JP2003222653A - Measurement program production support system and support program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To produce a measurement program proper for each measurement device with quality similar to an expert even without knowledge and experiences in a short time in a production step of the measurement program used for operating a semiconductor measurement device. <P>SOLUTION: The system comprises a measurement program database 31 containing a measurement program, a measured result database 32 containing the measured results in tests conducted with a semiconductor measuring device, a measured result analyzer 42 analyzing the measured results, an analyzed result database 33 containing the analyzed results, an analyzed result judging part 43 judging whether the analyzed results fulfill a specific standard, a knowledge data base 34 containing response methods of measurement program correction based on past analysis results, a correction candidate production part 44 surveying the knowledge database 34 based on the judged results with the analyzed result judging part 43, obtaining the corresponding response methods and producing correction candidates of the measurement program, and an output device 2 outputting the correction candidates. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device and a program for supporting the creation of a measurement program used for operating a semiconductor measuring device.

[0002]

2. Description of the Related Art A semiconductor measuring device is used for inspecting a semiconductor product in a manufacturing process of the semiconductor product. The semiconductor measuring device operates according to a preset measurement program and measures the electrical characteristics and mechanical characteristics of semiconductor products. Conventionally, the process from creation to introduction of a measurement program used to operate such a semiconductor measuring device is performed according to the flow shown in FIG.

As shown in FIG. 15, the measurement program is
After being created and debugged, each test measurement and verification is actually performed in the semiconductor measuring device. The part surrounded by the dotted line frame shown in FIG. 15 is a step in which three-step test measurement and analysis verification are performed, and the measurement program is corrected based on the analysis result. By executing these steps, the measurement program and each measuring device are adjusted, and the subtle error of each measuring device is adjusted.

[0004]

Conventionally, a measurement engineer analyzes a measurement data and considers a program correction point based on the analysis result. The measurement engineer finds the correction points of the program by comparing the analysis result with past experience. Therefore, it takes a lot of labor and time to create the measurement program, and it relies on human labor, so human error is likely to occur. Furthermore, there is a problem that the quality such as reliability and execution time of the measurement program is greatly influenced by the knowledge and experience of the measurement engineer.

Therefore, according to the present invention, at the stage of creating a measurement program used for operating a semiconductor measuring apparatus, a person who has a short knowledge of a measuring program suitable for each measuring apparatus and has little knowledge or experience can be provided. Another object of the present invention is to provide a measurement program creation support device and a measurement program creation support program that can be created with the same quality as an expert.

[0006]

A measurement program creation support apparatus of the present invention comprises a measurement program database for storing a measurement program used for operating a semiconductor measurement apparatus, and a test by the semiconductor measurement apparatus using the measurement program. A measurement result database that stores the measurement results obtained, a measurement result analysis unit that analyzes the measurement results, an analysis result database that stores the analysis results by the measurement result analysis unit, and the analysis results satisfy a predetermined standard. An analysis result determination unit that determines whether or not, a knowledge database that stores a coping method for correcting a measurement program based on past analysis results, and a search of the knowledge database based on the determination result by the analysis result determination unit The same as the revision candidate creation unit that obtains the coping method and creates revision candidates for the measurement program. It is obtained and an output section for outputting a correction candidates created by the positive candidate creating unit.

According to the present invention, the measurement result which is tested by the semiconductor measuring apparatus is analyzed by using the measurement program being prepared, and the appropriate coping method for the correction of the measurement program is selected based on the past analysis result. Correction candidates for the measurement program are automatically created and output. The user can correct the measurement program based on the output correction candidate of the measurement program.

Further, the measurement program creation support program of the present invention acquires the measurement result from a measurement result database that stores the measurement result of the test performed using the measurement program used to operate the semiconductor measuring device. Analysis step, a step of determining whether or not the analysis result satisfies a predetermined standard, and a knowledge database that stores a coping method of the measurement program correction based on the past analysis result is searched based on the determination result. Then, the computer is made to execute the step of obtaining the corresponding coping method and creating the correction candidate of the measurement program, and the step of outputting the created correction candidate.

The measurement program creation support program of the present invention is supplied to a computer via a computer-readable recording medium such as a hard disk, a CD-ROM, a flexible disk or a magneto-optical disk, or
Alternatively, by recording in a memory in a computer in advance and executing the program by the computer, the same operation as that of the measurement program creation support device of the present invention can be obtained.

[0010]

1 is a block diagram showing the configuration of a measurement program creation support device according to an embodiment of the present invention. In FIG. 1, a measurement program creation support device according to an embodiment of the present invention mainly includes an input device 1 such as a keyboard and a pointing device, an output device 2 such as a display and a printer, a storage device 3 such as a memory and a hard disk, and , A central processing unit 4 that controls each of the devices 1 to 3 and performs arithmetic processing.

The storage device 3 stores a measurement program database 31 for storing a measurement program used for operating a semiconductor measuring device (not shown) for measuring electrical characteristics and mechanical characteristics of semiconductor products, and the measurement program database 31. A measurement result database 32 that stores a measurement result of a test performed by the semiconductor measuring apparatus using the measurement program stored in the program database 31, and an analysis result database 33 that stores an analysis result by a measurement result analysis unit 41 described later, A knowledge database 34 that stores the coping method of the measurement program modification based on the past analysis result is constructed.

The central processing unit 4 uses a measurement program stored in the measurement program database 31 to perform a measurement by the semiconductor measurement device, and a measurement result analysis unit 4 which analyzes the measurement result by the test measurement unit 41.
2, the analysis result determination unit 43 that determines whether or not the analysis result by the measurement result analysis unit 42 satisfies a predetermined standard, and the knowledge database 34 is searched based on the determination result by the analysis result determination unit 43, and the corresponding It functions as a correction candidate creation unit 44 that acquires a coping method and creates a correction candidate for the measurement program.

The measurement result analysis section 42 has a correlation analysis function for analyzing the correlation of the measurement results tested by the test measurement section 41 using the reference measurement program and the measurement program being prepared, respectively. The test measurement unit 41 performs the test by using the accuracy analysis function that analyzes the measurement accuracy of the measurement result that is tested by the test measurement unit 41 using the measurement program, and the reference measurement program and the measurement program that is being created. And a difference analysis function for analyzing the measured results.

In the correlation analysis function, the same semiconductor product is
Correlation as shown in Fig. 2 for each test item from the measurement result data (reference data) measured using the reference measurement program and the measurement result data (startup data) measured using the measurement program being created. While creating the graph, the intercept, slope, and
The correlation coefficient is calculated and stored in the analysis result database 33.

The analysis result obtained by the correlation analysis function is sent to the output device 2 for output, and the analysis result judging section 43 is also provided.
Is sent to the computer and the correlation is determined. FIG. 3 shows an example of the intercept, the slope, and the correlation coefficient stored in the analysis result database 33 as the analysis result. In the analysis result determination unit 43, when the correlation is obtained between the reference data and the start-up data, the measurement program of the test item is determined to be “no problem”, and as shown in the example of FIG. If the graph is biased to, it is determined that the measurement program has a problem.

The correction candidate creating section 44 uses the knowledge database 3 based on the determination result by the analysis result determining section 43.
4 is searched, the corresponding coping method is acquired, and the correction candidate of the measurement program is created. FIG. 4 shows the knowledge database 34.
7 shows an example of a method of coping with the correction of the measurement program based on the analysis result of the past correlation stored in. In the example of FIG.
Regarding the test item "TEST5", "intercept = 0.30"
Since the value of “2” has the same tendency as “Countermeasure 3” of the knowledge database 34 shown in FIG. 4, “Countermeasure 3” is a correction candidate.
Select. This correction candidate is sent to the output device 2 and output as shown in FIG.

In the accuracy analysis function of the measurement result analysis unit 42,
The maximum (MAX) value, the minimum (MIN) value, and the standard deviation (σ) of the measurement result data when the semiconductor product is measured a plurality of times by using the program being created are calculated and stored in the analysis result database 33. In addition, the calculated analysis result is sent to the output device 2 and output, and is also sent to the analysis result determination unit 43 to determine whether or not a predetermined standard value is satisfied. FIG. 6 shows an example of the maximum value, the minimum value and the standard deviation calculated for each test item and the standard value for each test item.

The analysis result judging unit 43 compares the standard deviation as an analysis result for each test item with each standard value, and if the standard value is not satisfied, it is sent to the correction candidate preparing unit 44 as "problem". FIG. 7 shows an example of the test items determined as “problem” by the analysis result determination unit 43.

The correction candidate creating section 44 uses the knowledge database 3 based on the determination result by the analysis result determining section 43.
4 is searched, the corresponding coping method is acquired, and the correction candidate of the measurement program is created. FIG. 8 shows the knowledge database 34.
The example of the coping method of the correction of the measurement program based on the analysis result of the past accuracy stored in FIG. In the example of FIG.
Since “Countermeasure 3” and “Countermeasure 4” for the test item “TEST4” are registered in the knowledge database 34 of FIG. 8, “Countermeasure 3” and “Countermeasure 4” are selected as correction candidates.
This correction candidate is sent to the output device 2 and output as shown in FIG.

In the difference analysis function of the measurement result analysis unit 42,
Test the same semiconductor product with the measurement result data (reference data) measured with the reference measurement program and the measurement result data (startup data) measured with the measurement program being created, as shown in FIG. The items are collected and stored in the analysis result database 33. Further, the analysis result is sent to the output device 2 and output, and also sent to the analysis result determination unit 43 to make a difference determination.

The analysis result judging section 43 extracts a semiconductor product whose measurement result data is different between the reference data and the start-up data. Here, since it is the result of measuring the same semiconductor product, if even one value is different, it is assumed that there is some abnormality in the measurement program. FIG. 11 shows an example of a list of extracted semiconductor product test items and measurement results.

The correction candidate creating section 44 uses the knowledge database 3 based on the determination result by the analysis result determining section 43.
4 is searched, the corresponding coping method is acquired, and the correction candidate of the measurement program is created. FIG. 12 shows the knowledge database 3
4 shows an example of a method of coping with the correction of the measurement program based on the analysis result of the past difference stored in 4. In the example of FIG. 11, the reference data test item “TEST1” and the start-up data test item “TEST1” have the same tendency as “measure 1” and “measure 2” registered in the knowledge database 34 of FIG. “Countermeasure 1” and “Countermeasure 2” are selected as correction candidates.

Similarly, the reference data test item "OK",
Start-up data test item “TEST2” has the same tendency as “Countermeasure 3” in the knowledge database 34, reference data test item “TEST6”, start-up data test item “TES”
Since "ST3" has the same tendency as "Countermeasure 5" in the knowledge database 34, "Countermeasure 3" and "Countermeasure 5" are selected as correction items. These correction candidates are sent to the output device 2 and output as shown in FIG.

The processing by the measurement program creation support device having the above configuration will be described below with reference to the flow chart shown in FIG. As shown in FIG. 14, first, an operator creates a measurement program (step S101) and confirms that the measurement program operates (step S10).
2). Here, the created measurement program is stored in the measurement program database 31.

Next, test measurement is performed by the test measuring section 41 using the created measurement program (step S1).
03). When the test measurement is completed, the measurement result is stored in the measurement result database 32. The measurement result in the measurement result database 32 is extracted to the measurement result analysis unit 42, and the correlation analysis is performed by the correlation analysis function (step S104). The analysis results are collected in the form of a form, output from the output device 2, and stored in the analysis result database 33 (step S105).

Subsequently, in the analysis result judging section 52, it is judged whether or not each analysis value analyzed by the measurement result analyzing section 42 is within the standard value, and the test item name which is out of the standard value is corrected. It is sent to the candidate creation unit 44 (step S1)
06). In the correction candidate creation unit 44, the analysis result determination unit 52
When the analysis items are received from the analysis result database 33
The inside is referred to and the analysis result and the standard value of the analysis item which is out of the standard value are extracted from the corresponding analysis result (step S10
7).

The correction candidate creation unit 44 searches the knowledge database 34 under the condition of the difference between the analysis result of the extracted analysis item and the standard value (step S108), acquires a coping method that matches the condition, and acquires it. It is set as a correction candidate (step S109) and is output from the output device 2 (step S1).
10).

The operator corrects the measurement program based on the correction candidates output from the output device 2 (step S111), and a series of processes of steps S103 to S111 until all the analyzed values are within the standard values. Is repeated (step S112). Further, the above step S1
Similar to 03 to S112, the accuracy check by the accuracy analysis function of the measurement result analysis unit 42 (step S113) and the difference check by the difference analysis function (step S114).
And repeat until all the analyzed values are within the standard values.

As described above, in the measurement program creation support device according to this embodiment, the operator can correct the measurement program based on the analysis result and the correction candidate output from the output device 2. Then, by repeating this series of operations in all the analysis functions (correlation analysis function, accuracy analysis function, difference analysis function) of the measurement result analysis unit 42 until all analysis items fall within the standard values, each measurement device It is possible to create a measurement program that enables good measurement in.

Further, since the correction candidates output from the output device 2 are created based on the knowledge database 34 in which the coping method of the measurement program correction based on the past analysis result is stored, the knowledge and experience of the operator, etc. It is a correction candidate that does not depend on. Therefore, even a worker with little knowledge or experience can create a measurement program with the same quality as an expert.

[0031]

According to the present invention, the measurement result of the test performed by using the measurement program used for operating the semiconductor measuring device is analyzed, and it is determined whether or not the analysis result satisfies a predetermined standard. Based on the result of this judgment, the appropriate method of coping with the correction of the measurement program based on the past analysis results is selected, and the correction candidates for the measurement program are automatically created and output. By modifying the measurement program based on the candidates, you can quickly create a measurement program that is suitable for each measuring device.
Moreover, even a person with little knowledge or experience can create it with the same quality as an expert.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a measurement program creation support device according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a correlation graph created by a measurement result analysis unit in FIG.

FIG. 3 is a diagram showing an example of analysis results stored in the analysis result database of FIG.

FIG. 4 is a diagram showing an example of a method of coping with correction of a measurement program stored in the knowledge database of FIG.

5 is a diagram showing an example of correction candidates created by a modification candidate creation unit in FIG. 1. FIG.

FIG. 6 is a diagram showing an example of a result analyzed by a measurement result analysis unit in FIG.

[Fig. 7] "There is a problem" by the analysis result judging unit of Fig. 1
It is a figure which shows the example of the test item determined as.

FIG. 8 is a diagram showing an example of a method of coping with correction of a measurement program stored in the knowledge database of FIG.

FIG. 9 is a diagram showing an example of correction candidates created by a correction candidate creation unit in FIG. 1.

10 is a diagram showing an example of a result analyzed by a measurement result analysis unit in FIG.

11 is a diagram showing an example of a list of test items and measurement results extracted by the analysis result determination unit of FIG.

FIG. 12 is a diagram showing an example of a method of coping with correction of a measurement program stored in the knowledge database of FIG. 1.

FIG. 13 is a diagram showing an example of correction candidates created by a modification candidate creation unit in FIG. 1.

FIG. 14 is a flow chart of processing by the measurement program creation support device of FIG. 1.

FIG. 15 is a flowchart showing a process from creation of a conventional measurement program to introduction thereof.

[Explanation of symbols]

1 input device 2 output devices 3 storage devices 4 Central processing unit 31 Measurement program database 32 measurement result database 33 Analysis result database 34 Knowledge Database 41 Test measurement section 42 Measurement result analysis unit 43 Analysis Result Judgment Section 44 Correction Candidate Creation Department

Claims (3)

[Claims] 1. A measurement program database storing a measurement program used for operating a semiconductor measuring device, a measurement result database storing a measurement result of a test performed by the semiconductor measuring device using the measurement program, A measurement result analysis unit that analyzes the measurement result, an analysis result database that stores the analysis result by the measurement result analysis unit, an analysis result determination unit that determines whether or not the analysis result satisfies a predetermined standard, and a past A knowledge database that stores a coping method for correction of a measurement program based on an analysis result, and a search of the knowledge database based on a determination result by the analysis result determination unit, obtain a corresponding coping method, and create a correction candidate for a measurement program. Outputs the correction candidate creation unit and the correction candidate created by the correction candidate creation unit Measurement program creation support apparatus and a radical 19. 2. The measurement result analysis unit has a function of analyzing a correlation of measurement results tested by a semiconductor measuring device using a reference measurement program and a measurement program being created, respectively, and A function that analyzes the measurement accuracy of the measurement result that is tested by the semiconductor measurement device using the measurement program, and a measurement that is performed by the semiconductor measurement device using the reference measurement program and the measurement program being created. A function for analyzing a result is provided.
The described measurement program creation support device. 3. A step of acquiring and analyzing the measurement result from a measurement result database that stores a measurement result obtained by performing a test using a measurement program used for operating a semiconductor measuring device, and the analysis result A step of determining whether or not a predetermined standard is satisfied, and based on this determination result, a knowledge database that stores a coping method for the measurement program modification based on the past analysis result is searched, and a corresponding coping method is acquired. A measurement program creation support program for causing a computer to execute a step of creating a modification candidate of a measurement program and a step of outputting the created modification candidate.