A method and device for processing semiconductor production data

Abstract

The present application discloses a method and apparatus for processing semiconductor production data, which is used to generate and output a recommendation report of influencing factors corresponding to the test results of semiconductor test items, thereby narrowing the browsing scope of semiconductor production data, improving the speed of accessing data that affects the test results of semiconductor test items, and thus improving work efficiency. The method provided by the present application includes: generating a first recommendation report and / or a second recommendation report corresponding to the test results of the semiconductor test items; wherein the first recommendation report is a recommendation report generated based on a correlation coefficient, wherein the correlation coefficient refers to the correlation coefficient between the test results of the semiconductor test items and the reasons for generating the test results, wherein the reasons include semiconductor production data; the second recommendation report is a recommendation report generated based on the frequency of users querying semiconductor production data; and outputting the first recommendation report and / or the second recommendation report.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for processing semiconductor production data. Background Art

[0002] The semiconductor industry is highly automated and therefore highly data-driven, with large amounts of data recorded at different production and testing sites.

[0003] However, in the prior art, there is no fully automatic reporting system for automatically finding factors influencing the yield rate, and manual query and analysis of data is required. Summary of the Invention

[0004] An embodiment of the present application provides a method and device for processing semiconductor production data, which is used to generate and output a recommendation report on influencing factors corresponding to the test results of semiconductor test projects, thereby narrowing the browsing scope of semiconductor production data, improving the speed of reviewing data that affects the test results of semiconductor test projects, and thus improving work efficiency.

[0005] An embodiment of the present application provides a method for processing semiconductor production data, including:

[0006] generating a first recommendation report and / or a second recommendation report corresponding to the test results of the semiconductor test item; wherein the first recommendation report is a recommendation report generated based on a correlation coefficient, wherein the correlation coefficient refers to a correlation coefficient between the test results of the semiconductor test item and the reasons for the test results, wherein the reasons include semiconductor production data; and the second recommendation report is a recommendation report generated based on the frequency of user queries for semiconductor production data;

[0007] Output the first recommendation report and / or the second recommendation report.

[0008] The method generates and outputs a recommendation report of factors influencing the test results of semiconductor test items. The recommendation report includes a first recommendation report and / or a second recommendation report. The first recommendation report is generated based on a correlation coefficient between the test results of the semiconductor test items and the causes of the test results, wherein the causes include semiconductor production data. The second recommendation report is generated based on the frequency with which users query semiconductor production data. Therefore, by automatically generating and outputting the recommendation report to the user, the user's browsing scope of semiconductor production data is narrowed, the user's access speed to data affecting the test results of the semiconductor test items is increased, and work efficiency is thereby improved.

[0009] In some embodiments, the method further comprises:

[0010] An alarm message is issued when one or a combination of the following conditions is met:

[0011] The test results exceed preset specifications;

[0012] The rate of change of the test result compared to the previous test result is greater than a preset value;

[0013] The test results did not exceed the preset specifications.

[0014] In some implementations, when one or a combination of the conditions is met, the first recommendation report and / or the second recommendation report is output.

[0015] In some implementations, outputting the first recommendation report and / or the second recommendation report includes: outputting the first recommendation report and / or the second recommendation report to a preset knowledge base.

[0016] In some embodiments, the method further comprises:

[0017] Counting the frequency of users' queries for the first recommendation report and the frequency of users' queries for the second recommendation report;

[0018] According to the user's query frequency for the first recommendation report and the user's query frequency for the second recommendation report, the recommendation proportions of the first recommendation report and the second recommendation report in the knowledge base are changed.

[0019] In some embodiments, the method further comprises:

[0020] Based on the recommendation ratios of the changed first recommendation report and the second recommendation report in the knowledge base, a recommendation result is output to the user, wherein the recommendation result includes the monitoring result of the test item and at least one recommendation reason for generating the monitoring result, as well as correlation information between each recommendation reason and the monitoring result.

[0021] In some implementations, the monitoring results, the recommendation reasons, and the correlation information are respectively displayed via curve graphs.

[0022] In some embodiments, the method further comprises:

[0023] The users are grouped, and different groups of users correspond to different weighting coefficients, where the weighting coefficients are weighting coefficients used to generate the frequency of users querying semiconductor production data.

[0024] In some embodiments, users are pre-divided into three groups, namely, an expert user group, an ordinary user group, and a new user group, wherein the weighting coefficient corresponding to the expert user group is greater than the weighting coefficient corresponding to the ordinary user group, and the weighting coefficient corresponding to the ordinary user group is greater than the weighting coefficient corresponding to the new user group.

[0025] In some implementations, generating the first recommendation report includes:

[0026] Determining correlation coefficients between the test results and multiple categories of causes, and sorting the multiple categories of causes according to the correlation coefficients;

[0027] Multiple categories of reasons are selected from the sorting results to generate the first recommendation report, wherein the correlation coefficients corresponding to the selected reasons exceed a preset threshold.

[0028] In some implementations, generating the second recommendation report includes:

[0029] Based on the log records of users querying semiconductor production data, the user behaviors of querying semiconductor production data are divided into monitoring and exploration categories;

[0030] Perform frequency statistics on user behaviors of the monitoring category and the exploration category respectively;

[0031] Based on the statistical results of the frequency statistics of user behaviors of the monitoring type and the exploration type, the user query frequency of the semiconductor production data represented by the same identifier is counted to generate the user behavior frequency statistics corresponding to the test results of the semiconductor test project;

[0032] The user behavior frequency statistics results of the user behavior frequencies greater than the preset threshold are obtained to generate the second recommendation report.

[0033] An embodiment of the present application provides a semiconductor production data processing device, comprising:

[0034] a memory for storing program instructions;

[0035] The processor is configured to call the program instructions stored in the memory and execute any one of the above methods according to the obtained program.

[0036] An embodiment of the present application provides a semiconductor production data processing device, comprising:

[0037] a first unit configured to generate a first recommendation report and / or a second recommendation report corresponding to a test result of a semiconductor test item; wherein the first recommendation report is generated based on a correlation coefficient, wherein the correlation coefficient refers to a correlation coefficient between a test result of the semiconductor test item and a cause of the test result, wherein the cause includes semiconductor production data; and wherein the second recommendation report is generated based on a frequency of user queries for semiconductor production data;

[0038] The second unit is configured to output the first recommendation report and / or the second recommendation report.

[0039] Another embodiment of the present application provides a computing device, which includes a memory and a processor, wherein the memory is used to store program instructions, and the processor is used to call the program instructions stored in the memory and execute any of the above methods according to the obtained program.

[0040] Furthermore, according to an embodiment, a computer program product for a computer is provided, for example, comprising software code portions for executing the steps of the method defined above when the product is executed on the computer. The computer program product may include a computer-readable medium having the software code portions stored thereon. Furthermore, the computer program product may be directly loaded into the internal memory of the computer and / or transmitted via a network through at least one of an upload process, a download process, and a push process.

[0041] Another embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute any of the above methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0043] Figure 1 A schematic diagram of the overall architecture of the technical solution provided in the embodiments of this application;

[0044] Figure 2 A schematic diagram of the overall process of a method for processing semiconductor production data provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of an alarm condition provided in an embodiment of the present application;

[0046] Figure 4 A flowchart of a method for changing the recommendation ratios of a first recommendation report and a second recommendation report in a knowledge base provided in an embodiment of the present application;

[0047] Figure 5 A schematic diagram of monitoring results of test items provided in an embodiment of the present application;

[0048] Figure 6 A schematic diagram of a recommended reason for the monitoring result of a test item provided in an embodiment of the present application;

[0049] Figure 7 A schematic diagram of another recommended reason for the monitoring result of a test item provided in an embodiment of the present application;

[0050] Figure 8 A schematic diagram illustrating the correlation between the recommendation reasons and monitoring results provided in an embodiment of the present application;

[0051] Figure 9 A flowchart of a method for generating a second recommendation report provided in an embodiment of the present application;

[0052] Figure 10 A schematic diagram of a one-to-one frequency recording of user behavior provided in an embodiment of the present application;

[0053] Figure 11 A schematic diagram of one-to-one frequency recording collection for exploring user behavior provided in an embodiment of the present application;

[0054] Figure 12 A schematic diagram of a semiconductor production data processing device provided in an embodiment of the present application;

[0055] Figure 13 A schematic structural diagram of another device for processing semiconductor production data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0057] An embodiment of the present application provides a method and device for processing semiconductor production data, which is used to generate and output a recommendation report on influencing factors corresponding to the test results of semiconductor test projects, thereby narrowing the browsing scope of semiconductor production data, improving the speed of reviewing data that affects the test results of semiconductor test projects, and thus improving work efficiency.

[0058] Among them, the method and the device are based on the same application concept. Since the principles of solving problems by the method and the device are similar, the implementation of the device and the method can refer to each other, and the repeated parts will not be repeated.

[0059] The terms "first", "second", etc. (if any) in the specification and claims of the embodiments of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0060] The following examples and embodiments are to be understood as illustrative examples only. Although this specification may refer to "one," "an," or "some" examples or embodiments at several places, this does not mean that each such reference relates to the same example or embodiment, nor does it mean that the feature applies only to a single example or embodiment. Individual features of different embodiments may also be combined to provide further embodiments. Furthermore, terms such as "comprises" and "comprising" should be understood as not limiting the described embodiments to consisting only of those features already mentioned; such examples and embodiments may also include features, structures, units, modules, etc. that are not specifically mentioned.

[0061] The following describes in detail the various embodiments of the present application in conjunction with the accompanying drawings. It should be noted that the order in which the embodiments of the present application are presented only represents the order of the embodiments, and does not represent the advantages or disadvantages of the technical solutions provided by the embodiments.

[0062] The technical solutions provided in the embodiments of this application relate to the fields of integrated circuits and semiconductors, and mainly to the Yield Management System (YMS) and the Java programming language. Figure 1The embodiment of the present application uses machine learning to obtain the popularity between data and make recommendations. Based on user operation behavior, the system records the direction of the search and recommends that the system automatically narrows the data browsing range, thereby achieving high-speed data browsing. Furthermore, a software interface is provided to display the correlation between the test results and causes (semiconductor production data) of semiconductor test items and the connection with time trends; and further, real-time monitoring can be performed through a background monitoring program. When an anomaly is detected, an abnormal alarm message is generated, such as sending an alarm email, etc., to notify users such as yield and process engineers of the abnormal situation.

[0063] See also Figure 2 The present invention provides a method for processing semiconductor production data, which generally includes:

[0064] S101. Generate a first recommendation report and / or a second recommendation report corresponding to the test results of a semiconductor test item; wherein the first recommendation report is generated based on a correlation coefficient, wherein the correlation coefficient refers to a correlation coefficient between the test results of the semiconductor test item and the reasons for the test results, wherein the reasons include semiconductor production data; and the second recommendation report is generated based on the frequency of user queries for semiconductor production data.

[0065] In some implementations, the first recommendation report may be arranged in descending order of the correlation coefficients;

[0066] Similarly, in some implementations, the second recommendation report may be arranged in descending order of frequency.

[0067] S102: Output the first recommendation report and / or the second recommendation report.

[0068] The embodiments of this application take into account the highly automated and digitized nature of the semiconductor industry, recording a large amount of data at various production and testing sites. The current universal yield data acquisition platform has hundreds of active users daily, leaving behind a large number of query records and saved regularly executed tasks. This data allows for user behavior analysis. DRAM products are currently in the extensive R&D phase, requiring a significant amount of data exploration compared to typical logic circuits.

[0069] Therefore, the method described in the embodiment of the present application generates and outputs a recommendation report of factors affecting the test results of semiconductor test items, the recommendation report including a first recommendation report and / or a second recommendation report; wherein the first recommendation report is generated based on a correlation coefficient, wherein the correlation coefficient refers to the correlation coefficient between the test results of the semiconductor test items and the reasons for the test results, wherein the reasons include semiconductor production data; and the second recommendation report is generated based on the frequency with which users query semiconductor production data. Therefore, by automatically generating and outputting the recommendation report to the user, the user's browsing scope of semiconductor production data is narrowed, the user's browsing speed for data affecting the test results of the semiconductor test items is increased, and thus work efficiency is improved.

[0070] In some embodiments, the production management unit of semiconductor products is a batch, which is called a lot and named Lot-ID.

[0071] In some embodiments, data of the same type is stored in a large table, which can be called a data class, such as MET measurement, DEF defect, WAT chip acceptance test, CP probe test, etc.

[0072] In some implementations, the test and production steps of data in a data class are recorded as "STEP-ID", the test method is recorded as "Program", and the specific parameters are recorded as "Paramter".

[0073] In some implementations, when statistics are collected, the same data features include: ProdID, StepID, program_id, param_group, and param_id.

[0074] In some embodiments, the method further comprises:

[0075] An alarm message is issued when one or a combination of the following conditions is met:

[0076] The test results exceed preset specifications;

[0077] The rate of change of the test result compared to the previous test result is greater than a preset value;

[0078] The test results did not exceed the preset specifications.

[0079] That is to say, the embodiment of the present application provides a combined monitoring method of result data (early warning scheme combination):

[0080] The data currently recorded in the semiconductor production and yield fields includes: data type * product * step * procedure * parameter, which can reach hundreds of thousands of types. In the existing technology, specifications and upper and lower limits of specifications need to be manually defined. In the embodiment of this application, a "pre-warning scheme combination" is provided: a combination of p-value + cpmk + automatic specifications, which periodically monitors the result data. Among them:

[0081] Auto spec assumes the past 90 days of data are normal and then performs statistical analysis. It then performs outlier removal, averaging, and sigma analysis. The average value is used as the Spec-Target, and 6 sigma is used as the upper and lower specification limits (spec low and spec high).

[0082] The P-value and cpmk can be calculated using standard statistical methods, as shown in the following examples.

[0083] The early warning scheme combination sets the following three alarm conditions. When one or a combination of them is met, an alarm is triggered and the knowledge base is opened:

[0084] Condition 1: Auto spec. When the specification is exceeded, an alarm is issued and the knowledge base is triggered to open.

[0085] First, we retrieve the 90-day history data according to the five grouping conditions (Prod / step / program / paragroup / para id).

[0086] Then, remove gross errors (remove outliers).

[0087] Next, take the average AVG of the data in each set of conditions and use it as the spec target. Also take the standard deviation s corresponding to each set of conditions.

[0088] Finally, for each set of conditions, set AVG+3s=spec high (upper limit) and AVG-3s=spec low (lower limit).

[0089] That is, for each set of conditions, if the corresponding upper or lower limit is exceeded, that is, it is not within the upper or lower limit range, an alarm message will be triggered, the knowledge base will be opened, and a recommendation report will be output.

[0090] Condition 2: P-value (data change rate), that is, if the data changes greatly and exceeds the preset threshold, an alarm will be triggered.

[0091] First, in the data detection, the number of values ​​greater than the upper limit and the number of values ​​less than the lower limit are summed.

[0092] Next, find the number of data samples.

[0093] Finally, the effectiveness ratio is calculated, and data greater than 1 to 99.7% enters the alarm list, generating an alarm grouping condition list (5 conditions).

[0094] Condition three: Cpmk (measurement process accuracy), that is, if the Cpmk of the data does not exceed the specification, an alarm will also be issued.

[0095] First, calculate the Cpk of the data, for example, using the following formula 1:

[0096]

[0097] in:

[0098] Cpk = process capability index;

[0099] USL upper control limit;

[0100] LSL: Lower control limit;

[0101] μ: mean value of data;

[0102] σ (Sigma): standard deviation.

[0103] Then, calculate Cpmk, for example, using the following formula 2:

[0104]

[0105] in:

[0106] T: The difference between the upper control limit and the lower control limit.

[0107] For example, Figure 3 As shown, referring to the Six Sigma theory, the probability of data being outside 3 sigma is 1-99.73%; the probability of data being outside 2 sigma is 1-95.45%; and the probability of data being outside 1 sigma is 1-68.27%. To test the data distribution, the data must first conform to a normal distribution.

[0108] In some implementations, the P-value and Cpmk of newly added data are monitored daily, and a list of alarm data is generated upon a trigger, i.e., a combination of early warnings.

[0109] In some implementations, when one or a combination of the conditions is met, the first recommendation report and / or the second recommendation report is output.

[0110] In some implementations, outputting the first recommendation report and / or the second recommendation report includes: outputting the first recommendation report and / or the second recommendation report to a preset knowledge base.

[0111] That is, in some embodiments, when one or a combination of the above conditions is met, an alarm may be triggered, and a recommendation report may be output to a knowledge base (preset database) for the user to review and analyze the specific reasons.

[0112] In some embodiments, see Figure 4 , the method further comprises:

[0113] Counting the frequency of users' queries for the first recommendation report and the frequency of users' queries for the second recommendation report;

[0114] According to the user's query frequency for the first recommendation report and the user's query frequency for the second recommendation report, the recommendation proportions of the first recommendation report and the second recommendation report in the knowledge base are changed.

[0115] See also Figure 4 , where ABS-R-Ranking generates a recommendation report based on the correlation coefficient; User Behavior Frequency Analysis generates a recommendation report based on user query frequency;

[0116] A causal relationship knowledge base, i.e., the preset knowledge base, is opened for user query when the alarm condition is met;

[0117] Of the two recommendation reports, the one that users click on more often will have a larger recommendation share, and the recommendation report with a larger share will eventually be displayed to users.

[0118] For example, the recommendation report based on the correlation coefficient (first recommendation report) and the recommendation report based on the user query frequency (second recommendation report) initially have a recommendation ratio of 50% each. However, in actual application, the user click rate of the recommendation report based on the correlation coefficient is high, so the recommendation ratio of the recommendation report based on the correlation coefficient is gradually increased. For example, the recommendation ratio of the recommendation report based on the correlation coefficient is 60%, while the recommendation ratio of the recommendation report based on the user query frequency is adjusted to 40%.

[0119] In some embodiments, see Figure 4 , the method further comprises:

[0120] Based on the recommendation ratios of the changed first recommendation report and the second recommendation report in the knowledge base, a recommendation result is output to the user, wherein the recommendation result includes the monitoring result of the test item and at least one recommendation reason for generating the monitoring result, as well as correlation information between each recommendation reason and the monitoring result.

[0121] In some embodiments, the monitoring results, the recommendation reasons, and the correlation information are displayed in curve graphs. Figure 5 As shown (the horizontal axis is time, the vertical axis is monitoring result data), Figure 5 The trend of monitoring results of a test project in the past 5 days is shown. The sampling period is 7-6 days. There are two recommended reasons for this monitoring result, which are: Figure 6 The recommended reasons 1 and Figure 7 The recommended reason 2 is shown; wherein, the monitoring result is the monitoring result of a certain test item (Y), and the recommended reason 1 and the recommended reason 2 are the two recommended reasons leading to the monitoring result of Y. Of course, there can be only one reason or more reasons.

[0122] in, Figure 6 and Figure 7 The horizontal axis is time, and the vertical axis is the recommended reason data, which shows different MET data in production. Figure 6 and Figure 7 It can be seen that the time is earlier than Figure 5 The time shown is not arranged in chronological order, but in Figure 5 The monitoring results are arranged in the order of lot ID.

[0123] The correlation information is as follows: Figure 8 As shown, the horizontal axis is X, which represents production data; the vertical axis is Y, which represents test items. Figure 8 It can be seen that the curve corresponding to result 1 and cause 2 indicates that the recommended reason 2 is strongly correlated with the monitoring result 1 (because the curve is more linear), while the curve corresponding to result 1 and cause 1 indicates that the correlation between the recommended reason 1 and the monitoring result 1 is not high.

[0124] In some embodiments, the method further comprises:

[0125] The users are grouped, and different groups of users correspond to different weighting coefficients, where the weighting coefficients are weighting coefficients used to generate the frequency of users querying semiconductor production data.

[0126] In some embodiments, users are pre-divided into three groups, namely, an expert user group, an ordinary user group, and a new user group, wherein the weighting coefficient corresponding to the expert user group is greater than the weighting coefficient corresponding to the ordinary user group, and the weighting coefficient corresponding to the ordinary user group is greater than the weighting coefficient corresponding to the new user group.

[0127] For example, users are divided into the following three groups, and the weighting coefficients of users in different groups are different:

[0128] Expert user group: Grant 5 times the learning intensity to the principal level by reviewing personnel information;

[0129] Normal user group: Senior engineers are granted double learning intensity through personnel database.

[0130] New user group: New engineers are assigned 1x learning intensity based on the personnel database.

[0131] The learning process of user query frequency for semiconductor production data is reinforced through intensity. For example, if an expert user queries once, the corresponding weighting factor is 5, which is equivalent to 5 queries; if an ordinary user queries once, the corresponding weighting factor is 2, which is equivalent to 2 queries; and if a new user queries once, the corresponding weighting factor is 1, which is equivalent to 1 query.

[0132] In some implementations, generating the first recommendation report includes:

[0133] Determining correlation coefficients between the test results and multiple categories of causes, and sorting the multiple categories of causes according to the correlation coefficients;

[0134] Multiple categories of reasons are selected from the sorting results to generate the first recommendation report, wherein the correlation coefficients corresponding to the selected reasons exceed a preset threshold.

[0135] That is to say, an automatic recommendation method for causal data provided in the embodiments of the present application is as follows:

[0136] causal classification and temporal ordering of data;

[0137] Perform a reverse wide table combination traversal on the historical 180-day data to calculate the ABS-R-Rank;

[0138] For each result class, a list of the top five correlation coefficients (R) with absolute values ​​greater than 0.5 is saved as a source of recommendation.

[0139] Therefore, when counting, the same data features are: ProductID, StepID, program_id, param_group, parammeter_id, which are connected as a character and expressed as a "key metadata chain", abbreviated as: keymate. For example, the standard format of the data chain ID is as follows:

[0140] AAAA|AAAA.0000.00000|AAAA|AAAA00|AAAA

[0141] The first AAAA is the ProductID; AAAA.0000.00000 is the StepID; the second AAAA is the programID; AAAA00 is the parameter-group; and the third AAAA is the parameter id.

[0142] The data categories used in the embodiments of the present application are as follows: cause category MET, DEF, FDC result category: WAT CP.

[0143] For historical data, find the temporary table (Y.lot id = X.lot id) and run a loop based on the corresponding relationship. Where Y represents the test project and X represents the production data.

[0144] Use machine learning to get the ranking of R and take the absolute value. Code command (code in AI, used to regress R, R is the score) for example:

[0145] scores=mutual_info_regression(x,y,random_state=0)

[0146] scores=pd.Series(scores,name="R-Ranking",index=x.columns)

[0147] In some embodiments, see Figure 9 Generating the second recommendation report includes:

[0148] Based on the log records of users querying semiconductor production data, the user behaviors of querying semiconductor production data are divided into monitoring and exploration categories;

[0149] Perform frequency statistics on user behaviors of the monitoring category and the exploration category respectively;

[0150] According to the statistical results of the frequency statistics of the user behaviors of the monitoring type and the exploration type, the user query frequency of the semiconductor production data represented by the same identifier (ID) is counted, for example Figure 10 As shown, the user behavior frequency statistics corresponding to the test results of the semiconductor test project are generated;

[0151] The user behavior frequency statistics results of the user behavior frequencies greater than the preset threshold are obtained to generate the second recommendation report.

[0152] In the embodiment of the present application, user data behaviors are divided into: monitoring and exploration.

[0153] Monitoring is the continuous acquisition of data with a clear objective in mind, used to identify trends;

[0154] Exploration is based on a specific yield target to a wider range of data, making a one-to-many (i.e. 1:N, N is greater than 1) data connection, and then looking for high correlation. Many users' broader query behaviors have a lot of overlapping relationships, and the record frequency can be analyzed, such as Figure 11 The embodiment of the present application extracts regularity from these behavior records as a source of recommendation.

[0155] One-to-one frequency recording (i.e. recording frequency) formula is as follows:

[0156] Single-point to single-point frequency (Singal point to singal point Freq) = Acount*weight1+Bcount*weight2;

[0157] Among them, A count represents the frequency statistics of monitoring user data behavior, and weight1 represents the weight corresponding to monitoring user data behavior; B count represents the frequency statistics of exploration user data behavior, and weight2 represents the weight corresponding to exploration user data behavior.

[0158] The following is an introduction to the equipment or device provided in the embodiments of the present application, in which the explanations or examples of technical features that are the same as or corresponding to those described in the above method are not repeated hereafter.

[0159] See also Figure 12 , an embodiment of the present application provides a semiconductor production data processing device, comprising:

[0160] The processor 600 is configured to read the program in the memory 620 and execute the following process:

[0161] generating a first recommendation report and / or a second recommendation report corresponding to the test results of the semiconductor test item; wherein the first recommendation report is a recommendation report generated based on a correlation coefficient, wherein the correlation coefficient refers to a correlation coefficient between the test results of the semiconductor test item and the reasons for the test results, wherein the reasons include semiconductor production data; and the second recommendation report is a recommendation report generated based on the frequency of user queries for semiconductor production data;

[0162] Output the first recommendation report and / or the second recommendation report.

[0163] In some implementations, the processor 600 is further configured to read a program in the memory 620 and execute the following process:

[0164] An alarm message is issued when one or a combination of the following conditions is met:

[0165] The test results exceed preset specifications;

[0166] The rate of change of the test result compared to the previous test result is greater than a preset value;

[0167] The test results did not exceed the preset specifications.

[0168] In some implementations, when one or a combination of the conditions is met, the first recommendation report and / or the second recommendation report is output.

[0169] In some implementations, outputting the first recommendation report and / or the second recommendation report includes: outputting the first recommendation report and / or the second recommendation report to a preset knowledge base.

[0170] In some implementations, the processor 600 is further configured to read a program in the memory 620 and execute the following process:

[0171] Counting the frequency of users' queries for the first recommendation report and the frequency of users' queries for the second recommendation report;

[0172] According to the user's query frequency for the first recommendation report and the user's query frequency for the second recommendation report, the recommendation proportions of the first recommendation report and the second recommendation report in the knowledge base are changed.

[0173] In some implementations, the processor 600 is further configured to read a program in the memory 620 and execute the following process:

[0174] Based on the recommendation ratios of the changed first recommendation report and the second recommendation report in the knowledge base, a recommendation result is output to the user, wherein the recommendation result includes the monitoring result of the test item and at least one recommendation reason for generating the monitoring result, as well as correlation information between each recommendation reason and the monitoring result.

[0175] In some implementations, the monitoring results, the recommendation reasons, and the correlation information are respectively displayed via curve graphs.

[0176] In some implementations, the processor 600 is further configured to read a program in the memory 620 and execute the following process:

[0177] The users are grouped, and different groups of users correspond to different weighting coefficients, where the weighting coefficients are weighting coefficients used to generate the frequency of users querying semiconductor production data.

[0178] In some embodiments, users are pre-divided into three groups, namely, an expert user group, an ordinary user group, and a new user group, wherein the weighting coefficient corresponding to the expert user group is greater than the weighting coefficient corresponding to the ordinary user group, and the weighting coefficient corresponding to the ordinary user group is greater than the weighting coefficient corresponding to the new user group.

[0179] In some implementations, generating the first recommendation report includes:

[0180] Determining correlation coefficients between the test results and multiple categories of causes, and sorting the multiple categories of causes according to the correlation coefficients;

[0181] Multiple categories of reasons are selected from the sorting results to generate the first recommendation report, wherein the correlation coefficients corresponding to the selected reasons exceed a preset threshold.

[0182] In some implementations, generating the second recommendation report includes:

[0183] Based on the log records of users querying semiconductor production data, the user behaviors of querying semiconductor production data are divided into monitoring and exploration categories;

[0184] Perform frequency statistics on user behaviors of the monitoring category and the exploration category respectively;

[0185] Based on the statistical results of the frequency statistics of user behaviors of the monitoring type and the exploration type, the user query frequency of the semiconductor production data represented by the same identifier is counted to generate the user behavior frequency statistics corresponding to the test results of the semiconductor test project;

[0186] The user behavior frequency statistics results of the user behavior frequencies greater than the preset threshold are obtained to generate the second recommendation report.

[0187] In some embodiments, the semiconductor production data processing apparatus further includes a transceiver 610 for receiving and sending data under the control of the processor 600 .

[0188] Among them, Figure 12 In the embodiment of the present invention, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits such as one or more processors represented by processor 600 and memory represented by memory 620. The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore not further described herein. The bus interface provides an interface. The transceiver 610 can be multiple components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices over a transmission medium.

[0189] In some embodiments, the semiconductor production data processing device further includes a user interface 630, which may be an interface capable of connecting external or internal devices. The connected devices include but are not limited to a keypad, display, speaker, microphone, joystick, etc.

[0190] The processor 600 is responsible for managing the bus architecture and general processing, and the memory 620 can store data used by the processor 600 when performing operations.

[0191] In some embodiments, the processor 600 may be a CPU (central processing unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a CPLD (Complex Programmable Logic Device).

[0192] See also Figure 13 Another semiconductor production data processing device provided by an embodiment of the present application includes:

[0193] A first unit 11 is configured to generate a first recommendation report and / or a second recommendation report corresponding to a test result of a semiconductor test item; wherein the first recommendation report is generated based on a correlation coefficient, wherein the correlation coefficient refers to a correlation coefficient between a test result of the semiconductor test item and a cause of the test result, wherein the cause includes semiconductor production data; and the second recommendation report is generated based on a frequency of user queries for semiconductor production data.

[0194] The second unit 12 is configured to output the first recommendation report and / or the second recommendation report.

[0195] In some embodiments, the first unit 11 is further configured to:

[0196] An alarm message is issued when one or a combination of the following conditions is met:

[0197] The test results exceed preset specifications;

[0198] The rate of change of the test result compared to the previous test result is greater than a preset value;

[0199] The test results did not exceed the preset specifications.

[0200] In some implementations, when one or a combination of the conditions is met, the second unit 12 outputs the first recommendation report and / or the second recommendation report.

[0201] In some implementations, outputting the first recommendation report and / or the second recommendation report includes: outputting the first recommendation report and / or the second recommendation report to a preset knowledge base.

[0202] In some embodiments, the second unit 12 is further configured to:

[0203] Counting the frequency of users' queries for the first recommendation report and the frequency of users' queries for the second recommendation report;

[0204] According to the user's query frequency for the first recommendation report and the user's query frequency for the second recommendation report, the recommendation proportions of the first recommendation report and the second recommendation report in the knowledge base are changed.

[0205] In some embodiments, the second unit 12 is further configured to:

[0206] Based on the recommendation ratios of the changed first recommendation report and the second recommendation report in the knowledge base, a recommendation result is output to the user, wherein the recommendation result includes the monitoring result of the test item and at least one recommendation reason for generating the monitoring result, as well as correlation information between each recommendation reason and the monitoring result.

[0207] In some implementations, the monitoring results, the recommendation reasons, and the correlation information are respectively displayed via curve graphs.

[0208] In some embodiments, the first unit 11 is further configured to:

[0209] The users are grouped, and different groups of users correspond to different weighting coefficients, where the weighting coefficients are weighting coefficients used to generate the frequency of users querying semiconductor production data.

[0210] In some embodiments, users are pre-divided into three groups, namely, an expert user group, an ordinary user group, and a new user group, wherein the weighting coefficient corresponding to the expert user group is greater than the weighting coefficient corresponding to the ordinary user group, and the weighting coefficient corresponding to the ordinary user group is greater than the weighting coefficient corresponding to the new user group.

[0211] In some implementations, generating the first recommendation report includes:

[0212] Determining correlation coefficients between the test results and multiple categories of causes, and sorting the multiple categories of causes according to the correlation coefficients;

[0213] Multiple categories of reasons are selected from the sorting results to generate the first recommendation report, wherein the correlation coefficients corresponding to the selected reasons exceed a preset threshold.

[0214] In some implementations, generating the second recommendation report includes:

[0215] Based on the log records of users querying semiconductor production data, the user behaviors of querying semiconductor production data are divided into monitoring and exploration categories;

[0216] Perform frequency statistics on user behaviors of the monitoring category and the exploration category respectively;

[0217] Based on the statistical results of the frequency statistics of user behaviors of the monitoring type and the exploration type, the user query frequency of the semiconductor production data represented by the same identifier is counted to generate the user behavior frequency statistics corresponding to the test results of the semiconductor test project;

[0218] The user behavior frequency statistics results of the user behavior frequencies greater than the preset threshold are obtained to generate the second recommendation report.

[0219] It should be noted that the division of units in the embodiments of the present application is schematic and is merely a logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0220] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0221] An embodiment of the present application provides a computing device, which may specifically be a desktop computer, a portable computer, a smart phone, a tablet computer, a personal digital assistant (PDA), etc. The computing device may include a central processing unit (CPU), a memory, input / output devices, etc. The input devices may include a keyboard, a mouse, a touch screen, etc., and the output devices may include a display device such as a liquid crystal display (LCD) or a cathode ray tube (CRT).

[0222] The memory may include a read-only memory (ROM) and a random access memory (RAM), and provides program instructions and data stored in the memory to the processor. In an embodiment of the present application, the memory may be used to store the program of any of the methods provided in the embodiments of the present application.

[0223] The processor calls the program instructions stored in the memory, and the processor is used to execute any of the methods provided in the embodiments of the present application according to the obtained program instructions.

[0224] The present application embodiment also provides a computer program product or computer program, which includes computer instructions, which are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions so that the computer device performs any of the methods described in the above embodiments. The program product can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0225] The present invention provides a computer-readable storage medium for storing computer program instructions used by the apparatus provided in the above embodiments of the present invention, which includes a program for executing any of the methods provided in the above embodiments of the present invention. The computer-readable storage medium may be a non-transitory computer-readable medium.

[0226] The computer-readable storage medium can be any available medium or data storage device that can be accessed by a computer, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (such as ROMs, EPROMs, EEPROMs, non-volatile memories (NANDFLASH), solid-state drives (SSDs)), etc.

[0227] It should be understood that:

[0228] The access technology through which entities in the communication network transmit traffic can be any suitable current or future technology, such as WLAN (Wireless Local Access Network), WiMAX (Worldwide Interoperability for Microwave Access), LTE, LTE-A, 5G, Bluetooth, infrared, etc.; in addition, the embodiments can also apply wired technology, for example, IP-based access technology, such as a wired network or a fixed line.

[0229] Embodiments suitable for being implemented as software code or a portion thereof and run using a processor or processing functionality are independent of the software code and may be specified using any known or future developed programming language, such as a high-level programming language such as objective-C, C, C++, C#, Java, Python, Javascript, other scripting languages, etc., or a low-level programming language such as machine language or assembler.

[0230] The implementation of the embodiments is hardware independent and may be implemented using any known or future developed hardware technology or any mixture thereof, such as a microprocessor or CPU (Central Processing Unit), MOS (Metal Oxide Semiconductor), CMOS (Complementary MOS), BiMOS (Bipolar MOS), BiCMOS (Bipolar CMOS), ECL (Emitter Coupled Logic) and / or TTL (Transistor-Transistor Logic).

[0231] Embodiments may be implemented as separate devices, apparatuses, units, components or functions, or in a distributed manner, for example, one or more processors or processing functions may be used or shared in a process, or one or more processing segments or processing portions may be used and shared in a process, where one physical processor or more than one physical processor may be used to implement one or more processing portions dedicated to a particular process as described.

[0232] The apparatus may be implemented by a semiconductor chip, a chipset, or a (hardware) module including such a chip or chipset.

[0233] The embodiments may also be implemented as any combination of hardware and software, such as ASIC (Application Specific IC (Integrated Circuit)) components, FPGA (Field Programmable Gate Array) or CPLD (Complex Programmable Logic Device) components or DSP (Digital Signal Processor) components.

[0234] The embodiments may also be implemented as a computer program product including a computer usable medium having computer readable program code embodied therein, the computer readable program code being adapted to perform the processes as described in the embodiments, wherein the computer usable medium may be a non-transitory medium.

[0235] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) that contain computer-usable program code.

[0236] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0237] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0238] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0239] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A method for processing semiconductor production data, characterized in that: The method comprises: generating a first recommendation report and a second recommendation report corresponding to a test result of a semiconductor test item; wherein the first recommendation report is generated based on a correlation coefficient, wherein the correlation coefficient refers to a correlation coefficient between the test result of the semiconductor test item and a cause of the test result, wherein the cause includes semiconductor production data; and the second recommendation report is generated based on a frequency of user queries for semiconductor production data; outputting the first recommendation report and the second recommendation report; Generating the first recommendation report includes: Determining correlation coefficients between the test results and multiple categories of causes, and sorting the multiple categories of causes according to the correlation coefficients; Selecting multiple categories of reasons from the sorting results to generate the first recommendation report, wherein the correlation coefficients corresponding to the selected reasons exceed a preset threshold; Generating the second recommendation report includes: Based on the log records of users querying semiconductor production data, the user behaviors of querying semiconductor production data are divided into monitoring and exploration categories; Perform frequency statistics on user behaviors of the monitoring category and the exploration category respectively; Based on the statistical results of the frequency statistics of user behaviors of the monitoring type and the exploration type, the user query frequency of the semiconductor production data represented by the same identifier is counted to generate the user behavior frequency statistics corresponding to the test results of the semiconductor test project; The user behavior frequency statistics results of the user behavior frequencies greater than the preset threshold are obtained to generate the second recommendation report.

2. The method according to claim 1, characterized in that The method further comprises: An alarm message is issued when one or a combination of the following conditions is met: The test results exceed preset specifications; The rate of change of the test result compared to the previous test result is greater than a preset value; The Cpmk value of the test result is between the preset upper control limit and the preset lower control limit; The value of Cpmk is determined by the following formula: in: Cpk represents the process capability index; USL represents the preset upper control limit; LSL represents the preset lower control limit; μ represents the average value of the semiconductor production data; σ represents the preset standard deviation; T represents the difference between the preset upper control limit and the lower control limit; Cpmk represents the measurement process accuracy.

3. The method according to claim 2, characterized in that When one or a combination of the conditions is satisfied, the first recommendation report and the second recommendation report are output.

4. The method according to claim 1, wherein The outputting the first recommendation report and the second recommendation report includes: outputting the first recommendation report and the second recommendation report to a preset knowledge base.

5. The method according to claim 4, characterized in that The method further comprises: Counting the frequency of users' queries for the first recommendation report and the frequency of users' queries for the second recommendation report; According to the user's query frequency for the first recommendation report and the user's query frequency for the second recommendation report, the recommendation proportions of the first recommendation report and the second recommendation report in the knowledge base are changed.

6. The method according to claim 5, characterized in that The method further comprises: Based on the recommendation ratios of the changed first recommendation report and the second recommendation report in the knowledge base, a recommendation result is output to the user, wherein the recommendation result includes the monitoring result of the test item and at least one recommendation reason for generating the monitoring result, as well as correlation information between each recommendation reason and the monitoring result.

7. The method according to claim 6, characterized in that The monitoring results, the recommendation reasons, and the correlation information are respectively displayed through curve graphs.

8. The method according to claim 1, characterized in that The method further comprises: The users are grouped, and different groups of users correspond to different weighting coefficients, where the weighting coefficients are weighting coefficients used to generate the frequency of users querying semiconductor production data.

9. The method according to claim 8, characterized in that Users are pre-divided into three groups, namely, expert user group, ordinary user group, and new user group. The weighting coefficient corresponding to the expert user group is greater than that corresponding to the ordinary user group, and the weighting coefficient corresponding to the ordinary user group is greater than that corresponding to the new user group.

10. The method according to claim 2, characterized in that The test results that exceeded the preset specifications include: The automatic specification of the test result exceeds a preset upper limit or a preset lower limit.

11. A device for processing semiconductor production data, characterized in that: include: a memory for storing program instructions; A processor, configured to call the program instructions stored in the memory and execute the method according to any one of claims 1 to 10 according to the obtained program.

12. A device for processing semiconductor production data, characterized in that: include: a first unit configured to generate a first recommendation report and a second recommendation report corresponding to a test result of a semiconductor test item; wherein the first recommendation report is generated based on a correlation coefficient, wherein the correlation coefficient refers to a correlation coefficient between the test result of the semiconductor test item and a cause of the test result, wherein the cause includes semiconductor production data; and the second recommendation report is generated based on a frequency of user queries for semiconductor production data; A second unit is configured to output the first recommendation report and the second recommendation report; Generating the first recommendation report includes: Determining correlation coefficients between the test results and multiple categories of causes, and sorting the multiple categories of causes according to the correlation coefficients; Selecting multiple categories of reasons from the sorting results to generate the first recommendation report, wherein the correlation coefficients corresponding to the selected reasons exceed a preset threshold; Generating the second recommendation report includes: Based on the log records of users querying semiconductor production data, the user behaviors of querying semiconductor production data are divided into monitoring and exploration categories; Perform frequency statistics on user behaviors of the monitoring category and the exploration category respectively; Based on the statistical results of the frequency statistics of user behaviors of the monitoring type and the exploration type, the user query frequency of the semiconductor production data represented by the same identifier is counted to generate the user behavior frequency statistics corresponding to the test results of the semiconductor test project; The user behavior frequency statistics results of the user behavior frequencies greater than the preset threshold are obtained to generate the second recommendation report.

13. A computer program product for a computer, characterized in that The invention comprises a software code portion for executing the method according to any one of claims 1 to 10 when the product is run on the computer.

14. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute the method according to any one of claims 1 to 10.

Images