers manufactured in that critical chamber. Alternatively, as mentioned above, the batch of wafers can primarily include low-yield and high-yield wafers. Therefore, in step S11, low-yield and high-yield wafers can be further categorized. Thus, in step S12, based on the categorization identifiers used to identify low-yield and high-yield wafers and the chamber number of each wafer's chamber, the critical chamber that produces low-yield wafers can be determined.

[0039] Considering that in practice, there are often chambers with no low-yield wafers (e.g., all high-yield wafers), chambers with no high-yield wafers (e.g., all low-yield wafers), and chambers containing both high-yield and low-yield wafers. For example, suppose the batch of wafers being analyzed contains 80 wafers, of which 40 are high-yield and 40 are low-yield. The equipment has three chambers, labeled -P, -R, and -Q. By classifying the 80 wafers according to the chambers, we might find that chamber -P contains 32 wafers, all of which are high-yield (i.e., no low-yield wafers); chamber -R contains 28 wafers, of which 8 are high-yield and 20 are low-yield; and chamber -Q contains 20 wafers, all of which are low-yield. Therefore, it is known that chambers -R and -Q are the key chambers for producing low-yield wafers.

[0040] For cavities without low-yield wafers (e.g., all wafers in the -P cavity exhibit high yield), it indicates that the cavity may not have process issues or the impact may be minor. For cavities without high-yield wafers (e.g., all wafers in the -Q cavity exhibit low yield), it indicates that the cavity may have significant process issues and is a critical cavity affecting yield. A comprehensive inspection and analysis of these cavities, along with optimization of the process flow within the cavity, is necessary. However, it is difficult to analyze individually which specific operating parameters within these cavities without high-yield wafers cause low yield. Therefore, for critical cavities with both high-yield and low-yield wafers (e.g., the -R cavity), the specific impact of operating parameters on wafer yield can be further determined by using the wafer assemblies manufactured in these critical cavities (e.g., the -R cavity contains 28 wafers). Alternatively, if no chambers containing both high-yield and low-yield wafers are found after classification, the impact of operating parameters on wafer yield can be analyzed by using the difference in operating parameter values ​​between chambers without low-yield wafers (these chambers without low-yield wafers can be understood as reference chambers) and key chambers without high-yield wafers.

[0041] In step S13, based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield are determined. The analysis results of the factors affecting the wafer yield include the key operating parameters in the critical chamber that lead to low-yield wafers and the parameter thresholds corresponding to the key operating parameters. The parameter thresholds represent the parameter value limits that produce low-yield wafers.

[0042] As described above, the critical chambers for producing low-yield wafers can include: chambers without high-yield wafers (i.e., all low-yield wafers), and / or chambers containing both high-yield and low-yield wafers. Specifically, for critical chambers containing both high-yield and low-yield wafers, the differences between the parameter values ​​of various operating parameters when producing high-yield wafers and those when producing low-yield wafers can be analyzed. Operating parameters with significant differences (e.g., exceeding a certain threshold) can be identified as the critical operating parameters leading to low-yield wafers. Furthermore, it can be determined that wafers manufactured when this critical operating parameter is below or above a certain value are low-yield wafers, thus obtaining the parameter threshold corresponding to this critical operating parameter. For example, assuming the operating parameters include temperature and pressure parameters, for the 8 high-yield wafers and 20 low-yield wafers contained in the -R chamber, the parameter values ​​of temperature and pressure corresponding to the 8 high-yield wafers in the -R chamber can be compared and analyzed with the parameter values ​​of temperature and pressure corresponding to the 20 low-yield wafers. The operating parameter with a large difference in parameter value (e.g., greater than a certain threshold) can be identified as the key operating parameter causing the low-yield wafer (e.g., if the parameter value corresponding to the temperature parameter has a large difference, then the temperature parameter is the key operating parameter). At the same time, it can also be found that the wafers manufactured when the key operating parameter is below a certain value or above a certain value are low-yield wafers, thus obtaining the parameter threshold corresponding to the key operating parameter (e.g., when the temperature is higher than 35°C, the wafers produced are low-yield wafers, then 35°C is the parameter threshold corresponding to the temperature parameter).

[0043] Specifically, for critical chambers lacking high-yield wafers, the differences in various operating parameters between chambers lacking low-yield wafers and critical chambers lacking high-yield wafers can be calculated. Then, operating parameters with significant differences (e.g., exceeding a certain threshold) between the two types of chambers can be identified as the critical operating parameters leading to low-yield wafers. Furthermore, wafers manufactured when this critical operating parameter is below or above a certain value are considered low-yield wafers, thus obtaining the parameter threshold corresponding to this critical operating parameter. For example, assuming operating parameters include temperature and pressure parameters, the parameter values ​​for temperature and pressure corresponding to the 32 wafers in the -P chamber can be compared and analyzed with the parameter values ​​for temperature and pressure corresponding to the 20 wafers in the -Q chamber. This allows for the identification of operating parameters with significant differences (e.g., exceeding a certain threshold) between the -Q and -P chambers as the critical operating parameters leading to low-yield wafers, and the determination of the corresponding parameter threshold.

[0044] It should be understood that the implementation methods for analyzing the key operating parameters and corresponding parameter thresholds affecting the low-yield wafer can be implemented by using a key chamber that contains both high-yield and low-yield wafers. Alternatively, the implementation methods for analyzing the key operating parameters and corresponding parameter thresholds affecting the low-yield wafer can be implemented by using a reference chamber without low-yield wafers and a key chamber without high-yield wafers. Of course, the analysis results of the above two implementation methods can also be combined to obtain the final analysis results of the influencing factors. For example, the intersection or union of the above two implementation methods can be taken to obtain the final analysis results of the influencing factors. This disclosure does not limit this aspect.

[0045] In one possible implementation, a decision tree algorithm can be used to implement step S12, which determines the critical chamber that produces low-yield wafers and the set of wafers manufactured within the critical chamber based on the yield data and equipment data of the batch of wafers. Step S13, based on the yield of each wafer in the set of wafers manufactured within the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, determines the analysis results of the factors affecting wafer manufacturing equipment on wafer yield. That is, a decision tree analysis is used to examine the critical chamber through which the low-yield wafers pass, the key operating parameters within that critical chamber that cause low yield, and the parameter thresholds of those key operating parameters. This method allows for efficient execution of steps S12 to S13 and provides accurate results for the analysis of the influencing factors.

[0046] To facilitate understanding of the process of implementing steps S12 to S13 using the decision tree algorithm, this embodiment first provides a brief introduction to the basic principles of the decision tree algorithm. The decision tree algorithm includes: a splitting process, where the construction of the decision tree starts from the root node, which contains the entire dataset (such as the dataset generated by associating the yield data of batch wafers and equipment data obtained in step S11). The algorithm then selects split points based on features (attributes) to divide the dataset into two or more subsets. The splitting is based on selecting a feature and its threshold (such as the chamber number or the parameter threshold of the operating parameters mentioned above), ensuring that the data in the split subsets is as "pure" as possible, i.e., the target variable (such as the category label in a classification problem) is relatively concentrated. The splitting attribute (Feature Selection) mainly uses information gain, which measures the information difference before and after splitting (such as the difference between chambers or parameter values). Specifically, information gain can be calculated based on entropy, which is an indicator used to measure uncertainty. Node splitting... Splitting involves splitting a node into child nodes based on selected features and thresholds. These child nodes continue to split until certain stopping conditions are met, such as reaching the maximum depth, having too few samples in the node, or all data being pure (i.e., having only one class). Stopping criteria define when splitting in a decision tree stops: reaching a preset tree depth (max depth), having fewer samples in the split child nodes than a certain threshold (minsamples split), or the samples in the current node being too "pure" to split effectively further. This process is designed to prevent overfitting, as decision trees can become overly complex. Pruning simplifies the generated decision tree. Pre-pruning stops splitting during tree generation, while post-pruning removes less important branches after the tree is fully generated. The structure of a decision tree mainly includes the root node, which contains all the data and is where the split begins; internal nodes, each representing a test condition for a feature; and leaf nodes, which are the decision results (such as class labels or regression values).

[0047] For example, based on the decision tree algorithm described above, assume the dataset contains the device's chamber number (-P, -R, -Q), operating parameter CONT_AE_FIELD, and yield category (G: high yield, B: low yield). Missing values ​​can be imputed using the mean, and the chamber (e.g., chamber ID) and operating parameter CONT_AE_FIELD can be used as splitting attributes. Based on this, the decision tree algorithm can divide the wafers into multiple groups. For example, the first layer can split based on the chamber ID (e.g., chambers -P, -R, -Q), which is the first split point in the decision tree. The second layer can further split within each chamber using the operating parameter CONT_AE_FIELD, for example, whether the operating parameter is less than a certain value (e.g., ≤26.987091), to further subdivide high-yield and low-yield wafers. Through decision tree splitting, the key chambers and key operating parameter combinations affecting yield can be analyzed, thereby enabling the analysis of the performance of these wafers in different device chambers and the key influencing factors using the decision tree algorithm.

[0048] In practical applications, for example, the DecisionTreeClassifier tool can be used to build decision tree models, selecting information gain as the splitting criterion, limiting the maximum tree depth to 3 to avoid overfitting, and setting the text format for the output decision tree to display the splitting process. The matplotlib tool can also be used to plot the decision tree graphically for easier visualization and understanding. In other words, the results of the influencing factor analysis output by the decision tree algorithm can be graphically visualized, allowing users to intuitively obtain the results of the analysis of factors affecting wafer yield. For example, Figure 2 This diagram illustrates a tree-like visualization of the influencing factor analysis results output by a decision tree algorithm, as shown below. Figure 2As shown, the root node of the decision tree has a total of 80 wafers (N=80), of which 40 are high-yield wafers and the other 40 are low-yield wafers. Green represents high-yield wafer c1 and red represents low-yield wafer c2. -P, -R, and -Q are the device chamber IDs, and CONT_AE_FIELD is the operating parameter. Based on the root node and the first-level factor condition (i.e., the device chamber ID), it is divided into three intermediate nodes (i.e., chambers): -P, -R, and -Q. -P (the green intermediate node) contains 32 wafers (N=32), all of which are high-yield wafers (c1), with no low-yield wafers (c2). -R (the mixed-color intermediate node) contains 28 wafers (N=28), of which 8 are high-yield wafers (c1) and 20 are low-yield wafers (c2). -Q (the red intermediate node) contains 20 wafers (N=20), all of which are low-yield wafers (c2). For chamber-R, CONT_AE_FIELD is determined as a critical operating parameter based on the second-level factor condition (i.e., the parameter value of the operating parameter). This critical operating parameter can be split into two leaf nodes. That is, under the path of chamber-R, the impact of the critical operating parameter CONT_AE_FIELD on low-yield wafers is further analyzed. Specifically, when CONT_AE_FIELD ≤ 26.987091, all 20 wafers are low-yield wafers (c2); when CONT_AE_FIELD > 26.987091, all 8 wafers are high-yield wafers (c1). Therefore, 26.987091 is the parameter threshold corresponding to this critical operating parameter. It should be understood that for any other operating parameter, it is possible to determine whether it is a critical operating parameter and to determine its corresponding parameter threshold.

[0049] Understandable, through Figure 2 It can be visually observed that all wafers in the -P chamber exhibit high yield, meaning that high-yield wafers are mainly distributed in the -P chamber, while low-yield wafers are mainly distributed in the -Q and -R chambers. Among them, all wafers in the -Q chamber exhibit low yield, indicating that there may be significant process problems in this chamber, making it a key chamber affecting yield. As for the -R chamber, it is further affected by the CONT_AE_FIELD parameter. That is, the operating parameter CONT_AE_FIELD significantly affects the yield, and when CONT_AE_FIELD ≤ 26.987091, the wafer exhibits low yield, while CONT_AE_FIELD > 26.987091, the wafer exhibits high yield.

[0050] Decision tree algorithms require minimal data preprocessing, such as feature standardization, and can handle both numerical and categorical features. The output is visually interpretable and easy to understand. The decision tree output text can display splitting decisions based on chamber ID and equipment parameters, helping to identify key operating parameters affecting yield. The graphical representation of the decision tree illustrates the impact of different equipment chambers and operating parameters on yield, clearly showing which chambers and parameters lead to low yield. This visualization of the analysis process helps engineers quickly identify the causes of yield declines and develop optimization strategies.

[0051] In practical applications, identifying the factors influencing wafer yield helps engineers quickly pinpoint the root causes of low yield problems and develop targeted improvement measures, ultimately enhancing the overall efficiency and quality of wafer manufacturing. Therefore, after obtaining the analysis results of the factors influencing wafer manufacturing equipment on wafer yield, the method may further include:

[0052] Based on the analysis of factors affecting wafer yield by wafer manufacturing equipment, optimization suggestions are generated for key chambers and / or key operating parameters within key chambers of wafer manufacturing equipment.

[0053] Based on the analysis results of influencing factors and optimization suggestions, an analysis report is generated and pushed to relevant personnel so that they can optimize key chambers and / or key operating parameters within key chambers of wafer manufacturing equipment with the goal of improving wafer yield.

[0054] For example, regarding the above Figure 2 The analysis results of the influencing factors shown can generate optimization suggestions that include: 1. Focusing on and adjusting the setting value of the operating parameter CONT_AE_FIELD, especially ensuring that its value is higher than the parameter threshold of 26.987091, to improve yield. For example, the CONT_AE_FIELD parameter of the -R chamber can be adjusted to be greater than 26.987091. 2. For chambers with low yield (such as the -Q chamber), detailed key inspections and maintenance are required, and the process flow of the chamber should be optimized. In practical applications, some suggestion templates can be set, and after obtaining the analysis results of the influencing factors, the analysis results of the influencing factors can be combined with the corresponding suggestion templates to generate optimization suggestions; or, large language models in the field of artificial intelligence can be used to generate optimization suggestions for key chambers and / or key operating parameters in key chambers of wafer manufacturing equipment based on the analysis results of the influencing factors of wafer manufacturing equipment on wafer yield. This disclosure does not limit this aspect.

[0055] It should be understood that the analysis report may include the analysis results of the aforementioned influencing factors and optimization suggestions. This disclosure does not limit the style, layout, or delivery method of the analysis report. For example, the analysis report can be delivered to relevant personnel via email, SMS, or any application. These personnel may include engineers who maintain and manage wafer manufacturing equipment, enabling them to optimize the process flow of key chambers and / or adjust the parameter values ​​of key operating parameters used in the key chambers in a timely manner based on the analysis report. In actual production, the effects of optimization can be verified, and further iterative optimization can be performed based on the verification results until the predetermined goal is achieved. For example, after optimization, the yield of wafers produced in the -R chamber may be increased to over 90%, the low yield problem of the -Q chamber may be significantly improved, and the overall production yield may be increased by approximately 15%. This disclosure does not limit the scope of the optimization.

[0056] According to the method of this disclosure, by acquiring the yield data and equipment data of the wafers manufactured by the equipment, the key chambers that produce low-yield wafers are identified. Then, based on the yield of the wafers manufactured in the key chambers and the parameter values ​​of the operating parameters, the key operating parameters and corresponding parameter thresholds that cause low-yield wafers in the key chambers can be determined. Overall, the key chambers, key operating parameters and corresponding parameter thresholds that affect the yield can be quickly and accurately located, which helps relevant personnel to locate equipment problems in a timely manner and formulate corresponding improvement measures to effectively improve the yield of wafers manufactured by the equipment on the production line.

[0057] As mentioned above, decision tree algorithms can be used to determine the analysis results of factors affecting wafer yield from wafer manufacturing equipment. However, considering that using only decision tree algorithms may result in incomplete analysis results—for example, when multiple key operating parameters are identified, the degree of influence (or priority) of each key operating parameter on wafer yield cannot be known, thus preventing engineers from knowing the order in which to investigate each key operating parameter (e.g., a higher degree of influence of a key operating parameter should be investigated first, while a lower degree of influence can be investigated later)—in one possible implementation, step S13, based on the yield of each wafer in the wafer set manufactured in the key chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, determines the analysis results of factors affecting wafer yield from wafer manufacturing equipment. This may include:

[0058] Step S131: Using a first analysis algorithm, determine a first analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. The first analysis result includes the first operating parameter of the wafer in the critical chamber that causes low yield and the parameter threshold corresponding to the first operating parameter.

[0059] Step S132: Using the second analysis algorithm, determine the second analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. The second analysis result includes the second operating parameters of the wafers in the critical chamber that lead to low yield and the degree of influence of the second operating parameters on the wafer yield.

[0060] Step S133: Based on the first analysis results and the second analysis results, determine the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield;

[0061] Among them, the second operating parameter is the same as or partially the same as the first operating parameter. The results of the influencing factor analysis also include the degree of influence of the key operating parameters that lead to low yield wafers in the key chamber on the wafer yield and / or the results of ranking the key operating parameters according to the degree of influence.

[0062] In step S131, the first analysis algorithm can be a decision tree algorithm. That is, it can refer to the implementation method of using the decision tree algorithm to determine the key operating parameters affecting the yield in the key chamber in step S13 above, so as to determine the first operating parameter and the parameter threshold corresponding to the first operating parameter in the key chamber that leads to low yield. In other words, the decision tree algorithm can be used to implement the above steps S12 and S131 to obtain the first analysis result of the influencing factor analysis using the decision tree algorithm, which will not be elaborated here.

[0063] In step S132, the second analysis algorithm may employ, for example, the T-detection algorithm, or a machine learning algorithm such as the random forest algorithm or the support vector machine algorithm, to determine the second operating parameters of the wafers in the critical chamber that cause low yield and the degree of influence of the second operating parameters on the wafer yield based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. This disclosure embodiment does not limit this.

[0064] Optionally, step S132 above, which determines the second analysis result based on the yield of each wafer in the wafer assembly manufactured within the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, may include:

[0065] Step S1321: Based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, determine the degree of influence of each operating parameter on the wafer yield.

[0066] Step S1322: Based on the degree of influence of each of the multiple operating parameters on the wafer yield, determine the second operating parameter in the critical chamber that leads to the low yield of the wafer.

[0067] In step S1321, based on the yield of each wafer in the wafer assembly manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, the degree of influence of each of the multiple operating parameters on the wafer yield is determined, which may include:

[0068] Step S13211: Based on the yield of each wafer in the wafer set manufactured in the critical chamber, determine the first parameter group corresponding to the low-yield wafer and the second parameter group corresponding to the high-yield wafer. The first parameter group includes the parameter values ​​of multiple operating parameters when manufacturing each low-yield wafer in the critical chamber, and the second parameter group includes the parameter values ​​of multiple operating parameters when manufacturing each high-yield wafer in the critical chamber.

[0069] Step S13212: By comparing and analyzing the first parameter group and the second parameter group, the parameter difference of each operating parameter in the multiple operating parameters is obtained. The parameter difference characterizes the degree of difference between the parameter values ​​of the operating parameters corresponding to high-yield wafers and the parameter values ​​of the operating parameters corresponding to low-yield wafers.

[0070] Step S13213: Determine the parameter difference corresponding to each of the multiple operating parameters as the degree of influence of each of the multiple operating parameters on the wafer yield.

[0071] In step S13211, the critical chamber can be a chamber containing both high-yield and low-yield wafers (such as a -R chamber). Given the high-yield and low-yield wafers in the critical chamber, the operating parameters of the wafers can be divided into two groups: a first parameter group corresponding to the low-yield wafers and a second parameter group corresponding to the high-yield wafers. For example, assuming that the operating parameters include temperature and pressure parameters, for the 8 high-yield wafers and 20 low-yield wafers in the -R chamber, the first parameter group corresponding to the low-yield wafers includes the parameter values ​​of the temperature and pressure parameters for each of the 20 low-yield wafers, and the second parameter group corresponding to the high-yield wafers includes the parameter values ​​of the temperature and pressure parameters for each of the 8 high-yield wafers.

[0072] In step S13212, for example, a T-test algorithm can be used to compare and analyze the first parameter group and the second parameter group. The T-test algorithm is an algorithm that can be used to determine whether there is a significant difference between the means of two groups of data. Specifically, for each operating parameter, the T-test algorithm shown in formula (1) can be used to calculate the parameter difference T corresponding to each operating parameter:

[0073]

[0074] in, This represents the average value of a certain operating parameter in the first parameter group (such as the average value of the temperature parameter for low-yield wafers). n1 represents the average value of the same operating parameter in the second parameter group (e.g., the average value of the temperature parameter for high-yield wafers), n2 represents the number of parameter values ​​for a certain operating parameter in the first parameter group (i.e., the number of low-yield wafers corresponding to the first parameter group), and n2 represents the number of parameter values ​​for the same operating parameter in the second parameter group (i.e., the number of high-yield wafers corresponding to the second parameter group). Represents the pooled variance, where, Formula (2) can be used:

[0075]

[0076] in, This represents the variance of the parameter values ​​for a certain operating parameter in the first parameter group (such as the variance of the temperature parameter values ​​for low-yield wafers). This represents the variance of the parameter values ​​for the same operating parameter in the second parameter group (such as the variance of the temperature parameter values ​​for high-yield wafers).

[0077] It should be understood that the above-described method of using the T-detection algorithm to calculate the parameter variability corresponding to the running parameters is one possible implementation provided by this public embodiment. In fact, those skilled in the art can also use other known methods of calculating variability in the art. For example, the difference, variance, or mean squared error between the mean values ​​of the same running parameter in the first parameter group and the second parameter group can be used as the parameter variability corresponding to the running parameter. Alternatively, other known machine learning algorithms in the art (such as random forest, support vector machine, etc.) can be used to determine the parameter variability corresponding to each running parameter. This public embodiment does not limit this.

[0078] The greater the difference in the operating parameters, the greater the impact of the operating parameter on the wafer yield. The more likely the operating parameter is to cause low wafer yield, the greater the impact of the operating parameter on the wafer yield. Therefore, the difference in the parameters corresponding to each operating parameter among multiple operating parameters can be determined as the degree of impact of each operating parameter on the wafer yield among multiple operating parameters.

[0079] In step S1322, for example, based on the degree of influence of each operating parameter on wafer yield among multiple operating parameters, the operating parameter with an influence greater than a certain threshold can be selected as the second operating parameter that leads to low yield wafers; or, based on the degree of influence of each operating parameter on wafer yield among multiple operating parameters, multiple operating parameters can be sorted (e.g., in descending order), and then based on the result of the descending order, the operating parameter with the first m positions in the influence column can be selected as the determined second operating parameter. The second operating parameter is also the operating parameter with significant difference. This embodiment of the present disclosure does not limit this.

[0080] Optionally, the second operating parameter determined above can also be directly marked as a key operating parameter affecting wafer yield. That is, the implementation of steps S1321 to S1322 above can be directly referred to to determine the analysis results of the factors affecting wafer yield of the wafer manufacturing equipment. In other words, the second operating parameter in the key chamber that leads to low-yield wafers in the determined second analysis results can be used as the key operating parameter. Then, based on the parameter value of the determined key operating parameter, the parameter threshold corresponding to the key operating parameter can be determined. This disclosure embodiment does not limit this. For example, for the 28 wafers in the -R chamber above, they can be divided into a first parameter group corresponding to 20 low-yield wafers and a second parameter group corresponding to 8 high-yield wafers. Then, the parameter difference between the parameter values ​​of the two parameter groups can be compared by the T detection algorithm. If it is found that the parameter value of the temperature parameter has a significant difference between the two parameter groups (that is, the parameter difference is large), then the temperature parameter can be determined as the key operating parameter leading to low-yield wafers.

[0081] As described above, in step S12, it is also possible to identify critical chambers without high-yield wafers (i.e., critical chambers full of low-yield wafers). For such critical chambers without high-yield wafers, in step S13211, the parameter values ​​of multiple operating parameters corresponding to the low-yield wafers in such critical chambers can be directly used to form a first parameter group, and the parameter values ​​of multiple operating parameters corresponding to the high-yield wafers in a reference chamber without low-yield wafers (i.e., a reference chamber full of high-yield wafers) can be used to form a second parameter group. Then, steps S13212 to S13213 are executed to obtain the degree of influence of each operating parameter on the wafer yield. Then, step S1322 is executed to determine the second operating parameter in the critical chamber that causes the low-yield wafers. This will not be elaborated here.

[0082] In step S133, the first operating parameter contained in the first analysis result may be the same as or partially the same as the second operating parameter contained in the second analysis result. Therefore, the intersection or union of the first operating parameter contained in the first analysis result and the second operating parameter contained in the second analysis result can be used as the key operating parameter in the key chamber that leads to low yield wafers.

[0083] If the union of the first operating parameter and the second analysis result is adopted, and the first operating parameter and the second operating parameter are partially the same, then for the part of the second operating parameter that is different from the first operating parameter (that is, the part of the second operating parameter for which the parameter threshold has not been calculated), referring to the above implementation method of using the decision tree algorithm to determine the parameter threshold of the first operating parameter, the parameter value exhibited by this part of the second operating parameter on high-yield wafers and low-yield wafers is used to determine the parameter threshold corresponding to this part of the second operating parameter. This disclosure embodiment does not limit this.

[0084] It should be understood that the degree of influence of each operating parameter on wafer yield is known, and therefore the degree of influence of key operating parameters on wafer yield is also known. Furthermore, the results of ranking key operating parameters according to their degree of influence (e.g., in descending order) can also be obtained. Therefore, the results of the influencing factor analysis can also include the degree of influence of key operating parameters in the critical chamber that lead to low-yield wafers on wafer yield and / or the results of ranking key operating parameters according to their degree of influence.

[0085] As mentioned above, an analysis report can be generated based on the results of the influencing factor analysis, and the report can be pushed to relevant personnel to optimize the key operating parameters in the critical chamber. Thus, the analysis report pushed to relevant personnel can also include the degree of influence of key operating parameters on wafer yield and / or the results of ranking key operating parameters according to the degree of influence. In other words, key operating parameters can be ranked according to the degree of influence and provided to engineers to investigate and optimize parameter problems based on the priority of key operating parameters in the process. This is equivalent to prioritizing key operating parameters so that key operating parameters with a higher degree of influence can be optimized first, which is conducive to timely improvement of the wafer yield of the equipment.

[0086] In practical applications, the analysis report may also include the degree of influence of each operating parameter on wafer yield and the results of sorting all operating parameters according to the degree of influence. Key operating parameters, their degree of influence, and parameter thresholds may be highlighted (e.g., using different colors, fonts, or bolding) to facilitate users in viewing the degree of influence of all operating parameters and intuitively viewing information such as the degree of influence of key operating parameters and parameter thresholds. This disclosure does not limit this aspect.

[0087] According to the method of this disclosure, by integrating wafer yield data from the CP site and equipment data from the FDC system, and combining it with a decision tree algorithm, a highly efficient and accurate method for analyzing factors affecting wafer yield is proposed. This method can quickly locate key chambers and key performance parameters affecting yield, and propose targeted optimization measures, thereby significantly improving production efficiency and product quality.

[0088] According to the method of this disclosure, full-process data traceability and correlation are achieved by combining wafer yield data (test data from the CP site) with equipment data synchronized by the FDC system. A decision tree algorithm is used to quickly identify key chambers and key operating parameters affecting yield. Specific parameter optimization directions and threshold references (e.g., CONT_AE_FIELD>26.987091) can be provided.

[0089] According to the method of this disclosure, parameter data corresponding to low-yield wafers and high-yield wafers are grouped, and a geothermal analysis algorithm is used to identify parameters with significant differences, thereby further clarifying the key factors affecting yield. After grouping all parameter data of the key chambers found through the decision tree, a second analysis algorithm is used to perform difference analysis on the two groups of data, and the operating parameters are sorted according to the significance value of the difference, which facilitates engineers to troubleshoot and optimize according to priority.

[0090] According to the method of this disclosure, analysis results can be instantly pushed to relevant engineers through related applications and email systems, ensuring rapid response to optimization measures. It can directly locate and optimize influencing factors based on data analysis results, avoiding repeated trials in traditional methods and improving optimization efficiency. Furthermore, it is not only applicable to wafer yield optimization but can also be extended to quality improvement scenarios in other semiconductor manufacturing processes.

[0091] The method according to embodiments of this disclosure enables a more comprehensive understanding and resolution of abnormal machine parameter problems by constructing an algorithmic model to identify these problems. It can learn the relationship between parameters and machine performance from large amounts of data, combining domain expertise to achieve accurate identification and effective resolution of abnormal machines and their parameter problems.

[0092] The method according to embodiments of this disclosure implements a method for analyzing the relationship between wafer yield and machine chambers and machine parameters using a decision tree algorithm. It implements a technique for segmenting wafer data based on machine chamber ID, machine operating parameters, and wafer yield. This method can identify abnormal machines affecting wafer yield while also identifying key operating parameters within critical chambers. It can provide specific suggestions for machine parameter optimization and chamber maintenance based on the decision tree analysis results. Software tools and algorithms can be developed to automate the above decision tree analysis process. Visualization techniques can be used to intuitively display the decision tree analysis results. A method can be used to provide process improvement decision support based on the decision tree analysis results.

[0093] According to the method of this disclosure, using a decision tree algorithm to find key chambers and abnormal key operating parameters is an effective method that can solve the problem of yield decline caused by abnormal operating parameters. By carefully selecting features, handling data imbalance problems, applying pruning techniques, and performing model evaluation and optimization, an accurate and robust decision tree model can be constructed, providing strong support for the detection of machine tools and abnormal machine parameters.

[0094] According to the method of this disclosure, decision tree analysis helps engineers quickly identify key influencing factors through conditional segmentation and optimize processes. For example, by analyzing equipment and process parameters, the root causes of anomalies can be identified, improving the overall wafer yield. It can accurately identify key equipment chambers and parameters affecting wafer yield, which is difficult to achieve through traditional analysis methods. Optimizing key operating parameters can directly improve wafer yield, thereby increasing overall production output. Identifying and optimizing low-yield chambers (such as -O chambers) can reduce scrap rates, lower production costs, and improve resource utilization.

[0095] The method according to embodiments of this disclosure provides a systematic approach to evaluate and optimize process flows, particularly for the maintenance and improvement of low-yield chambers, contributing to improved efficiency and stability of the entire production line. It provides data-driven decision support for engineers and decision-makers, making process optimization and troubleshooting more scientific and efficient. Automated decision tree analysis tools enable rapid analysis of large amounts of data, improving the efficiency of problem analysis and resolution. It supports continuous process improvement by constantly analyzing new data to adjust and optimize production parameters, achieving continuous quality enhancement. Intuitive data visualization technology enhances the transparency of the production process, making it easier for relevant personnel to understand and monitor production status.

[0096] While existing technologies can detect mean differences between different groups (e.g., normal and abnormal machines) to identify abnormal machines, this method only provides a macroscopic perspective and cannot delve into the specific parameters of the machine. The results typically cannot directly pinpoint which specific parameters caused the machine's abnormality. This means that although abnormal machines can be identified, it is impossible to determine which parameter anomalies led to the overall performance degradation. The method provided in this disclosure, however, utilizes decision tree analysis to examine the chambers through which low-yield wafers pass and the operating parameters within those chambers that cause low yield. This allows for in-depth analysis at the specific parameter level, identifying the specific factors affecting yield. Furthermore, it provides an intuitive tree diagram that is easy to understand and interpret, enabling even non-experts to grasp the analysis results.

[0097] The method according to the embodiments of this disclosure achieves an innovation in the analysis method of commonalities of equipment and chambers. It uses a multi-level factor condition to split samples and specifically describes how to group according to different equipment chamber IDs and equipment parameters, thereby accurately identifying equipment or process steps that may lead to low yield.

[0098] The method according to embodiments of this disclosure specifically applies the decision tree algorithm to failure analysis in wafer manufacturing, rather than general decision tree algorithm applications. In particular, the splitting nodes (chamber ID or operating parameters) and discrimination criteria selected in the decision tree algorithm reflect the unique characteristics of its application in semiconductor process flows. That is, it innovatively realizes a unique technical path that utilizes splitting nodes to maximize the decision criteria for distinguishing between high-yield and low-yield wafers, and gradually narrows down the range of root causes of anomalies through these splitting processes. In other words, the influencing factor analysis implemented using the decision tree algorithm in embodiments of this disclosure has significant advantages in identifying abnormal equipment and specific equipment parameter problems, providing more in-depth and detailed analysis results, and helping to fundamentally solve the problem of yield decline.

[0099] Figure 3 This diagram shows a block diagram of a wafer yield influencing factor analysis apparatus according to an embodiment of the present disclosure, such as... Figure 3 As shown, the device includes:

[0100] The acquisition module 301 is used to acquire yield data and equipment data of a batch of wafers manufactured by a wafer manufacturing equipment. The wafer manufacturing equipment includes multiple chambers for manufacturing wafers. The yield data includes the yield of each wafer in the batch. The equipment data includes the chamber in which each wafer in the batch is manufactured and the set of operating parameters of the chamber. The set of operating parameters includes the parameter values ​​of multiple operating parameters when manufacturing wafers in the chamber.

[0101] The first determining module 302 is used to determine, based on the yield data of the batch wafers and the equipment data, the key chamber that produces low-yield wafers and the set of wafers manufactured in the key chamber, wherein the low-yield wafers are wafers with a yield lower than a first threshold.

[0102] The second determining module 303 is used to determine the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield based on the yield of each wafer in the wafer set manufactured in the key chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. The analysis results of the factors affecting the wafer yield include the key operating parameters in the key chamber that lead to low-yield wafers and the parameter thresholds corresponding to the key operating parameters. The parameter thresholds represent the parameter value limits that produce low-yield wafers.

[0103] In one possible implementation, determining the analysis results of the factors affecting wafer yield of the wafer manufacturing equipment based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: using a first analysis algorithm to determine a first analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, wherein the first analysis result includes a first operating parameter in the critical chamber that leads to low yield wafers and a parameter threshold corresponding to the first operating parameter; and using a second analysis algorithm to determine the factors affecting wafer yield based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer. The yield of the wafer and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer are used to determine a second analysis result. The second analysis result includes the second operating parameter in the critical chamber that leads to low-yield wafers and the degree of influence of the second operating parameter on the wafer yield. Based on the first analysis result and the second analysis result, the analysis result of the influencing factors of the wafer manufacturing equipment on the wafer yield is determined. The second operating parameter is the same as or partially the same as the first operating parameter. The analysis result of the influencing factors also includes the degree of influence of the critical operating parameter in the critical chamber that leads to low-yield wafers on the wafer yield and / or the result of ranking the critical operating parameters according to the degree of influence.

[0104] In one possible implementation, determining the second analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: determining the degree of influence of each of the multiple operating parameters on the wafer yield based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer; and determining a second operating parameter in the critical chamber that leads to low-yield wafers based on the degree of influence of each of the multiple operating parameters on the wafer yield.

[0105] In one possible implementation, determining the degree of influence of each of the multiple operating parameters on the wafer yield based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: determining a first parameter group corresponding to low-yield wafers and a second parameter group corresponding to high-yield wafers based on the yield of each wafer in the wafer set manufactured in the critical chamber, wherein the high-yield wafers are wafers with a yield higher than a second threshold, the second threshold being greater than the first threshold, and the first parameter group including the yield of each low-yield wafer manufactured in the critical chamber. The first parameter group contains the parameter values ​​of multiple operating parameters during high-yield wafer fabrication, and the second parameter group contains the parameter values ​​of multiple operating parameters during the fabrication of each high-yield wafer in the critical chamber. By comparing and analyzing the first parameter group and the second parameter group, the parameter difference corresponding to each of the multiple operating parameters is obtained. The parameter difference characterizes the degree of difference between the parameter values ​​of the operating parameters corresponding to high-yield wafers and the parameter values ​​of the operating parameters corresponding to low-yield wafers. The parameter difference corresponding to each of the multiple operating parameters is determined as the degree of influence of each of the multiple operating parameters on the wafer yield.

[0106] In one possible implementation, the first analysis algorithm includes a decision tree algorithm, and the second analysis algorithm includes a T-detection algorithm.

[0107] In one possible implementation, the apparatus further uses a decision tree algorithm to determine, based on the yield data of the batch of wafers and equipment data, the critical chambers that produce low-yield wafers and the set of wafers manufactured within the critical chambers; and, based on the yield of each wafer in the set of wafers manufactured within the critical chambers and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, to determine the analysis results of the factors affecting wafer manufacturing equipment on wafer yield.

[0108] In one possible implementation, the apparatus further includes: a generation module, configured to generate optimization suggestions for key chambers and / or key operating parameters within key chambers of the wafer manufacturing equipment based on the analysis results of factors affecting wafer yield of the wafer manufacturing equipment; and a push module, configured to generate an analysis report based on the analysis results of the factors affecting wafer yield and the optimization suggestions, and push the analysis report to relevant personnel so that the relevant personnel can optimize the key chambers and / or key operating parameters within key chambers of the wafer manufacturing equipment with the goal of improving wafer yield based on the analysis report.

[0109] According to the apparatus of this disclosure, by acquiring the yield data and equipment data of the wafers manufactured by the equipment, the key chambers that produce low-yield wafers are identified. Then, based on the yield of the wafers manufactured in the key chambers and the parameter values ​​of the operating parameters, the key operating parameters and corresponding parameter thresholds that cause low-yield wafers in the key chambers can be determined. Overall, the apparatus can quickly and accurately locate the key chambers, key operating parameters and corresponding parameter thresholds that affect the yield, thereby helping relevant personnel to locate equipment problems in a timely manner and effectively improve the yield of wafers manufactured by the equipment on the production line.

[0110] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0111] This disclosure also proposes a computer-readable storage medium storing computer program instructions that, when executed by a processor, implement the above-described method. The computer-readable storage medium can be volatile or non-volatile.

[0112] This disclosure also proposes an electronic device, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above method when executing the instructions stored in the memory.

[0113] This disclosure also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above-described method.

[0114] Figure 4 A block diagram of an electronic device 1900 according to an embodiment of the present disclosure is shown. For example, the electronic device 1900 may be provided as a server or a terminal device. (Refer to...) Figure 4 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0115] Electronic device 1900 may also include a power supply component 1926 configured to perform power management of electronic device 1900, a wired or wireless network interface 1950 configured to connect electronic device 1900 to a network, and an input / output interface 1958 (I / O interface). Electronic device 1900 can operate on an operating system, such as Windows Server, stored in memory 1932. TM Mac OS X TM Unix TM Linux TM FreeBSD TM Or similar.

[0116] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions that can be executed by a processing component 1922 of an electronic device 1900 to perform the above-described method.

[0117] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0118] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0119] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0120] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0121] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0122] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0123] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0124] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0125] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for analyzing factors affecting wafer yield, characterized in that, include: The process involves acquiring yield data and equipment data for a batch of wafers manufactured by a wafer manufacturing equipment. The wafer manufacturing equipment includes multiple chambers for manufacturing wafers. The yield data includes the yield of each wafer in the batch. The equipment data includes the chamber in which each wafer in the batch is manufactured and the set of operating parameters for the chamber. The set of operating parameters includes the parameter values ​​of multiple operating parameters when manufacturing wafers in the chamber. Based on the yield data of the batch wafers and the equipment data, the key chamber for producing low-yield wafers and the set of wafers manufactured in the key chamber are determined, wherein the low-yield wafers are wafers with a yield lower than a first threshold. Based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield are determined. The analysis results of the factors affecting the wafer yield include the key operating parameters in the critical chamber that lead to low-yield wafers and the parameter thresholds corresponding to the key operating parameters. The parameter thresholds represent the parameter value limits that produce low-yield wafers. The step of determining the influencing factors of the wafer manufacturing equipment on the wafer yield based on the yield of each wafer in the wafer assembly manufactured in the key chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: The first analysis algorithm is used to determine the first analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. The first analysis result includes the first operating parameter of the wafer in the critical chamber that leads to low yield and the parameter threshold corresponding to the first operating parameter. The second analysis algorithm is used to determine the second analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. The second analysis result includes the second operating parameters in the critical chamber that lead to low yield wafers and the degree of influence of the second operating parameters on the wafer yield. Based on the first analysis results and the second analysis results, the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield are determined; Wherein, the second operating parameter is the same as or partially the same as the first operating parameter, and the influencing factor analysis results also include the degree of influence of the key operating parameters that lead to low yield wafers in the key chamber on the wafer yield and / or the results of ranking the key operating parameters according to the degree of influence, the first analysis algorithm includes the decision tree algorithm, and the second analysis algorithm includes the T detection algorithm.

2. The method according to claim 1, characterized in that, The determination of the second analysis result based on the yield of each wafer in the wafer assembly manufactured within the key chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: Based on the yield of each wafer in the wafer set manufactured in the key chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer, determine the degree of influence of each of the multiple operating parameters on the wafer yield; Based on the degree of influence of each of the multiple operating parameters on wafer yield, a second operating parameter that leads to low-yield wafers in the critical chamber is determined.

3. The method according to claim 2, characterized in that, The step of determining the degree of influence of each of the multiple operating parameters on the wafer yield based on the yield of each wafer in the wafer assembly manufactured in the critical chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: Based on the yield of each wafer in the wafer set manufactured in the critical chamber, a first parameter group corresponding to low-yield wafers and a second parameter group corresponding to high-yield wafers are determined. The high-yield wafers are wafers with a yield higher than a second threshold, which is greater than the first threshold. The first parameter group includes parameter values ​​of multiple operating parameters when manufacturing each low-yield wafer in the critical chamber, and the second parameter group includes parameter values ​​of multiple operating parameters when manufacturing each high-yield wafer in the critical chamber. By comparing and analyzing the first parameter group and the second parameter group, the parameter difference corresponding to each of the multiple operating parameters is obtained. The parameter difference characterizes the degree of difference between the parameter values ​​of the operating parameters corresponding to high-yield wafers and the parameter values ​​of the operating parameters corresponding to low-yield wafers. The parameter differences among the plurality of operating parameters are determined as the degree of influence of each of the plurality of operating parameters on wafer yield.

4. The method according to claim 1, characterized in that, The method also uses a decision tree algorithm to determine the key chamber that produces low-yield wafers and the set of wafers manufactured in the key chamber based on the yield data and equipment data of the batch of wafers. Furthermore, it determines the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield based on the yield of each wafer in the set of wafers manufactured in the key chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: Based on the analysis results of the factors affecting wafer yield of the wafer manufacturing equipment, optimization suggestions are generated for key chambers and / or key operating parameters within the key chambers of the wafer manufacturing equipment. Based on the analysis results of the influencing factors and the optimization suggestions, an analysis report is generated and pushed to relevant personnel so that they can optimize the key chambers and / or key operating parameters within the key chambers of the wafer manufacturing equipment with the goal of improving wafer yield.

6. A device for analyzing factors affecting wafer yield, characterized in that, include: The acquisition module is used to acquire yield data and equipment data of a batch of wafers manufactured by a wafer manufacturing equipment. The wafer manufacturing equipment includes multiple chambers for manufacturing wafers. The yield data includes the yield of each wafer in the batch. The equipment data includes the chamber in which each wafer in the batch is manufactured and the set of operating parameters for the chamber. The set of operating parameters includes the parameter values ​​of multiple operating parameters when manufacturing wafers in the chamber. The first determining module is used to determine, based on the yield data of the batch wafers and the equipment data, the key chamber that produces low-yield wafers and the set of wafers manufactured in the key chamber, wherein the low-yield wafers are wafers with a yield lower than a first threshold. The second determining module is used to determine the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield based on the yield of each wafer in the wafer set manufactured in the key chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. The analysis results of the factors affecting the wafer yield include the key operating parameters in the key chamber that lead to low yield wafers and the parameter thresholds corresponding to the key operating parameters. The parameter thresholds represent the parameter value limits that produce low yield wafers. The step of determining the influencing factors of the wafer manufacturing equipment on the wafer yield based on the yield of each wafer in the wafer assembly manufactured in the key chamber and the parameter values ​​of multiple operating parameters during the manufacturing of each wafer includes: The first analysis algorithm is used to determine the first analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. The first analysis result includes the first operating parameter of the wafer in the critical chamber that leads to low yield and the parameter threshold corresponding to the first operating parameter. The second analysis algorithm is used to determine the second analysis result based on the yield of each wafer in the wafer set manufactured in the critical chamber and the parameter values ​​of multiple operating parameters when manufacturing each wafer. The second analysis result includes the second operating parameters in the critical chamber that lead to low yield wafers and the degree of influence of the second operating parameters on the wafer yield. Based on the first analysis results and the second analysis results, the analysis results of the factors affecting the wafer manufacturing equipment on the wafer yield are determined; Wherein, the second operating parameter is the same as or partially the same as the first operating parameter, and the influencing factor analysis results also include the degree of influence of the key operating parameters that lead to low yield wafers in the key chamber on the wafer yield and / or the results of ranking the key operating parameters according to the degree of influence, the first analysis algorithm includes the decision tree algorithm, and the second analysis algorithm includes the T detection algorithm.

7. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method of any one of claims 1 to 5 when executing instructions stored in the memory.

8. A non-volatile computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 5.

Images