JP2008151585A - Method and apparatus for analyzing distribution, method and apparatus for estimating abnormal equipment, program for allowing computer to execute distribution analyzing method or abnormal equipment estimation method and recording medium capable of reading computer on which program is recorded - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform evaluation of high precision in a distribution analyzing method for evaluating whether analyzing target data having characteristics distributed along a plane is similar to a specific distribution shape pattern. <P>SOLUTION: The distribution shape pattern is defined by demarcating the plane corresponding to an analyzing target data decided plane into a plurality of rectangular regions in a lattice like state and imparting a density value for acquiring a multivalue in a certain numerical value range so as to show the relative magnitude relation of the characteristic value of the analyzing target data (S101). The density value of the distribution shape pattern is adjusted so that the magnitude of the characteristic value of a characteristic data group is reflected to the evaluation by using the characteristic value of a predetermined distribution shape pattern adjustment characteristic data group (S102 and S103). The analyzing target data is compared with the distribution shape pattern after adjustment to calculate the quantitative similarity of the distribution of the characteristic value of the analyzing target data and of the distribution of the density value of the distribution shape pattern after adjustment (S104). <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a distribution analysis method and apparatus, and more particularly to a distribution analysis method and apparatus for evaluating whether or not analysis target data having characteristics distributed along a plane is similar to a specific distribution shape.

  The present invention also relates to an abnormal facility estimation method and apparatus for automatically estimating a manufacturing apparatus in which an abnormality has occurred from among a plurality of manufacturing apparatuses respectively used in a plurality of manufacturing processes sequentially performed on a substrate.

  The present invention also relates to a program for causing a computer to execute such a distribution analysis method or abnormal facility estimation method.

  The present invention also relates to a computer-readable recording medium recording such a program.

  In a production line for thin film devices such as semiconductor wafers, semiconductor displays, and hard disk magnetic heads, various inspections are performed for the purpose of improving yield and stabilizing. These inspections include, for example, a pattern inspection for detecting a defect in a circuit pattern caused by a foreign substance attached on a substrate, an electrical characteristic inspection for inspecting an electrical characteristic of a formed circuit, and the like. The production line monitors these inspection results on a daily basis.For example, abnormalities have occurred in the manufacturing process based on the increase in the number of defects detected by pattern inspection and fluctuations in electrical characteristics measured by electrical characteristic inspection. Check whether there is any. If an abnormality occurs in a certain manufacturing process among a plurality of manufacturing processes that are sequentially performed on the board, the inspection results should be investigated and analyzed to quickly identify the cause and take countermeasures. Thus, damage due to yield reduction can be minimized. Therefore, a system for collecting inspection information and processing history of each manufacturing apparatus is provided in the manufacturing line.

When an abnormality occurs in the manufacturing process, the inspection result often appears as a characteristic distribution on the substrate surface. Since these distributions vary depending on the state of the abnormality and are closely related to the cause of the abnormality, examples of past characteristic distributions are compiled into a database together with the cause of the abnormality, and the cause of the abnormality is investigated and decision-making measures are taken. To be used. Therefore, it is important to detect the tendency of the characteristic distribution in the inspection process at an early stage, but in reality it is difficult to express the characteristic distribution numerically. It is difficult to detect and monitor the characteristic distribution because it often cannot be found even if the distribution occurs.
Japanese Patent Laid-Open No. 11-45919 JP 2003-100825 A

  In order to overcome these difficulties, a method has been proposed in which a geometric characteristic of a pre-registered distribution is compared with a characteristic distribution of the substrate, and a substrate having a characteristic distribution similar to the registered distribution is detected. As such a technique, for example, in Patent Document 1 (Japanese Patent Laid-Open No. 11-45919), a characteristic distribution is treated as a binary image, a binary image having a distribution shape registered in advance and a binary image having a characteristic distribution of a substrate. Describes a template matching method used in image processing. Specifically, in the method of this document, the number of defects detected on the substrate in the inspection process is totaled for each grid-like pixel set with respect to the substrate surface. Next, threshold processing is performed with a predetermined threshold on the number of defects for each pixel, and the number of defects for each pixel is binarized. From this result, a binary image of the entire surface of the substrate is created, and template matching in image recognition is performed on the defect distribution template illustrated in FIG. 23, which is also created in advance as a binary image. When the matching rate with is high, it is determined that it matches the defect distribution defined in the template. However, the defect distribution detected in the actual inspection process has a different concentration of defects depending on the shape, and the number of defects also changes depending on the time of inspection. Therefore, it is not possible to uniquely determine a threshold value when creating a binary image. very difficult. In addition, since template matching between binary images is performed, there is a problem that matching accuracy is low.

  As another method, for example, in Patent Document 2 (Japanese Patent Laid-Open No. 2003-100825), a region where defects are concentrated as shown in FIG. 24A is defined as a template. As shown in FIG. 24B, there is described a method for determining that a substrate having a higher number of defects in the template region with respect to the total number of defects on the substrate has a higher degree of coincidence. In another method described in the same document, a template is defined that includes density values for lattice-shaped small areas set on the substrate surface, and the density value of the template and the analysis target substrate are defined for each small area. Describes a method of using a correlation coefficient with the number of defects counted as a degree of coincidence. However, in this method, since the number of defects is not reflected in the degree of coincidence, the obtained result does not coincide with human visual coincidence. In addition, a small number of defects are generated at random positions in the normal state on the substrate surface, but if such defects occur in an area defined by a template by chance, the degree of coincidence with the template is very high. Since it is calculated, there is a problem that detection accuracy is poor.

  Further, a problem common to the above two conventional techniques is that the evaluation is made only with the characteristic distribution shape and the characteristic value is not taken into account. When the person in charge of analysis confirms the analysis result, it is necessary to confirm how large the characteristic value is with the target distribution shape. However, the above-described conventional technique does not perform such evaluation. For this reason, there is a problem that accuracy is not good.

  Accordingly, an object of the present invention is a distribution analysis method and apparatus for evaluating whether or not analysis target data having characteristics distributed along a plane is similar to a specific distribution shape pattern. To provide an analysis method and apparatus.

  Another object of the present invention is to use a degree of similarity based on such a distribution analysis method on the inspection result of a substrate, thereby making it possible to use a plurality of manufacturing apparatuses sequentially used for a substrate. An object of the present invention is to provide an abnormal facility estimation method and apparatus for automatically estimating a manufacturing apparatus in which an abnormality has occurred.

  Another object of the present invention is to provide a program for causing a computer to execute such a distribution analysis method or abnormal facility estimation method.

  Another object of the present invention is to provide a computer-readable recording medium in which such a program is recorded.

In order to solve the above problems, the distribution analysis method of the present invention is:
A distribution analysis method for evaluating whether analysis target data having characteristic values distributed along a plane is similar to a specific distribution shape pattern,
The distribution shape pattern is divided into a plurality of rectangular areas in a lattice shape on a plane corresponding to the plane defining the analysis target data, and the relative magnitude of the characteristic value of the analysis target data is determined for each rectangular area. In order to express the relationship, define a concentration value that can take multiple values within a certain numerical range,
Using the characteristic value of the predetermined distribution shape pattern adjustment characteristic data group, the density value of the distribution shape pattern is adjusted so that the size of the characteristic value of the characteristic data group is reflected in the evaluation,
Comparing the analysis target data with the adjusted distribution shape pattern to obtain a quantitative similarity between the characteristic value distribution of the analysis target data and the density value distribution of the adjusted distribution shape pattern It is characterized by.

  Here, the “analysis target data” may be any data as long as some characteristic value is distributed along the plane.

  “Multi-value” refers to two or more values, preferably three or more values.

  If processing corresponding to the distribution analysis method of the present invention is executed, for example, by a computer, the quantitative similarity between the distribution of characteristic values of the analysis target data and the distribution of concentration values of the adjusted distribution shape pattern can be obtained. It can be determined automatically. Based on this similarity, it is possible to evaluate whether the distribution of characteristic values of the analysis target data is similar to a specific distribution shape pattern. Here, the adjusted distribution shape pattern uses the characteristic value of a predetermined distribution shape pattern adjustment characteristic data group, and the distribution value pattern reflects the magnitude of the characteristic value of the characteristic data group in the evaluation. Since the density value of the shape pattern is adjusted, the evaluation result is not easily influenced by the fluctuation factor. In other words, the size of the characteristic value of the data to be analyzed differs for each distribution target pattern, or the size of the characteristic value of the analysis target data changes depending on the time of analysis even for the same distribution target pattern. Even if such a variable factor exists, the evaluation result is not easily affected by such a variable factor. Therefore, according to this distribution analysis method, evaluation can be performed with high accuracy.

In the distribution analysis method of one embodiment,
Adjustment of the above density value
First step of extracting, from the analysis target data group formed by the analysis target data, data whose characteristic value distribution is similar to the density value distribution of the distribution shape pattern as the distribution shape pattern adjustment characteristic data group When,
Using the distribution statistics of the characteristic values of the distribution shape pattern adjustment characteristic data group obtained in the first step and the distribution statistics of the density values of the distribution shape patterns, the density value of the distribution shape pattern A second step of adjusting the value is provided.

  In the distribution analysis method of this embodiment, the distribution of characteristic values of the distribution shape pattern adjustment characteristic data group extracted from the analysis target data group is reflected in the adjustment of the density value of the distribution shape pattern. Therefore, even if there is a difference in distribution shape pattern or a difference in characteristic values between populations in the same distribution shape, the similarity can be evaluated with high accuracy.

  In the distribution analysis method according to one embodiment, the first step includes a correlation coefficient between a characteristic value corresponding to each rectangular area of each analysis target data and a density value assigned to each rectangular area of the distribution shape pattern. Is extracted as the above distribution shape pattern adjustment characteristic data group.

  In the distribution analysis method of this embodiment, by using the correlation coefficient, a useful distribution shape pattern adjustment characteristic data group can be extracted regardless of the characteristic value.

In the distribution analysis method of one embodiment,
The statistic of the distribution of characteristic values of the distribution shape pattern adjustment characteristic data group in the second step is a standard deviation of the average value of each characteristic data constituting the characteristic data group for each analysis target data, and the distribution shape The statistic of the pattern density value distribution is the standard deviation of the density value of the distribution shape pattern,
The similarity is given by covariance between a density value given to each rectangular area of the adjusted distribution shape pattern and a characteristic value corresponding to each rectangular area of the analysis target data. .

In the distribution analysis method of one embodiment,
The statistic of the distribution of the characteristic values of the distribution shape pattern adjustment characteristic data group in the second step is an average value for each analysis target data of the characteristic values forming the characteristic data group, and the density of the distribution shape pattern The value distribution statistic is the standard deviation of the density values of the distribution shape pattern,
The similarity is given by covariance between a density value given to each rectangular area of the adjusted distribution shape pattern and a characteristic value corresponding to each rectangular area of the analysis target data. .

  In the distribution analysis method of this embodiment, the similarity can be obtained with high accuracy regardless of changes in the distribution shape pattern and the characteristic value of the population of the analysis target data.

In the distribution analysis method of one embodiment,
In order to obtain the above similarity,
A third step of determining whether the relative magnitude relationship of the characteristic values of the analysis target data is similar to the relative magnitude relationship of the density values of the distribution shape pattern;
A fourth step of evaluating the similarity using the determination result of the third step is provided.

  In the distribution analysis method according to an embodiment, the third step includes assigning the characteristic value corresponding to each rectangular area of the analysis target data and the rectangular area of the distribution shape pattern or the adjusted distribution shape pattern. It is determined that analysis target data having a correlation coefficient with a density value that is equal to or greater than a predetermined threshold value is similar to the distribution shape pattern.

  In this embodiment, since the shape of the characteristic distribution of the analysis target data is used as pre-processing for evaluating the similarity, the similarity can be evaluated with high accuracy even if a very large value exists in the characteristic value. .

  In one embodiment of the distribution analysis method, the characteristic value of the analysis target data is distributed along the surface of the substrate.

  Here, the “substrate” widely refers to a glass substrate on which a thin film device is manufactured, a wafer on which a semiconductor device is manufactured, and the like.

  In the distribution analysis method of this embodiment, the distribution of the characteristic values on the substrate is distributed without depending on the difference in the characteristic value due to the variation factor generated in the manufacturing process of the substrate and the temporal change of the characteristic value. Whether it is similar to the density value distribution of the shape pattern is evaluated with high accuracy.

The abnormal equipment estimation method of this invention is
An abnormal facility estimation method for estimating a manufacturing apparatus in which an abnormality has occurred, from among the manufacturing apparatuses used in a plurality of manufacturing steps sequentially performed on a substrate,
Perform the distribution analysis method to determine the similarity,
Based on a processing history representing a manufacturing apparatus that has processed each of the substrates in each of the manufacturing steps, a manufacturing apparatus commonly used for the substrates having the similarity equal to or higher than a predetermined threshold is extracted. And

  According to the abnormal equipment estimation method of the present invention, the manufacturing apparatus used in common for the substrates whose similarity is equal to or higher than a predetermined threshold is extracted, so that the manufacturing apparatus in which the abnormality has occurred is accurately estimated. be able to.

The distribution analyzer of this invention is
A distribution analysis device that evaluates whether analysis target data having characteristic values distributed along a plane is similar to a specific distribution shape pattern,
The distribution shape pattern is divided into a plurality of rectangular areas in a lattice shape on a plane corresponding to the plane defining the analysis target data, and the relative magnitude of the characteristic value of the analysis target data is determined for each rectangular area. A pattern registration unit that defines and assigns density values that can take multiple values within a certain numerical range to represent the relationship;
Using a characteristic value of a predetermined distribution shape pattern adjustment characteristic data group, and a pattern adjustment unit for adjusting the density value of the distribution shape pattern so that the magnitude of the characteristic value of the characteristic data group is reflected in the evaluation; ,
Similarity in which the analysis target data is compared with the adjusted distribution shape pattern to obtain a quantitative similarity between the distribution of characteristic values of the analysis target data and the distribution of density values of the adjusted distribution shape pattern And a degree evaluation unit.

  Here, the “analysis target data” may be any data as long as some characteristic value is distributed along the plane.

  “Multi-value” refers to two or more values, preferably three or more values.

  According to the distribution analysis apparatus of the present invention, the distribution analysis method of the present invention can be implemented. That is, if the pattern registration unit, the pattern adjustment unit, and the similarity evaluation unit are configured by, for example, a computer program and execute processing, the distribution of characteristic values of the analysis target data and the density of the adjusted distribution shape pattern The quantitative similarity with the value distribution can be automatically obtained. Based on this similarity, it is possible to evaluate whether the distribution of characteristic values of the analysis target data is similar to a specific distribution shape pattern. Here, the adjusted distribution shape pattern uses the characteristic value of a predetermined distribution shape pattern adjustment characteristic data group, and the distribution value pattern reflects the magnitude of the characteristic value of the characteristic data group in the evaluation. Since the density value of the shape pattern is adjusted, the evaluation result is not easily influenced by the fluctuation factor. In other words, the size of the characteristic value of the data to be analyzed differs for each distribution target pattern, or the size of the characteristic value of the analysis target data changes depending on the time of analysis even for the same distribution target pattern. Even if such a variable factor exists, the evaluation result is not easily affected by such a variable factor. Therefore, according to this distribution analysis method, evaluation can be performed with high accuracy.

In the distribution analyzer of one embodiment,
The pattern adjustment unit
A first part for extracting, as the distribution shape pattern adjustment characteristic data group, data whose characteristic value distribution is similar to the distribution of density values of the distribution shape pattern from the analysis target data group formed by the analysis target data When,
The density value of the distribution shape pattern is obtained by using the distribution statistics of the characteristic values of the distribution shape pattern adjustment characteristic data group obtained in the first part and the distribution statistics of the density values of the distribution shape patterns. It is characterized by comprising a second portion for adjusting the angle.

  In the distribution analysis method of this embodiment, the distribution of characteristic values of the distribution shape pattern adjustment characteristic data group extracted from the analysis target data group is reflected in the adjustment of the density value of the distribution shape pattern. Therefore, even if there is a difference in distribution shape pattern or a difference in characteristic values between populations in the same distribution shape, the similarity can be evaluated with high accuracy.

In the distribution analyzer of one embodiment,
The similarity evaluation unit
A third part for determining whether the relative magnitude relationship between the characteristic values of the analysis target data is similar to the relative magnitude relationship between the density values of the distribution shape pattern;
A fourth portion for evaluating the similarity using the determination result by the third portion is provided.

  In this embodiment, since the shape of the characteristic distribution of the analysis target data is used as pre-processing for evaluating the similarity, the similarity can be evaluated with high accuracy even if a very large value exists in the characteristic value. .

  In one embodiment of the distribution analysis apparatus, the characteristic value of the analysis target data is distributed along the surface of the substrate.

  Here, the “substrate” widely refers to a glass substrate on which a thin film device is manufactured, a wafer on which a semiconductor device is manufactured, and the like.

  In the distribution analysis apparatus of this embodiment, the distribution of the characteristic values on the substrate is distributed without depending on the difference in the characteristic values due to the variation factors generated in the manufacturing process of the substrate and the temporal change of the characteristic values. Whether it is similar to the density value distribution of the shape pattern is evaluated with high accuracy.

The abnormal equipment estimation device of this invention is
An abnormal facility estimation device that estimates a manufacturing device in which an abnormality has occurred, from among the manufacturing devices used in each of a plurality of manufacturing steps that are sequentially performed on a substrate,
The distribution analysis apparatus is provided, and the similarity is obtained by the distribution analysis apparatus,
Based on a processing history representing a manufacturing apparatus that has processed each of the substrates in each of the manufacturing steps, an abnormal facility estimation is performed to extract a manufacturing apparatus that is commonly used for the substrates whose similarity is equal to or greater than a predetermined threshold. It has the part.

  According to the abnormal equipment estimation apparatus of the present invention, since the manufacturing apparatus used in common for the substrates whose similarity is equal to or higher than a predetermined threshold is extracted, the manufacturing apparatus in which the abnormality has occurred is accurately estimated. be able to.

  The program of the present invention is a distribution analysis program for causing a computer to execute the distribution analysis method of the present invention.

  In another aspect, the program of the present invention is an abnormal equipment estimation program for causing a computer to execute the abnormal equipment estimation method of the above invention.

  The recording medium of the present invention is a computer-readable recording medium in which the distribution analysis program of the above invention is recorded.

  In another aspect, the recording medium of the present invention is a computer-readable recording medium in which the abnormal facility estimation program of the present invention is recorded.

  Hereinafter, the present invention will be described in detail with reference to the illustrated embodiments.

  FIG. 1 schematically shows the concept of a distribution analysis method according to an embodiment of the present invention, taking as an example the result of pattern inspection that detects, as a characteristic value, a circuit pattern defect caused by a foreign substance or the like in a substrate manufacturing process. . This distribution analysis method is applied when, for example, a substrate in which defects are concentrated in a specific region is detected from a group of substrates inspected in a manufacturing process.

  In this distribution analysis method, in step S101, the distribution shape of the characteristic value for comparison with the analysis target substrate is registered as a distribution shape pattern. An example of the distribution shape pattern is shown in FIG. As shown in FIG. 2A, the distribution shape pattern is obtained by setting the plurality of rectangular areas 201, 201,... By dividing the surface of the substrate into a lattice shape, and setting the density value for each rectangular area 201. Defined. The density value can take multiple values (in this example, three or more values) within a range of 0 to 1. A set of density values set for each rectangular area 201 represents the relative magnitude relationship of the characteristic values on the substrate. In the pattern inspection, since the number of defects in each rectangular area corresponds to “characteristics”, the density value set for each rectangular area corresponds to a value obtained by converting the defect density into a relative value in the range of 0 to 1. To do. 2B and 2C show examples of the distribution shape pattern. The color of the area in the figure represents the density value, and the darker the color, the higher the density value. Accordingly, the distribution shape pattern 202 in FIG. 2B represents a pattern in which defects are concentrated in the lower right area of the substrate, and the distribution shape pattern 203 in FIG. 2C has defects in the upper left area of the substrate. Represents a concentrated pattern. In general, in the manufacturing process, if the distribution shape of defects is different, the number of defects (raw value) generated in each region is different. Since the above density value represents only the relative magnitude relationship between the characteristic values on the substrate surface, defects are concentrated in any region on the substrate surface regardless of the number of defects (raw value). It can be easily set only from a relative point of view. The distribution shape pattern may be manually input by an analyst using knowledge about the manufacturing process, or may be automatically created using past cases.

  In step S102 in FIG. 1, characteristic data of a substrate group for distribution shape pattern adjustment prepared separately from a substrate to be analyzed is input. Here, the substrate group for distribution shape pattern adjustment refers to a substrate group that can be regarded as a population of substrates to be analyzed, such as a substrate group manufactured in a manufacturing process at the same time as a substrate to be actually analyzed, A substrate to be analyzed may be included. The characteristic data input in step S102 will be described with reference to FIGS. In the pattern inspection, the coordinates of the defect 302 detected on the substrate 301 in FIG. 3 are detected. Therefore, the substrate surface to be analyzed is divided into a plurality of rectangular areas similar to the rectangular areas 201, 201,... Of the distribution shape pattern, and the number of defects belonging to the rectangular area is tabulated for each rectangular area. Then, the distribution of the number of defects obtained for each rectangular area (indicated by numbers such as “1”, “2”, and “3” in the rectangular area 201 in FIG. 4) is the characteristic data 401 of the substrate. Get as. Note that the characteristic data of the analysis target substrate input in step S104, which will be described later, is similarly represented by the distribution of the number of defects obtained for each rectangular area.

  Next, in step S103 in FIG. 1, the distribution shape pattern registered in step S101 is adjusted using the characteristic distribution of the characteristic data group calculated in step S102. In step S104, the adjusted distribution shape pattern and the analysis target substrate are adjusted. Find the similarity to the characteristic data. The concept of the distribution shape pattern adjustment in step S103 and step S104 will be described next with reference to FIGS. 5 (a) to 5 (d), FIG. 6 (a), and FIG. 6 (b). 5A to 5D are examples of distribution of pattern defects in the substrate. The substrate 501 and the substrate 502 have the distribution shape pattern 202 in FIG. 2B, and the substrate 503 and the substrate 504 have the same structure as FIG. This corresponds to the distribution shape pattern 203 of c). Comparing the substrate 501 and the substrate 502, the substrate 501 has a higher concentration of the distribution shape pattern 202 than the substrate 502 because defects are concentrated in a region defined by a high density value in the distribution shape pattern 202. It is reasonable to judge. Similarly, when the substrate 503 and the substrate 504 are compared, it can be determined that the substrate 503 has a higher degree of similarity to the distributed shape pattern 203 than the substrate 504. On the other hand, FIG. 6A and FIG. 6B show the number of defects and the above distribution in a region where the density value of the distribution shape pattern is high, that is, a region where defects are concentrated, with respect to the distribution shape pattern 202 and the distribution shape pattern 203, respectively. The example of the relationship of the number of the board | substrates in the board | substrate group for pattern adjustment is represented. In this example, the number of defects of the substrate corresponding to the distribution shape pattern 202 is larger than the number of defects of the substrate corresponding to the distribution shape pattern 203. Further, the substrate 601 and the substrate 602 in FIG. 6 have the same number of defects in the concentrated region, and correspond to both the distribution shape pattern 202 and the distribution shape pattern 203, respectively. The substrate 601 has a small number of defects in the substrate group corresponding to the distribution shape pattern 202, and the substrate 602 has a large number of defects in the substrate group corresponding to the distribution shape pattern 203. Considering the number, the similarity between the defect distribution on the substrate 601 and the distribution shape pattern 202 needs to be evaluated to a low value, and the similarity between the defect distribution on the substrate 602 and the distribution shape pattern 203 needs to be evaluated to a high value. Therefore, in order to evaluate the quantitative similarity considering both the number of defects and the characteristic distribution, a group of boards having the same characteristic distribution is extracted from the population of the analysis target boards, and those boards are extracted. It is necessary to evaluate the relative similarity from the characteristic value of. Therefore, in step S103, the density value of the distribution shape pattern is adjusted by the procedure shown in FIG. 7 so that the density value of the distribution shape pattern represents the relative similarity measure.

  First, in the first step S1031 included in step S103, the relative number of defects in the characteristic data 401 is not based on the size of the defect itself forming the characteristic data 401 from the substrate group for distribution shape pattern adjustment. A substrate is extracted such that the size relationship is similar to the relative size relationship of the density values of the distribution pattern. At the same time, the characteristic data about these extracted substrates (this is called “distributed shape pattern adjustment characteristic data group”) is extracted.

  In step S1032, the distribution shape pattern density value is adjusted using the distribution shape pattern adjustment characteristic data group. Here, since the density value of the distribution shape data is adjusted so that the accuracy of the similarity calculated in step S104 is improved, the adjustment method depends on the similarity evaluation method in step S104.

  Finally, in step S104, the characteristic data for each rectangular area is calculated from the input inspection result of the analysis target substrate by the same process as in step S102, and the distribution of the characteristic data and the adjusted distribution shape obtained in step S103 are calculated. A quantitative similarity with the distribution of density values of the pattern is obtained.

  By performing the above processing, it is possible to accurately evaluate the similarity between the distribution shape pattern and the characteristic data of the analysis target substrate. In other words, a board similar to the distribution shape pattern defined by the relative magnitude relationship of the density values is extracted from the population of the analysis target boards, and the characteristic data group (distribution shape pattern adjustment characteristics) for those extracted boards is extracted. The density value of the distribution shape pattern is adjusted using the data group. By evaluating the degree of similarity between the adjusted distribution shape pattern and the characteristic data of the substrate to be analyzed, the difference in the number of defects for each distribution shape pattern as shown in FIG. The quantitative similarity can be evaluated without doing so.

  Note that the distribution analysis method according to the present invention is not applied only to the pattern inspection of a substrate, but can be applied to any data in which characteristics are distributed in a plane.

  FIG. 8 shows a configuration of a distribution analysis apparatus 800 according to an embodiment of the present invention suitable for implementing the above-described distribution analysis method (see FIG. 1). The distribution analysis apparatus 800 includes a pattern registration unit 802, a data collection unit 803 as an input unit, a characteristic data calculation unit 806, an adjustment characteristic data group registration unit 807, a pattern adjustment unit 808, and a similarity evaluation unit 809. ing. In addition, an input device 801, an inspection information collection system 804, and an inspection device 805 are connected to the data collection unit 803 of the distribution analysis device 800, and an output device 810 is connected to the similarity evaluation unit 809. .

  The input device 801 is composed of, for example, a keyboard and a mouse. In this example, the input device 801 is used to input a distribution shape pattern definition, an analysis target substrate identification ID, a distribution shape pattern adjustment characteristic data group condition specification, and the like to the distribution analysis device 800.

  The pattern registration unit 802 receives the distribution shape pattern defined by the input device 801 and records it in the database. The information received from the input device 801 is not only the distribution shape pattern, but also information related to the distribution shape pattern, for example, information about the device that causes the distribution shape pattern, is recorded in association with the distribution shape pattern. Also good.

  The data collection unit 803 generates the inspection information collection system 804 and the inspection information of the substrate that matches the identification information of the analysis target substrate transmitted from the input device 801 to the distribution analysis device 800 and the conditions of the distribution shape pattern adjustment characteristic data group. Collected from the device 805 and passed to the characteristic data calculation unit 806. At this time, if necessary, information such as the substrate identification ID and the inspection date may be collected together with the inspection data.

  The inspection information collection system 804 is an inspection information collection system that collects inspection information from inspection devices arranged in the manufacturing process. In this example, the inspection information collection system 804 is connected to the inspection device 805, and the inspection data and inspection date and time of the substrate processed by the inspection device 805, the identification ID of the substrate, and the like are accumulated. In addition, the inspection information collection system 804 is connected to the distribution analyzer 800, and the data collection unit 803 collects necessary inspection data from the inspection information collection system 804.

  The inspection device 805 is arranged in the manufacturing process and actually inspects the substrate. The inspection device 805 is connected to the distribution analysis device 800, and the data collection unit 803 collects necessary inspection data from the inspection device 805.

  The characteristic data calculation unit 806 aggregates the inspection data of the substrate collected by the data collection unit 803 from the inspection information collection system 804 and the inspection apparatus 805 for each rectangular area 201 and creates characteristic data. Of the inspection data received from the data collection unit 803, the distribution shape pattern adjustment characteristic data group is transferred to the adjustment characteristic data group registration unit 807, and the analysis target substrate characteristic data is transmitted to the similarity evaluation unit 809.

  The adjustment characteristic data group registration unit 807 records the distribution shape pattern adjustment characteristic data group received from the characteristic data calculation unit 806. At this time, if necessary, information such as the substrate identification ID and the inspection date may be recorded together with the characteristic data.

  The pattern adjustment unit 808 uses the distribution shape pattern registered in the pattern registration unit 802 and the distribution shape pattern adjustment characteristic data group information registered in the adjustment characteristic data group registration unit 807 to perform the distribution in step S103 of FIG. Shape pattern adjustment processing is performed, and the adjusted distribution shape pattern is sent to the similarity evaluation unit 809. The internal configuration of the pattern adjustment unit 808 is shown in FIG. The similar data extraction unit 901 performs the process of step S1301 in FIG. 7, and selects a data group similar to the distribution shape pattern received from the pattern registration unit 802 from the characteristic data group received from the adjustment property data group registration unit 807. This is extracted as a distribution shape pattern adjustment characteristic data group and transferred to the distribution shape pattern adjustment unit 902 together with the distribution shape pattern. The distribution shape pattern adjustment unit 902 adjusts the density value of the distribution shape pattern by performing the processing in step S1302 of FIG. 7 on the received similar characteristic data group and distribution shape pattern, and sends the result to the similarity evaluation unit 809.

  The similarity evaluation unit 809 receives the analysis target substrate characteristic data calculated by the characteristic data calculation unit 806 and the distribution shape pattern whose density value has been adjusted by the pattern adjustment unit 808, and performs a process of calculating the similarity in step S104. The calculated similarity is sent to the output device 810. At this time, if necessary, information such as the substrate identification ID and the inspection date may be sent together with the similarity.

  The output device 810 outputs a distribution analysis result by the distribution analysis device 800 through a monitor, paper output, a magnetic disk, a portable semiconductor memory, or the like.

  Note that a plurality of distribution shape patterns may be registered in the pattern registration unit 802. In this case, the analysis target substrate is evaluated for similarity to each distribution shape pattern, and is output to the output device 810.

  Moreover, when all the information necessary for distribution analysis can be acquired from either one of the inspection information system 804 and the inspection apparatus 805, only one of the information that can be acquired may be connected.

  Also, conditions necessary for the analysis are stored in the distribution analysis apparatus 800, and when the inspection apparatus 805 inspects the substrate, the data collection unit 803 automatically adjusts the inspection data of the analysis target substrate and the distribution shape pattern adjustment. Inspection data for the substrate group may be acquired.

  In addition, the input device 801 and the output device 810 may be the same device, for example, a device in which a display unit serving as an output device is provided with a keyboard serving as an input device and integrated. Further, the input device 801 and the output device 810 may be included in the distribution analysis device 800.

  Further, the output device 810 may receive and output information other than information necessary for distribution analysis from the input device 801, the inspection information collection system 804, or the inspection device 805 through the distribution analysis device 800.

  With the distribution analysis apparatus 800 having this configuration, the above-described distribution analysis method can be implemented.

  Note that the above-described distribution analysis method may be constructed as a program for causing a computer to execute the distribution analysis method.

  Further, such a program may be recorded on a computer-readable recording medium such as a CD-ROM and distributed. By installing the program in a general-purpose computer, the distribution analysis method can be executed by the general-purpose computer.

  The present invention is not limited to the distribution analysis apparatus 800 for pattern inspection, and can be applied to any data whose characteristics are planarly distributed. In that case, the data collection unit 803 is connected to an information collection system or characteristic measurement device in which target data is stored.

  Next, as Example 1, a method for realizing Steps S1031 and S1032 in FIG. 7 and Step S104 in FIG. 1 will be specifically described.

  In the present invention, the distribution shape pattern registered in step S101 in FIG. 1, the distribution shape pattern adjustment characteristic data group input in step S102, and the analysis target substrate characteristic data input in step S104 are the number of rectangular areas. Consists of the same number of pairs. Hereinafter, in order to simplify the description, each of the above data is defined by two rectangular areas unless otherwise specified.

この相関係数ｒは−１から１の範囲を取り、値が大きいほど分布形状パターンＴと特性データＭの方向は近い。そこで、相関係数ｒが予め定めた閾値以上であるような特性データＭを分布形状パターンＴと類似した特性データ群として抽出する。 First, a method for extracting a similar substrate group in step S1031 will be described. FIG. 10 is an example of a graph in which density values and characteristic values in the area 1 and area 2 as rectangular areas are plotted for the distribution shape pattern and the characteristic data group. In FIG. 10, the horizontal axis represents the density value of the distribution shape pattern and the characteristic value of the characteristic data in the region 1, and the vertical axis represents the density value of the distribution shape pattern and the characteristic value of the characteristic data in the region 2 (FIG. 11. Same as in FIGS. 11, 12, 14, and 16.) Since the ratio of the characteristic values of the region 1 and the region 2 is the same for the data existing in the same direction as viewed from the origin on the graph, the direction of each point viewed from the origin in the graph represents the distribution shape of the values. Therefore, in order to extract characteristic data having a characteristic distribution similar to the distribution shape data from the characteristic data group, it is only necessary to extract characteristic data existing in the same direction as the distribution shape pattern. In FIG. 10, the characteristic data group 1003 existing in the same direction as the distribution shape pattern 1001 and the characteristic data group 1004 existing in the same direction as the distribution shape pattern 1002 are similar to the distribution shape pattern 1001 and the distribution shape pattern 1002, respectively. The characteristic data group can be determined. Whether or not the directions of the distribution shape pattern and the distribution shape pattern adjustment characteristic data coincide with each other can be determined using, for example, a correlation coefficient that is a statistic for evaluating the angle formed by the two data. Distribution shape pattern T characteristic data M for phrase distribution profile pattern adjustment has g-number of rectangular regions, each _{T = (t 1, ...,} t g), M = (m 1, ..., m g) tables in Suppose that At this time, the correlation coefficient r between the distribution shape pattern T and the characteristic data M is expressed by the following (Equation 1).

The correlation coefficient r ranges from −1 to 1, and the larger the value, the closer the direction of the distribution shape pattern T and the characteristic data M is. Therefore, characteristic data M whose correlation coefficient r is equal to or greater than a predetermined threshold is extracted as a characteristic data group similar to the distribution shape pattern T.

  Next, density value adjustment processing in step S1032 is performed. In this processing, since it is necessary to adjust density values so that the evaluation result in step S104 is obtained as a quantitative similarity, the similarity in step S104 is determined. It explains together with the evaluation method of degree.

ここで、ｃは（ｔ_ｉ−ｔ_ａｖ）と（ｐ_ｉ−ｐ_ａｖ）との正負が等しく、絶対値が大きいときに大きな値となる。つまり、分布形状パターンで濃度値が高い領域の特性値の大きな特性データＰほど大きな値となるので、図１２の類似度の概念を満たす。しかしながら、解析対象の基板の特性値が大きいほど特性値の平均値ｐ_ａｖと各領域の特性値ｐ_ｉの差は大きくなる傾向があるので、ある分布形状パターンに該当する基板が図６（ａ）に示す分布を持つ場合と図６（ｂ）に示す分布を持つ場合には、図６（ａ）の分布形状パターンに該当する基板の共分散ｃは全体的に高い値となる。また、分布形状パターンによって領域毎の特性値の分布が異なるので、分布形状パターン毎に（ｔ_ｉ−ｔ_ａｖ）の値がばらばらである。これらの理由から、共分散ｃでは、分布形状パターンが異なったり、分布形状パターンに該当する基板の特性データの分布が異なったりすると定量的な評価ができない。そこで、上記基板の分布による（ｐ_ｉ−ｐ_ａｖ）の違いを抑制するために、ステップＳ１０３で抽出した類似特性データ群の分布を用いる。ｎ個の特性データからなる類似特性データ群ＬはＬ＝（Ｌ_１，…，Ｌ_ｎ）で表され、ｊ番目の特性データはＬ_ｊ＝（ｌ_ｊ１，…，ｌ_ｊｇ）で表されるとする。このとき、この類似特性データ群Ｌの特性値の平均値Ｌ_ａｖは次の（数３）で表される。

平均値Ｌ_ａｖにおける各領域における値のばらつきはＬ_ａｖの標準偏差σ_Ｌａｖで表されるので、上記基板の分布による（ｐ_ｉ−ｐ_ａｖ）の違いを抑えるには、（ｐ_ｉ−ｐ_ａｖ）をσ_Ｌａｖで正規化すればよい。次に、分布形状パターンの違いによる（ｔ_ｉ−ｔ_ａｖ）のばらつきであるが、これも（ｐ_ｉ−ｐ_ａｖ）と同様にＴにおける領域毎の濃度値の標準偏差σ_Ｔで（ｔ_ｉ−ｔ_ａｖ）を正規化すればよい。以上のことから、図７中のステップＳ１０３２では、分布形状データＴを次の（数４）で調整した分布形状データＴ’を算出する。

また、ステップＳ１０４では、上記調整した分布形状データＴ’と特性データＰの共分散を用いて類似度ｓを評価する。

Next, FIG. 11 and FIG. 12 show the concept of similarity in step S104. FIG. 11 plots the distribution shape pattern 1101, the analysis target substrate characteristic data 1102, and another analysis target substrate characteristic data 1103 on a graph. As the distance between the characteristic data and the origin of the graph increases, the characteristic values of region 1 and region 2 also increase. That is, the distance between the data on the graph and the origin represents the overall size of the characteristic value, for example, the number of substrate defects in the case of pattern inspection. In FIG. 11, both the characteristic data 1102 and the characteristic data 1103 are similar in distribution shape to the distribution shape pattern 1101, and the characteristic data 1103 has a larger characteristic value than the characteristic data 1102. In this example, the characteristic data 1103 is evaluated with a higher degree of similarity than the characteristic data 1102 with respect to the distribution shape pattern 1101. FIG. 12 shows an example in which the characteristic data 1201 and the characteristic data 1202 are shifted in direction from the distribution shape pattern. In such a case, in order to determine how much the characteristic data 1201 and characteristic data 1202 have a characteristic value with respect to the distribution shape defined by the distribution shape pattern 1101, the origin and the straight line formed by the distribution shape pattern are determined. The size when the characteristic data is projected is used. In FIG. 12, the distance 1203 and the distance 1204 are the sizes of the characteristic data 1201 and the characteristic data 1202 with respect to the distribution shape pattern 1101, respectively. Such similarity can be evaluated using, for example, covariance. When the characteristic data P of the analysis target substrate is expressed by P = (p ₁ ,..., P _g ), the covariance c between the distribution shape pattern T and the characteristic data P is expressed by the following (Equation 2).

Here, c is a large value when (t _i −t _av ) is equal to (p _i −p _av ) and the absolute value is large. That is, since the characteristic data P having a larger characteristic value in a region having a high density value in the distribution shape pattern has a larger value, the concept of similarity in FIG. 12 is satisfied. However, since the difference between the average value p _av of the characteristic values and the characteristic value p _i of each region tends to increase as the characteristic value of the analysis target substrate increases, the substrate corresponding to a certain distribution shape pattern is shown in FIG. ) And the distribution shown in FIG. 6B, the substrate covariance c corresponding to the distribution shape pattern in FIG. 6A has a high value as a whole. In addition, since the distribution of the characteristic value for each region differs depending on the distribution shape pattern, the value of (t _i −t _av ) varies for each distribution shape pattern. For these reasons, the covariance c cannot be quantitatively evaluated if the distribution shape pattern is different or the distribution of the characteristic data of the substrate corresponding to the distribution shape pattern is different. Therefore, in order to suppress the difference in the (p i _{-p av)} by distribution of the substrate, using a similar distribution characteristic data group extracted in step S103. A similar characteristic data group L composed of n pieces of characteristic data is represented by L = (L ₁ ,..., L _n ), and the j-th characteristic data is represented by L _j = (l _j1 ,..., l _jg ). And At this time, the average value L _av of the characteristic values of the similar characteristic data group L is expressed by the following ( _Equation 3).

Since the variation of the values in each region in average _{L av} is expressed by a standard deviation sigma _Lav of _{L av,} to suppress the difference in the _(p i _{-p av)} by distribution of the _substrate, (p i _{-p av} ) _May be normalized by σ _Lav . Next, a variation in due to a difference in the distribution shape pattern _(t i _{-t av),} which is also _(p i _{-p av)} in the same manner as in the standard deviation sigma _T of the density values of each region in the T _{(t i} -T _av ) may be normalized. From the above, in step S1032 in FIG. 7, the distribution shape data T ′ obtained by adjusting the distribution shape data T by the following (Equation 4) is calculated.

In step S104, the similarity s is evaluated using the covariance of the adjusted distribution shape data T ′ and the characteristic data P.

As another method for obtaining the quantitative similarity, there is a method using the average value L _av of the characteristic data of the similar characteristic data group.

この（数６）で求めた分布形状パターンＴ’を用いると、より少ない計算量で定量的な類似度を評価することができる。 FIGS. 13A and 13B show the relationship between the number of defects in the rectangular area and the average number of defects in the pattern inspection of the substrate. FIG. 13A is an example of a substrate similar to a distributed shape pattern generated with a relatively large number of defects, and FIG. 13B is an example of a substrate similar to a distributed shape pattern generated with a relatively small number of defects. is there. Under normal conditions, there are almost no defects on the substrate, and if any abnormality occurs in the manufacturing process, defects are concentrated locally, so many of the rectangular areas on the substrate are defective. The region 1301 has a very small number and the region 1302 has an extremely large number of defects. As shown in FIGS. 13A and 13B, as the number of defects in the substrate increases, the difference between the average number of defects in each region and the number of defects in each region also increases. Therefore, the distribution shape data may be adjusted using the following (Equation 6) instead of (Equation 4).

By using the distribution shape pattern T ′ obtained by (Equation 6), it is possible to evaluate the quantitative similarity with a smaller amount of calculation.

  In this way, according to the covariance c, even when the distribution shape pattern is different or the distribution of the characteristic data of the substrate corresponding to the distribution shape pattern is different, the quantitative similarity can be evaluated. .

  Next, as a second embodiment, as another method of evaluating the similarity in step S104 in FIG. 1, not only the size of the characteristic value of the analysis target region but also the analysis of the region having a high density value in the distribution shape pattern is analyzed. The similarity when the distribution shape of the target characteristic data is taken into account will be described.

  When the similarity is calculated using (Equation 5) in step S104 of FIG. 1, data having extremely large characteristic values is evaluated highly even if the distribution shape is not similar to the distribution shape pattern. is there. FIG. 14 shows the characteristic data 1401 of the analysis target substrate similar to the distribution shape pattern 1101 and the characteristic value of the distribution shape pattern 1101 which is not similar to the distribution shape pattern 1101 and compared to the characteristic data 1401. The degree of similarity of the large characteristic data 1402 is evaluated. When each point has a positional relationship as shown in FIG. 14, the size when projected in the direction of the distribution shape pattern 1101 is larger than the characteristic data 1401. Therefore, step S104 in FIG. 1 may be composed of two steps as shown in FIG.

  In step S1041, the relative magnitude relationship between the density values of the distribution shape pattern is compared with the relative magnitude relationship between the characteristic values of the characteristic data of the analysis target substrate, and the characteristic distribution of the characteristic data of the analysis target pattern is the distribution shape pattern. It is determined whether it is similar to the distribution of density values. Since this process can be evaluated by the same method as in S1031 in FIG. 7, for example, the determination is performed based on the correlation coefficient between the characteristic data and the distribution shape pattern.

  In step S1042, the similarity between the characteristic data and the distribution shape pattern is evaluated in consideration of the determination result in step S1041. In this process, for example, similarity is calculated using only characteristic data (expression 5) such that the determination result obtained in step S1041 is equal to or greater than a predetermined threshold value, and the similarity of other characteristic data is calculated. Set to 0. The threshold value represents an allowable amount with respect to a difference between the distribution of the distribution shape pattern and the characteristic data, and can be set freely according to the application purpose of the present invention. Usually, the allowable value is higher than the threshold value determined in step S1031. Set a wide range. With this setting, as shown in FIG. 16, in step S1031, characteristic data is extracted from the characteristic distribution range 1601 very similar to the distribution shape pattern 1101, and the distribution shape pattern 1101 is adjusted in step S1032, and step S104. Then, regarding the shape of the characteristic distribution, the size of the characteristic data is evaluated by expanding the allowable range to a wider range 1602 than the processing in step S1031.

  By performing such processing, it is possible to prevent the similarity data from being erroneously highly evaluated for characteristic data having extremely large characteristic values, and to perform distribution analysis with higher accuracy.

  FIG. 17 shows an internal configuration of the similarity evaluation unit 809 in the distribution analysis apparatus 800 suitable for carrying out the distribution analysis method described in the second embodiment. The characteristic distribution determination unit 1701 uses the distribution shape pattern after density value adjustment received from the pattern adjustment unit 808 in the distribution analysis apparatus 800 and the analysis target substrate characteristic data received from the characteristic data calculation unit 806 in FIG. The processing in step S1041 is performed to determine whether the distribution shape pattern is similar to the characteristic data, and the determination result, the distribution shape data, and the characteristic data are sent to the similarity calculation unit 1702. The similarity calculation unit 1702 performs the process of step S1042 on the determination result received from the characteristic distribution determination unit 1701, the distribution shape pattern, and the characteristic data, and the characteristic value that forms the distribution shape pattern and the characteristic data. The degree of similarity with the distribution is evaluated, and the evaluation result is sent to the output device 810 connected to the distribution analysis device 800.

  With the distribution analysis apparatus 800 having this configuration, the distribution analysis method according to the second embodiment can be implemented.

  Next, as Example 3, using the substrate analysis result obtained by the distribution analysis method described above, an abnormality that estimates a manufacturing facility (that is, an apparatus in which an abnormality has occurred) that has caused a characteristic distribution in the manufacturing process A facility estimation method will be described.

  FIG. 18 schematically shows an example of a manufacturing process of a substrate to which this abnormal facility estimation method is applied. This manufacturing process includes a plurality of manufacturing steps a, b, c, and d that are sequentially performed on the substrate. The pattern inspection process e is performed after the completion of these manufacturing processes.

  In a certain manufacturing process included in the manufacturing process, in this example, the manufacturing processes b and d, a plurality of facilities each capable of executing the manufacturing process are arranged. In this example, three facilities, Unit 1, Unit 2, and Unit 3, are arranged in step b, and two units, Unit 1, and Unit 2, are arranged in step d. Each substrate that has flowed into the manufacturing process as the manufacturing process progresses is processed at any time by any equipment arranged in the manufacturing process in order to increase the production capacity. It is not determined which equipment is used to process each substrate in the process. In the case of such an arrangement, when it is examined in which equipment the substrate inspected in the pattern inspection step e in FIG. 18 is processed in each manufacturing step, as shown in FIG. The ratio of the substrates processed in each facility is 1/3, and the ratio of the substrates processed in each facility in step d is 1/2. Hereinafter, information such as which equipment has processed each substrate in each process is referred to as a processing history.

  On the other hand, if abnormalities occur in any of the equipment in the process and defects are concentrated in a specific area on the board, extracting the board with a similar characteristic distribution will cause the equipment that caused the problem to exist Then, the board | substrate processed with the causal equipment will be extracted unevenly. However, since the equipment for processing these substrates in processes other than the causal process is not constant, the above-described bias does not occur in other processes. For example, out of the substrates that have undergone the manufacturing process of FIG. 18, when a group of substrates in which defects are concentrated in the upper right part of the first machine in step b is extracted, as shown in FIG. Then, the ratio of the substrates processed in Unit 1 (4/5 in this example) was high, and the ratio of substrates processed in Units 2 and 3 (1/10 in this example) was reduced. There is a bias in the number of processed substrates. There is no deviation between the facilities in the ratio of the substrates processed in step d for the same substrate group. From this, it can be seen that abnormal equipment can be estimated by extracting a group of substrates having similar characteristic distributions.

  FIG. 20 shows a method for estimating the abnormal equipment that is the cause by using the distribution analysis result of the substrate characteristic value by the above-described distribution analysis method on the assumption of such a situation.

  First, in step S2001 including a series of steps S101 to S104 in FIG. 1, the degree of similarity with respect to the distribution shape to be estimated is evaluated for each board included in the board group to be subjected to abnormal equipment estimation. However, at step S102 and step S103, a distribution shape pattern adjustment characteristic data group is extracted from the characteristic data of all the substrates included in the substrate group, and the distribution shape pattern adjustment characteristic data group is used to extract the distribution shape pattern. Adjust the density value. In step S104, the similarity to the distribution shape pattern is evaluated for all the substrates included in the substrate group.

  In step S2002, substrates whose similarity obtained in step S2001 is equal to or greater than a predetermined threshold are extracted, and those substrates are set as a substrate group corresponding to the distribution shape pattern to be estimated. As the threshold value, for example, a value that allows a human to visually determine that the substrate corresponds to the distribution shape pattern to be estimated is employed. Since the similarity is evaluated in consideration of the variation factor in step S2001, the threshold value can be uniquely determined even when the distribution shape pattern and the substrate group to be estimated are different.

  In step S2003, the process history in each manufacturing process is acquired about the board | substrate included in the board | substrate group extracted by step S2002. Since this information is usually collected and managed by a process information collection system installed in the manufacturing process, it can be easily obtained.

  In step S2004, the processing history acquired in step S2003 is totaled for each process of the manufacturing process, and the number of processes for each facility is calculated.

  Finally, in step S2005, the number of processed sheets for each facility in each process is compared, and the facility having the largest deviation in the number of processed sheets is statistically examined to estimate that it is an abnormal facility. If there is no significant difference in the number of processed sheets, it is determined that the characteristic distribution is not caused by a specific facility.

  According to this abnormal equipment estimation, since the deviations in the processing history are compared from substrates having similar characteristic distributions, it is possible to perform abnormal equipment estimation with high accuracy.

  FIG. 21 shows a configuration of an abnormal equipment estimation device 2100 according to an embodiment of the present invention suitable for implementing the above-described abnormal equipment estimation method. The abnormal facility estimation apparatus 2100 includes a pattern registration unit 2102, a data collection unit 2103 as an input unit, a characteristic data calculation unit 2105, a history data storage unit 2106, a characteristic data registration unit 2107, a distribution analysis unit 2108, and a board group extraction unit 2109. The processing frequency calculation unit 2110 and the abnormal equipment estimation unit 2111 are configured. In addition, an input device 2101, an inspection information collection system 804, a process information collection system 2104, and an inspection device 805 are connected to the data collection unit 2103, and an output device 2112 is connected to the abnormal equipment estimation unit 2111.

  The input device 2101 is composed of, for example, a keyboard and a mouse. In this example, the input device 2101 is a target for the abnormal equipment estimation device 2100, the range of the inspection date and time of the board that is the target of abnormal equipment estimation, the number of defects detected by the inspection device, or the target of abnormal equipment estimation. It is used to input the conditions of the substrate for which abnormal equipment is to be estimated such as the range of the manufacturing process. Also, the distribution shape pattern used for abnormal equipment estimation is registered in the abnormal equipment estimation apparatus 2100 using the input device 2101.

  The pattern registration unit 2102 has the same function as the pattern registration unit 802 of the distribution analysis apparatus 800 shown in FIG. 8, and specifically receives the distribution shape pattern defined by the input device 2101 and records it in the database. The information received from the input device 2101 is not only the distribution shape pattern but also information related to the distribution shape pattern, for example, information about the device that causes the distribution shape pattern, and is recorded in association with the distribution shape pattern. Also good.

  The data collection unit 2103 collects, from the inspection information collection system 804 and the inspection apparatus 805, inspection data of the board group that matches the conditions of the estimation target board transmitted to the abnormal equipment estimation apparatus 2100 from the input device 2101 and the identification ID of the board. To the characteristic data calculation unit 2105. At this time, information such as the inspection date and time may be collected together with the inspection data. Further, the processing history data of the substrate group is collected from the process collection system 2104 and sent to the history data recording unit 2106 in association with the substrate ID.

  The inspection information collection system 804 is the same as the inspection information collection system 503 connected to the distribution analyzer 800. The data collection unit 2103 of the abnormal equipment estimation apparatus 2100 collects necessary inspection data from the inspection information collection system 804.

  The inspection device 805 is the same as the inspection device 805 connected to the distribution analysis device 800. The data collection unit 2103 of the abnormal equipment estimation apparatus 2100 collects necessary inspection data from the inspection apparatus 805.

  The process information collection system 2104 is a process information collection system that collects process history information from a processing apparatus arranged in a manufacturing process. The data collection unit 2103 of the abnormal equipment estimation apparatus 2100 collects necessary inspection data from the process information collection system 2104.

  The characteristic data calculation unit 2105 has the same function as the characteristic data calculation unit 806 of the distribution analysis apparatus 800. Specifically, the inspection data of the substrate collected by the data collection unit 2103 from the inspection information collection system 804 or the inspection apparatus 805 is used. The characteristic data is created from the data and transferred to the characteristic data group registration unit 2107.

  The history data storage unit 2106 stores the processing history of the estimation target substrate acquired by the data collection unit 2103 from the process information collection system 2104 in association with the substrate identification ID. The processing history is sent to the processing frequency calculation unit 2110 as necessary.

  The characteristic data group registration unit 2107 has the same function as the adjustment characteristic data group registration unit 807 of the distribution analysis apparatus 800. Specifically, the characteristic data calculated by the characteristic data calculation unit 2105 from the board group subject to abnormal equipment The group is stored in association with the identification ID of the substrate.

  The distribution analysis unit 2108 acquires the characteristic data group registered in the pattern registration unit 2101 and the characteristic data group registration unit 2107 and performs a series of processes in steps S102 to S104 in FIG. The data is transferred to the board group extraction unit 2109 in association with the board ID. The distribution analysis unit 2108 includes a pattern adjustment unit 808 and a similarity evaluation unit 809 in the same manner as the distribution analysis apparatus 800.

  The board group extraction unit 2109 extracts, from the similarity received from the distribution analysis unit 2108, characteristic data of the board having a similarity that is equal to or higher than a predetermined threshold, and the identification ID of the board is set as a processing frequency calculation unit 2110. To pass.

  The processing frequency calculation unit 2110 acquires the substrate processing history extracted by the substrate group extraction unit 2109 from the history data storage unit 2106 using the identification ID of the substrate received from the substrate group extraction unit 2109, and is arranged in each process. The number of processed sheets for each facility is calculated and passed to the abnormal facility estimation unit 2111.

  The abnormal equipment estimation unit 2111 performs statistical processing on the number of processed sheets received from the processing frequency calculation unit 2110, detects equipment having a large ratio of processed sheets, and outputs the detected result to the output device 2112 as an abnormal equipment estimation result.

  The output device 2112 outputs an abnormal facility estimation result by the abnormal facility estimation device 2100 through a monitor, paper output, a magnetic disk, a portable semiconductor memory, or the like.

  A plurality of distribution shape patterns may be registered in the pattern registration unit 2102 and abnormal equipment estimation may be performed for each distribution shape pattern.

  Moreover, when all the information necessary for abnormal equipment estimation can be acquired from any of the inspection information collection system 804, the process information collection system 2104, and the inspection apparatus 805, only the apparatus or system necessary for acquisition may be connected.

  Further, the input device 2101 and the output device 2112 may be the same device, for example, a device in which a display unit serving as an output device is provided with a keyboard serving as an input device and configured integrally. Further, the input device 2101 and the output device 2112 may be included in the abnormal equipment estimation device 2100.

  Also, the output device 2112 receives and outputs information other than information necessary for abnormal facility estimation from the input device 2101, inspection process information collection system 804, process information collection system 2104, and inspection device 805 through the abnormal facility estimation device 2100. Also good.

  The output device 2112 may receive and output the analysis result of the distribution analysis unit 2108 from the abnormal equipment estimation device 2100.

  As described above, the pattern registration unit 2102, the characteristic data calculation unit 2105, the characteristic data group registration unit 2107, and the distribution analysis unit 2108 are respectively the pattern registration unit 802, the characteristic data calculation unit 806, and the adjustment data for the distribution analysis device 800 described above. It has the same functions as the characteristic data group registration unit 807, the pattern adjustment unit 808, and the similarity evaluation unit 809. Therefore, as shown in FIG. 22, these components can be configured separately from the abnormal facility estimation apparatus 2100 as the distribution analysis apparatus 800. In the configuration shown in FIG. 22, abnormal equipment estimation apparatus 2100 includes only data collection unit 2103, history data storage unit 2106, board group extraction unit 2109, processing frequency calculation unit 2110, and abnormal equipment estimation unit 2111. The input device 2101 receives the distribution analysis result and identification ID of each substrate from the distribution analysis device 800 and transmits them to the data collection unit 2103. The data collection unit 2103 receives the distribution analysis result and identification ID of each substrate from the input device 2101 and passes them to the substrate group extraction unit 2103. Further, the processing history of each substrate is acquired from the process information collection system 2104 from the identification ID of each substrate, and is transferred to the history data storage unit 2106. The subsequent processing is performed in the same manner as the processing by the abnormal equipment analysis apparatus 2100 in FIG.

  According to this abnormal equipment estimation apparatus 2100, the above-mentioned abnormal equipment estimation method can be implemented.

  In addition, you may construct | assemble as a program for making a computer perform the above-mentioned abnormal equipment estimation method.

  Further, such a program may be recorded and distributed on a computer-readable recording medium such as a CD-ROM. By installing the program in a general-purpose computer, the abnormal equipment estimation method can be executed by the general-purpose computer.

It is a figure which shows typically the concept of the distribution analysis method of one Embodiment of this invention. It is a figure which shows typically the concept of a rectangular area and a distribution characteristic pattern. It is a figure which shows the example of the test result of a pattern test | inspection. It is a figure which shows typically the concept of the characteristic data in one Embodiment of this invention. It is a figure which shows the example of the board | substrate applicable to a distribution characteristic pattern. It is a figure which shows the difference in the characteristic value by a distribution characteristic pattern. It is a figure which shows the density value adjustment process of the distribution shape pattern of one Embodiment of this invention. It is a figure which shows the structure of the distribution analyzer of one Embodiment of this invention. It is a figure which shows the internal structure of the pattern adjustment part in FIG. It is a figure explaining the meaning of the direction of the origin and characteristic data in a graph. It is a figure explaining the meaning of the distance of the origin and characteristic data in a graph. It is a figure which shows the concept of the similarity of one Embodiment of this invention. It is a figure which shows typically the concept of the density adjustment method of the distribution shape pattern of one Embodiment of this invention. It is a figure which shows the similarity in case the characteristic data with an extremely large characteristic value exist. It is a figure which shows the similarity evaluation process of one Embodiment of this invention. It is a figure which shows typically the concept of the similarity evaluation process in FIG. It is a figure which shows the internal structure of the similarity evaluation part in FIG. It is a figure which shows typically the example of the manufacturing process of the board | substrate used as the application object of the abnormal facility estimation method of one Embodiment of this invention. It is a figure which shows the example which compared the number of sheets processed with the manufacturing equipment at the time of the normal in the said manufacturing process, and the time of abnormality. It is a figure which shows typically the concept of the abnormal equipment estimation method in one Embodiment of this invention. It is a figure which shows the structure of the abnormal equipment estimation apparatus of one Embodiment of this invention. It is a figure which shows the modification of the abnormal equipment estimation apparatus of FIG. It is a figure explaining a certain prior art. It is a figure explaining another prior art.

Explanation of symbols

800 Distribution analyzer 2100 Abnormal equipment estimation device

Claims (18)

A distribution analysis method for evaluating whether analysis target data having characteristic values distributed along a plane is similar to a specific distribution shape pattern,
The distribution shape pattern is divided into a plurality of rectangular areas in a lattice shape on a plane corresponding to the plane defining the analysis target data, and the relative magnitude of the characteristic value of the analysis target data is determined for each rectangular area. In order to express the relationship, define a concentration value that can take multiple values within a certain numerical range,
Using the characteristic value of the predetermined distribution shape pattern adjustment characteristic data group, the density value of the distribution shape pattern is adjusted so that the size of the characteristic value of the characteristic data group is reflected in the evaluation,
Comparing the analysis target data with the adjusted distribution shape pattern to obtain a quantitative similarity between the characteristic value distribution of the analysis target data and the density value distribution of the adjusted distribution shape pattern A distribution analysis method characterized by The distribution analysis method according to claim 1,
Adjustment of the above density value
First step of extracting, from the analysis target data group formed by the analysis target data, data whose characteristic value distribution is similar to the density value distribution of the distribution shape pattern as the distribution shape pattern adjustment characteristic data group When,
Using the distribution statistics of the characteristic values of the distribution shape pattern adjustment characteristic data group obtained in the first step and the distribution statistics of the density values of the distribution shape patterns, the density value of the distribution shape pattern A distribution analysis method comprising a second step of adjusting The distribution analysis method according to claim 2,
In the first step, the correlation coefficient between the characteristic value corresponding to each rectangular area of each analysis target data and the density value assigned to each rectangular area of the distribution shape pattern is equal to or greater than a predetermined threshold value. A distribution analysis method, characterized in that simple data is extracted as the distribution shape pattern adjustment characteristic data group. The distribution analysis method according to claim 2,
The statistic of the distribution of characteristic values of the distribution shape pattern adjustment characteristic data group in the second step is a standard deviation of the average value of each characteristic data constituting the characteristic data group for each analysis target data, and the distribution shape The statistic of the pattern density value distribution is the standard deviation of the density value of the distribution shape pattern,
The similarity is given by covariance between a density value given to each rectangular area of the adjusted distribution shape pattern and a characteristic value corresponding to each rectangular area of the analysis target data. Distribution analysis method. The distribution analysis method according to claim 2,
The statistic of the distribution of the characteristic values of the distribution shape pattern adjustment characteristic data group in the second step is an average value for each analysis target data of the characteristic values forming the characteristic data group, and the density of the distribution shape pattern The value distribution statistic is the standard deviation of the density values of the distribution shape pattern,
The similarity is given by covariance between a density value given to each rectangular area of the adjusted distribution shape pattern and a characteristic value corresponding to each rectangular area of the analysis target data. Distribution analysis method. The distribution analysis method according to claim 1,
In order to obtain the above similarity,
A third step of determining whether the relative magnitude relationship of the characteristic values of the analysis target data is similar to the relative magnitude relationship of the density values of the distribution shape pattern;
A distribution analysis method comprising a fourth step of evaluating the similarity using the determination result of the third step. The distribution analysis method according to claim 6,
The third step includes a correlation coefficient between the characteristic value corresponding to each rectangular area of the analysis target data and the density value given to each rectangular area of the distribution shape pattern or the adjusted distribution shape pattern. A distribution analysis method characterized by determining that analysis target data having a value equal to or greater than a predetermined threshold value is similar to the distribution shape pattern. The distribution analysis method according to claim 1,
A distribution analysis method, wherein the characteristic values of the analysis target data are distributed along the surface of the substrate. An abnormal facility estimation method for estimating a manufacturing apparatus in which an abnormality has occurred, from among the manufacturing apparatuses used in a plurality of manufacturing steps sequentially performed on a substrate,
The distribution analysis method according to claim 8 is performed to obtain the similarity,
Based on a processing history representing a manufacturing apparatus that has processed each of the substrates in each of the manufacturing steps, a manufacturing apparatus commonly used for the substrates having the similarity equal to or higher than a predetermined threshold is extracted. An abnormal equipment estimation method. A distribution analysis device that evaluates whether analysis target data having characteristic values distributed along a plane is similar to a specific distribution shape pattern,
The distribution shape pattern is divided into a plurality of rectangular areas in a lattice shape on a plane corresponding to the plane defining the analysis target data, and the relative magnitude of the characteristic value of the analysis target data is determined for each rectangular area. A pattern registration unit that defines and assigns density values that can take multiple values within a certain numerical range to represent the relationship;
Using a characteristic value of a predetermined distribution shape pattern adjustment characteristic data group, and a pattern adjustment unit for adjusting the density value of the distribution shape pattern so that the magnitude of the characteristic value of the characteristic data group is reflected in the evaluation; ,
Similarity in which the analysis target data is compared with the adjusted distribution shape pattern to obtain a quantitative similarity between the distribution of characteristic values of the analysis target data and the distribution of density values of the adjusted distribution shape pattern A distribution analysis device comprising a degree evaluation unit. In the distribution analyzer of Claim 10,
The pattern adjustment unit
A first part for extracting, as the distribution shape pattern adjustment characteristic data group, data whose characteristic value distribution is similar to the distribution of density values of the distribution shape pattern from the analysis target data group formed by the analysis target data When,
The density value of the distribution shape pattern is obtained by using the distribution statistics of the characteristic values of the distribution shape pattern adjustment characteristic data group obtained in the first part and the distribution statistics of the density values of the distribution shape patterns. A distribution analyzing apparatus comprising a second portion for adjusting the angle. In the distribution analyzer of Claim 10,
The similarity evaluation unit
A third part for determining whether the relative magnitude relationship between the characteristic values of the analysis target data is similar to the relative magnitude relationship between the density values of the distribution shape pattern;
A distribution analysis apparatus comprising a fourth portion that evaluates the degree of similarity using a determination result by the third portion. In the distribution analyzer of Claim 10,
A distribution analysis apparatus characterized in that the characteristic values of the analysis target data are distributed along the surface of the substrate. An abnormal facility estimation device that estimates a manufacturing device in which an abnormality has occurred, from among the manufacturing devices used in each of a plurality of manufacturing steps that are sequentially performed on a substrate,
The distribution analysis device according to claim 13 is provided, and the similarity is obtained by the distribution analysis device,
Based on a processing history representing a manufacturing apparatus that has processed each of the substrates in each of the manufacturing steps, an abnormal facility estimation is performed to extract a manufacturing apparatus that is commonly used for the substrates whose similarity is equal to or greater than a predetermined threshold. An abnormal equipment estimation device characterized by comprising a section.   A distribution analysis program for causing a computer to execute the distribution analysis method according to claim 1.   An abnormal equipment estimation program for causing a computer to execute the abnormal equipment estimation method according to claim 9.   A computer-readable recording medium on which the distribution analysis program according to claim 15 is recorded.   A computer-readable recording medium in which the abnormal equipment estimation program according to claim 16 is recorded.

Images