JP2002124445A - Manufacturing method of semiconductor device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To establish a sampling method of completely new analysis data for improving manufacturing yield. SOLUTION: This manufacturing method of a semiconductor device has a treatment process for treating a wafer or batch wafer units, an element characteristic measurement process for measuring the characteristics of an element formed in the treated wafer, a yield calculation process for calculating the yield according to the electrical characteristics of a chip formed in the wafer, and a correlation calculation process for calculating the correlation between the element characteristics and yield, and controls the wafer treatment process based on the calculated correlation. In the method, identification information is added to results measured in the element characteristic measurement process, where the identification information includes identification information for indicating the measured position in the wafer, identification information on the position in a lot where the measured wafer is inserted, identification information of the batch where the lot is inserted, and identification information on the position in the wafer or batch of the lot. The identification information is used for sampling the measurement result, and the correlation between the element characteristics and yield is calculated based on the sampled measurement result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for improving the yield in a pre-process production line for semiconductor products.

[0002]

2. Description of the Related Art In recent years, the product life of semiconductors has been shortened, and the price of the products has tended to fall sharply. For this reason, it is the most important issue for a semiconductor manufacturer to improve the yield early in the period when the product price is high and to secure a constant production amount from the beginning of mass production. As described above, there is an increasing tendency that the vertical startup of the yield is directly linked to the profit, and there is a demand for a shorter failure analysis TAT than before.

[0003] Against the background of strictly required specifications such as high speed and low power consumption, transistors that require a complicated structure and a manufacturing process are used in an LSI. Device characteristics are likely to vary. In order to improve the yield of LSIs, it is necessary to develop a method for analyzing a characteristic failure caused by element characteristics.

The measurement of the electrical element characteristics of an LSI is performed in an inspection process called a wafer inspection. In wafer inspection, a dedicated measuring device called a wafer tester is used to inspect basic characteristics (element characteristics) such as an ON voltage and an ON current of a transistor constituting a semiconductor product. The element for characteristic monitoring measured by the wafer tester is called TEG, and is usually arranged in an area called scribe between product chips.

FIG. 1 shows an example of a sectional view of an LSI, and FIG. 2 shows an outline of device characteristics measured in a general wafer inspection process. Item (1) in FIGS. 1 and 2 corresponds to the characteristics of the MOS transistor. For example, as shown in the figure, the ON current flowing between the source and the drain when an ON voltage is applied to the gate electrode is measured by wafer inspection. The on-voltage and on-current are items representing the most basic characteristics of a MOS transistor, and are determined when designing the transistor. Item (2) is a through hole for connecting a word line to a wiring layer, a diffusion layer to a wiring layer, and a wiring layer to a wiring layer. Item (3) is an item for inspecting the performance of various layer layers such as an oxide film. It is. That is, in the wafer inspection, the characteristics of the passive element such as the resistance of the through hole and the resistance of the film (sheet resistance) are measured. In the wafer inspection process, one or more TEGs on a wafer are used to measure the device characteristics of these tens to hundreds of items.

[0006] In contrast to a wafer inspection process for inspecting the basic device characteristics of a semiconductor product on a wafer-by-wafer basis, a probe inspection process involves logical operations, operating frequencies, power consumption, and the like.
A pass / fail decision is made for all chips on the product wafer. The percentage of non-defective chips in the total number of chips on the wafer obtained as a result of the probe inspection is called a probe yield (hereinafter, wafer yield or yield), and is used as the most basic index for performing failure analysis.

A method of analyzing a defect caused by element characteristics by a correlation analysis of a yield and a wafer inspection result is described in IE.
EE Int. Work. on Statica
1 Metrology, 1997, pp. 56-61, "A SYSTEMATIC APPROACH TO
IDENTIFY CRITICAL YIELDSEN
SITIVE PARAMETRIC PARAMETE
RS ".

[0008]

However, in the prior art, for example, the average value of the measured values of the device characteristics at one or more locations is used for the correlation analysis with the wafer yield, and the device characteristics between the measurement positions are used. Were averaged, and could not be extracted as a cause of characteristic failure.

It is an object of the present invention to establish a completely new analysis data sampling method, thereby improving the production yield. In particular, a method of narrowing down the cause of the variation in the extracted characteristic failure factor is established, thereby improving the manufacturing yield.

[0010]

SUMMARY OF THE INVENTION In order to achieve the above-mentioned object, the present invention is constituted as described in the appended claims.

For example, by creating analysis data to which stratified variables corresponding to each variation component between batches, within a batch, within a lot, and within a wafer of semiconductor product characteristics, analysis is performed based on stratified variables. The objective data is sampled to clarify the element characteristics that affect the yield, and the yield is improved by taking early measures for the element characteristics.

Also, for example, between batches, within batches, and within lots of semiconductor product characteristics, based on layer-by-layer variables corresponding to each variation component among semiconductor product characteristics between batches, within batches, within lots, and within wafers. By plotting the measured values of the characteristics in each region on the wafer along the axis of change of the characteristics corresponding to each of the variation components, that is, the batch number, the shelf number, and the wafer number, the process of narrowing down the characteristics variation is narrowed down, And the number of steps required for analysis.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A specific method for extracting a cause of characteristic failure and narrowing down a cause process will be described with reference to an example of characteristic failure analysis of an LSI.

FIG. 3A shows the cause of the characteristic variation. As shown in FIG. 1A, the characteristic variation is composed of four types of variation components between batches, within batches, within lots, and within wafers according to the processing mode of the process. Hereinafter, each variation component will be described.

(1) Batch-to-batch variation A process (batch process) in which a plurality of lots (one lot is composed of a plurality of wafers) are processed collectively.
Occurs in For example, it occurs in a batch heat treatment step for forming a gate oxide film shown in FIG. This is caused by a change in process conditions according to the order of construction.

(2) Intra-batch variation The same occurs in a batch process. For example, this is caused by a difference in heat history or a difference in the position of the lot to be processed in the furnace.

(3) Intra-lot variation This mainly occurs in a process of processing one wafer at a time (single wafer process). For example, it occurs in a gate etching step for forming the gate shown in FIG. This also occurs because the process conditions change according to the order of construction.

(4) In-wafer variation The variation is caused by both the batch process and the single-wafer process. This is caused by the state of the manufacturing apparatus, and is caused by the inability to uniformly process the inside of the wafer.

Therefore, when the correlation analysis between the wafer yield and the wafer inspection result is performed using the sample data lacking any of the variation components, for example, the variation component in the wafer shown in FIG. In the case where the characteristic failure occurs due to the in-wafer variation, the characteristic failure factor cannot be extracted. Or, the date and time of lot delivery in any process,
For example, when a lot subjected to the probe inspection is sampled within a certain period, and when all the lots to be analyzed belong to different batches and are processed at the same position in the furnace, the lot shown in FIG. Variation components in the batch are not included in the sample data. For this reason, in the case where characteristic failure occurs due to variation within the batch,
The cause of characteristic failure cannot be extracted.

As described above, depending on the method of sampling the data to be analyzed, the cause of the characteristic failure may not be extracted. Therefore, in extracting the characteristic failure, it is necessary to sample the analysis target data so as to include all of these variation components. Alternatively, it is necessary to perform sampling in consideration of a variation factor. By performing a correlation analysis of, for example, the yield and the wafer inspection result on the analysis data sampled in this way, it is possible to increase the possibility of extracting a characteristic failure factor.

FIGS. 3 (2) and 3 (3) show a method of generating analysis data and a format of the analysis data for enabling data sampling suitable for extracting a characteristic failure factor according to the present invention.

First, FIG. 3 (2) shows a method of generating a new index corresponding to the characteristic variation component in the wafer and changing to the wafer yield. First, the region on the wafer is virtually divided into a plurality of regions including the TEG measurement position. The divided area is hereinafter referred to as an intra-wafer. An identification tag is added to each intra-wafer so as to identify which region on the intra-wafer corresponds. For example, FIG.
As shown in (2), UL, U
Tags such as R, C, LL, and LR (hereinafter referred to as area identification tags) are associated with each other. Next, the yield for each intra-wafer is calculated. That is, the ratio of the number of non-defective chips in the intra wafer to the number of chips included in each intra wafer is calculated. Hereinafter, the yield of the intra-wafer is referred to as an intra-wafer yield.

Next, these intra-wafer yields and area identification tags are shaped into one record and one intra-wafer format as shown in FIG. In each column of “position information”, “intra-wafer yield”, and “wafer inspection result” in FIG. 3 (3), a region identification tag, an intra-wafer yield, and a corresponding TEG measurement result in the intra-wafer are described. I do.

Next, as information corresponding to a characteristic variation component in a lot, a number (hereinafter, referred to as a wafer number) representing a processing order of wafers in a single wafer process is used. The wafer number is described in the “wafer number” column in FIG.

Next, a shelf number in the batch heat treatment step is used as information corresponding to the characteristic variation component in the batch. The shelf number is tag information indicating the position in the furnace of the heat treatment apparatus. By referring to the shelf number, it is possible to know at which position in the furnace the lot has been set. The shelf number is described in the “shelf number” column in FIG.

Next, as information corresponding to the characteristic variation component between batches, it is assumed that the same tag information is assigned to lots constituting the same batch. At this time, a number indicating the processing order of the batch (hereinafter referred to as a batch number) is used as tag information. This batch number is described in the “batch number” column in FIG. Here, when a plurality of batch heat treatment apparatuses are used, tag information indicating which apparatus has been processed may be added as a new stratified variable. However, in this embodiment, the description will be simplified. Therefore, it is assumed that only one batch heat treatment apparatus is used.

In this embodiment, a new “batch number” and a new “batch number” corresponding to the variation component of each characteristic shown in FIG.
“Batch bin”, “Wafer number”, “Position information”, and “Intra-wafer yield” were introduced as stratified variables. A stratified variable is a variable used as a reference for sampling data to be analyzed when performing various statistical processes. In this embodiment, the batch number, the shelf number, the wafer number, and the position information are used. However, any variable may be used as long as it represents information equivalent to the contents represented by these variables. For example, the order of insertion into the batch heat treatment apparatus may be used instead of the shelf number, and cassette position information indicating the storage position in the wafer cassette may be used instead of the wafer number. Instead of the intra-wafer yield, a specific defect occurrence rate or the like in the intra-wafer may be used.

In extracting the cause of the characteristic failure, the analysis data of FIG. 3 (3) is used, and by referring to these stratified variables, all the variation components shown in FIG. 3 (1) are included. In this way, it is possible to sample the data to be analyzed. For example, by sampling records having continuous batch numbers, for example, batch numbers 1, 2, and 3, with reference to the stratified variable “batch number”, data sampling including all variation components is possible. By performing a correlation analysis between the “intra-wafer yield” and the “wafer inspection result” for the record, it is possible to extract the cause of the characteristic failure as compared with the case where the analysis data is sampled by the lot payout date and time or random sampling. The possibilities increase.

Referring to FIG. 4A, a correlation analysis between “intra-wafer yield” and “wafer inspection result” is performed using data sampled so as to include all variation components with reference to the layer-by-layer variable “batch number”. The following is an example. In this example, the diffusion layer resistance is related to the intra-wafer yield, and when the diffusion layer resistance becomes 25Ω or less, a phenomenon in which the intra-wafer yield is low can be confirmed, and the diffusion layer resistance is extracted as a characteristic failure factor. be able to. In the semiconductor used in this example, it is known that the variation in the resistance of the diffusion layer is mainly due to the variation in the wafer.
For the same lot as the target of this analysis,
FIG. 4B shows an example of a conventional correlation analysis performed using the average value of the wafer yield and the TEG measurement result. In the conventional analysis method, the average value of a plurality of measurement points on the wafer is used as the measured value of the diffusion layer resistance, so that the variation information within the wafer is excluded. As shown in FIG. The relationship between the yield cannot be grasped, and the resistance of the diffusion layer cannot be extracted as the cause of the characteristic failure.

Next, an example of an apparatus for extracting a cause of characteristic failure using the analysis data shown in FIG. 3 (3) will be described.

FIG. 5 shows a configuration example of the apparatus. This apparatus includes a central processing unit 501, input devices such as a keyboard 502 and a mouse 503, output devices such as a display 504 and a printer, and storage devices 505 such as a hard disk and a database.
Composed of The central processing unit 501 of the present apparatus stores a characteristic failure factor extraction processing program 508. This device is used for wafer tester 506 and probe tester 50
7, via a network 509 or a recording medium 510 such as a floppy (registered trademark) disk or MO, a wafer yield, an intra-wafer yield, or a wafer inspection result is obtained and stored in the storage device 505.

FIG. 6 shows an example of data of a wafer inspection result sent from the wafer tester 506, FIG. 7 shows an example of data of a probe inspection result sent from the probe tester 507, and FIG. The data example of the batch organization information and the shelf number information sent from the processing device or another system such as a factory automation system is shown.

In this embodiment, the wafer inspection result data of FIG. 6 includes a lot number 601, a wafer number 602, a TEG
The position information 603 indicating the measurement position and the measurement result 604 of the element characteristics to be inspected in the wafer inspection process are described in one record. The probe test result data in FIG.
Lot number 701, wafer number 702, position information 70
3. A format in which the intra-wafer yield 704 is described in one record. The data of the batch organization information and the shelf number information in FIG. 8 has a format in which a batch number 801, a shelf number 802, and a lot number 803 are described in one record. These data formats may be other equivalent formats. The inspection data shown in FIGS. 6 and 7 and the batch organization information and shelf number information shown in FIG. 8 are stored in the storage device 505 in FIG. 5 via a recording medium such as a network, a floppy disk, and an MO as appropriate. .

When the apparatus is started, first, the storage device 50
The inspection data, batch composition, and shelf number information stored in No. 5 are shaped to create the analysis data shown in FIG. The analysis data used by this apparatus has a format in which a batch number, a shelf number, a lot number, a wafer number, position information, an intra-wafer yield, and a wafer inspection result are described in one record. The data format may be another equivalent format. After creating the analysis data shown in FIG. 3C, the apparatus displays the sampling instruction screen shown in FIG. 9 on the display 504, and enters a standby state for input from the user.

The user inputs conditions for sampling the data to be analyzed from the sampling instruction screen shown in FIG. In the example shown in FIG. 9, from the lots constituting the batches of batch numbers 1 to 5, the lots set in the shelf numbers 1 to 3 of the furnace in the batch heat treatment step are selected, and further, the upper left ( UL) and an intra-wafer corresponding to the area of the central part (C) are instructed to be sampled as data to be analyzed. In this example, the range of the sampling condition is specified, but the format may be such that the target batch, shelf number, lot number, wafer number, and area are individually listed. After inputting the above sampling conditions,
The user presses the OK button to instruct the present apparatus to start the analysis processing. In this case, all the analysis data satisfying the input conditions are extracted.

The present apparatus receives an instruction from the user to start the analysis process, selects a record that matches the sampling conditions input by the user on the sampling instruction screen of FIG. 9 from the analysis data shown in FIG. Creates a scatter plot of intra-wafer yield and wafer inspection results for records, or calculates statistical indices that indicate the strength of the relationship between intra-wafer yield and wafer inspection results, such as correlation coefficient calculation processing. And display the processing result on the display 50.
4 and the process ends.

As described above, by specifying two or more batch numbers on the input screen for batches of stratified variables on the sampling instruction screen shown in FIG. 9, for example, by the characteristic failure factor extracting device, Between batches shown, within batches,
The analysis target data can be intentionally sampled so as to include all the characteristic variation components in the lot and wafer, and the characteristic failure is compared with the case where the analysis target data is sampled randomly or by other methods. The possibility of extracting a factor can be increased. More precisely, the batch number is limited, and the analysis data including all the remaining characteristic variation components can be sampled in the limited analysis target.

Therefore, as another effect of this embodiment, FIG.
From among the characteristic variation components between batches, within batches, within lots, and within wafers shown in (1), the influence of a specific variation component on the analysis can be removed, and the analysis accuracy can be improved. For example, on the sampling instruction screen of FIG. 9, after specifying one or more batch numbers on the input screen of the batch of the layer variable, a specific area, for example, only the central portion of the wafer, is specified on the input screen of the area of the layer variable. As a result, of the four types of variation components shown in FIG.
It is possible to remove the influence of the variation within the wafer.

As described above, if items considered to be the causes of variation are added to the inspection data and managed, sampling can be performed so as to include the characteristic variation components between batches, within batches, within lots, and within wafers. It becomes possible, and it is possible to specify a failure factor that has not been revealed so far. Further, since a sampling target can be specified, analysis data from which a specific variation component has been removed can also be sampled, whereby a defect factor that has not been revealed can be specified. There is an effect even if only sampling is performed so as to include at least one of these variation components in consideration of the variation factor. Although described in the embodiment, it is most effective to sample all the remaining analysis data by taking the batch identification information into consideration.

Next, a description will be given of a method of narrowing down the cause of variation in the measured characteristic value of the characteristic extracted as the cause of the defect.

As shown in FIG. 3A, the variation in characteristics is composed of four types of variation components between batches, within batches, within lots, and within wafers. Among these, variations among batches, within batches, and within lots are caused by characteristics changing according to the order of batch start, the position within batch of batch, and the order of wafer start. Therefore, the characteristic change axis corresponding to each variation component, that is, the batch number,
By plotting the measured values of the characteristics in each region on the wafer along the shelf number and the wafer number, it is possible to visualize the variation in the characteristics.

FIG. 10 shows an example in which variations in characteristics are visualized.

In this example, spatial coordinates are set with the batch number, the shelf number, and the characteristic measured value as the X axis, the Y axis, and the Z axis, respectively. Lot position), and any expression form may be used as long as the state of change in characteristics along the wafer number can be visualized. In this example, the wafer number is set on the same axis as the shelf number. FIG. 7 is a plot of characteristic change surfaces obtained by plotting measured values of characteristics for each of the regions A, B, and C on the wafer on the coordinate axes. By referring to this diagram, the user can intuitively understand which characteristic component is caused by variation among batches, within a batch, within a lot, or within a wafer, and obtain a hint to control the variation. Can be.
For example, in the example shown in the figure, the inter-batch variation shown in (1) and the intra-lot variation shown in (3) are all small, and do not significantly affect the characteristic variation. However, the changing surfaces are completely separated between the regions A, B, and C on the wafer, and the variation in the characteristics is mainly due to the variation in the wafer. The variation in the batch shown in (2) also indicates the variation in the characteristics. Affects variation. From the above results,
It can be inferred that the variation in the characteristics is caused by the variation in the batch and in the wafer. Here, from the correspondence table shown in FIG. 3A, it is estimated that the variation in the batch is caused by the difference in the thermal history in the batch heat treatment process, and the variation in the wafer is caused by the non-uniformity in the wafer surface in the single wafer processing process. Can be narrowed down. By narrowing down the cause process in advance by data analysis in this way, it is possible to reduce the cost and the number of analysis steps required for an experiment for determining the cause of the characteristic variation.

Next, as shown in FIG. 10, the characteristic change axis corresponding to each variation component between batches, within a batch, and within a lot, that is, each area on a wafer along a batch number, a shelf number, and a wafer number. An example of an apparatus for narrowing down the cause of the characteristic failure by visualizing the change in the characteristic in the above will be described. The configuration of the device is the same as that shown in FIG. Further, the device shown in FIG. 5 is activated, and FIG.
The processing up to the creation of the analysis data shown in FIG.

After creating the analysis data, the apparatus displays the sampling instruction screen shown in FIG. 9 on the display 504, and enters a standby state for input from the user. The user sees FIG.
The batch (one or more batches) to be analyzed is designated from the sampling instruction screen shown in FIG. After inputting the batch to be analyzed, the user presses the OK button to instruct the present apparatus to start the analysis process.

The present apparatus receives an instruction from the user to start the analysis processing, selects a record matching the batch number specified by the user on the sampling instruction screen of FIG. 9 from the analysis data shown in FIG. Based on the record, along with the characteristic change axis, that is, the batch number, the shelf number, and the wafer number, processing is performed to create a graph in which the measured values of the characteristics in each region on the wafer are plotted, and the processing result is displayed on the display 504. Is displayed and the processing ends.

As described above, by using the apparatus for narrowing down the cause of the characteristic defect, as shown in FIG. 10, the characteristic variation is classified into batch, batch, lot, and wafer variation components and visualized. As compared with the case where there is no clue, the process causing the characteristic variation can be narrowed down, and the cost and the number of analysis steps required for the experiment for investigating the cause of the characteristic variation can be reduced.

Further, if only one batch is specified in FIG. 9, the axis of the batch number, which was one of the three-dimensional components in FIG. 10, is eliminated. , Only the variation within the wafer) becomes apparent, and the analysis time can be reduced. For example, by displaying the three-dimensional graph shown in FIG. 10 and then specifying the batch number and switching to the two-dimensional graph, the analysis can be performed more efficiently. The same applies to the two-dimensional graph of the characteristic measured value by specifying the shelf number and the batch number.

These measurement results are displayed separately for each measurement area. It is empirically known that the cause of non-uniformity in the wafer surface is in the manufacturing equipment, and by monitoring in relation to the operating time and operating status of the equipment, the cause of the variation can be effectively revealed. It is intended to be made.

[0050]

According to the present invention, a completely new analysis data sampling method and a method of narrowing down the cause of variation in characteristics are established, and thereby the manufacturing yield can be improved.

[Brief description of the drawings]

FIG. 1 is a cross-sectional view illustrating an example of an LSI.

FIG. 2 is a diagram showing an example of an inspection item.

FIG. 3 is a diagram showing an example of a format of analysis data based on variation in characteristics of a semiconductor product.

FIG. 4 is a diagram showing an example of extracting characteristic failure factors.

FIG. 5 is a configuration diagram showing an example of a characteristic failure factor extraction device.

FIG. 6 is a diagram showing an example of a data format in which wafer inspection results are arranged in a table format.

FIG. 7 is a view showing an example of a data format in which probe inspection results are arranged in a table format.

FIG. 8 is a diagram showing an example of a data format in which batch numbers and shelf numbers are arranged in a table format.

FIG. 9 is a diagram showing an example of an input screen.

FIG. 10 is a diagram showing an example in which characteristics variations are visualized by being classified into respective variation components between batches, within a batch, within a lot, and within a wafer.

[Explanation of symbols]

501 ... central processing unit, 502 ... keyboard, 503 ...
Mouse, 504: display, 505: storage device, 5
06: wafer tester, 507: probe tester, 508
... characteristic failure factor extraction processing program, 509 ... network, 510 ... recording medium

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Shuji Kikuchi 292, Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside the Hitachi, Ltd. Production Technology Research Institute (72) Inventor Naofumi Iwata 292, Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Inside Hitachi, Ltd. Production Technology Laboratory

Claims (14)

[Claims] 1. A processing step of processing a wafer in a unit of a wafer or a batch, an element characteristic measuring step of measuring characteristics of an element formed in the processed wafer, and an electric characteristic of a chip formed in the wafer. A semiconductor device for managing a wafer processing step based on the calculated correlation, comprising: a yield calculation step of calculating a yield from the data; and a correlation calculation step of calculating a correlation between the element characteristics and the yield. A method comprising: identifying information indicating a measured position in a wafer, identification information regarding a position in a lot in which the measured wafer is inserted, and inserting the lot into a measurement result measured in the element characteristic measuring step. The identification information of the selected batch and the identification information of the position of the wafer or lot in the batch are added, and the measurement result is sampled using the identification information. The method of manufacturing an electronic device, characterized in that to calculate the correlation between device characteristics and yield based on the ring has been measured results. 2. The method according to claim 1, wherein the position information corresponds to a plurality of divided regions of the wafer, and the yield is calculated for each divided region. 3. A processing step of processing a wafer in units of wafers or batches, an element characteristic measuring step of measuring characteristics of elements formed on the processed wafer, and an electric characteristic of a chip formed on the wafer. A semiconductor device for managing a wafer processing step based on the calculated correlation, comprising: a yield calculation step of calculating a yield from the data; and a correlation calculation step of calculating a correlation between the element characteristics and the yield. A method, wherein identification information relating to a position in a lot into which the measured wafer is inserted, identification information of a batch into which the lot is inserted, or identification information of the wafer or the lot is added to the measurement result measured in the element characteristic measurement step. Adding at least one of identification information relating to a position in the batch, sampling the measurement result using the identification information, The method of manufacturing an electronic device, characterized in that to calculate the correlation between device characteristics and yield based on the ring has been measured results. 4. A processing step of processing a wafer in a unit of a wafer or a batch, an element characteristic measuring step of measuring characteristics of an element formed on the processed wafer, and an electric characteristic of a chip formed on the wafer. A semiconductor device for managing a wafer processing step based on the calculated correlation, comprising: a yield calculation step of calculating a yield from the data; and a correlation calculation step of calculating a correlation between the element characteristics and the yield. A method, wherein the identification information of the batch having the measured wafer is added to the measurement result measured in the device characteristic measurement step, and the corresponding measurement result is extracted based on the identification information of the designated batch. A method of manufacturing an electronic device, comprising calculating a correlation between element characteristics and a yield based on the extracted measurement result. 5. The measurement result further includes identification information relating to a position in the batch of the wafer or a lot into which the wafer is inserted, and the corresponding measurement is performed based on the identification information of the designated batch and the identification information relating to the position. 5. The method for manufacturing an electronic device according to claim 4, wherein a result is extracted, and a correlation between an element characteristic and a yield is calculated based on the extracted measurement result. 6. A method for identifying a specified batch by adding identification information indicating a measured position in a wafer and identification information regarding a position in a batch of a wafer or a lot into which the wafer is inserted to the measurement result. The corresponding measurement result is extracted based on the information and the identification information indicating the measured position in the wafer and the identification information regarding the position in the batch, and the correlation between the device characteristics and the yield is determined based on the extracted measurement result. The method for manufacturing an electronic device according to claim 4, wherein the calculation is performed. 7. A processing step for processing a wafer, and an element characteristic measuring step for measuring characteristics of elements formed on the processed wafer, and managing the wafer processing step based on the measurement result. A method of manufacturing a semiconductor device, comprising: adding position information in a measured wafer to the measurement result, displaying the measurement result by using the position information, displaying the measurement result in a graph, and displaying the measurement result on a wafer. A method for manufacturing an electronic device, wherein a graph is displayed according to a processing order. 8. The process according to claim 1, wherein the identification information of the processed batch is added to the measurement results of the wafers processed in batch units, and the measurement results are displayed in a graph in the order of the batches processed using the identification information of the batch. A method for manufacturing an electronic device according to claim 7. 9. The electronic device according to claim 8, wherein identification information of a lot in a batch is added to the measurement result, and the inside of the batch is graphically displayed in a lot order in the measurement result graphically displayed in the batch order. Manufacturing method. 10. The electronic device according to claim 9, wherein identification information of wafers in a lot is added to said measurement result, and said lot is displayed in a graph in a wafer order in said measurement result displayed in said lot order. Manufacturing method. 11. A method for adding identification information of a processed batch and identification information of a lot in a batch to a measurement result of the wafer processed in a batch unit, and performing the measurement using the identification information of the batch and the identification information of the lot. 8. The method for manufacturing an electronic device according to claim 7, wherein the results are displayed in a graph in the order of lots in the batch. 12. The electronic device according to claim 11, wherein identification information of wafers in a lot is added to said measurement result, and in said measurement result graphically displayed in said lot order, said inside of said lot is graphically displayed in wafer order. Manufacturing method. 13. A processing step of processing wafers in batches or lots, and an element characteristic measuring step of measuring characteristics of elements formed on the processed wafers. A method of manufacturing a semiconductor device for managing a wafer processing step, wherein identification information of a processed batch and identification information of a lot in a batch are added to the measurement result, and the identification information of the batch and the identification information of the lot are used. And displaying the measurement results in a graph in the order of lots in an arbitrary batch. 14. The electronic device according to claim 13, wherein identification information of wafers in a lot is added to said measurement result, and in said measurement result graphically displayed in said lot order, said lot is graphically displayed in wafer order. Manufacturing method.