JP2002366583A - Device and method for processing data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make processing courses comprehensive by observing them by human eyes and to easily determine whether the processing contents is correct or not in a short time when automatically processing multivariable data. SOLUTION: Data of operating conditions of manufacturing processes of a product and data of items to express characteristics of the product are time- sequentially acquired by using a measuring means, a prescribed processing procedures are performed to the acquired data by a control means based on processing conditions set by using an input means and the processing procedures are recorded in a storage means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data processing apparatus and a data processing method for recording the processing of data when processing and analyzing data such as items (characteristic items) representing the characteristics of a product. is there.

[0002]

2. Description of the Related Art A product is evaluated based on a number of items representing characteristics of the product during or after manufacturing, for example, characteristics such as temperature at the time of manufacturing the product, and weight and size of the product.
Much of this variation in characteristics is often due to some important work steps, even for products manufactured through multiple work steps. Therefore, it is important to find a process that causes the variation and pick up a processing factor having a high contribution rate to the variation from the process.

In such a case, the state of the product in each work process is acquired as time-series numerical data, and the multivariate data is analyzed. However, even if various processings are performed based on numerical data, it takes a lot of man-hours to derive important causal relationships from a huge amount of data. preferable. Therefore, conventionally, a data processing device that automatically processes such a huge amount of data and derives a causal relationship thereof has been used.

[0004] However, when many factors are intricately intertwined, erroneous conclusions are often obtained by automatic mechanical processing of data. For example, even if an attempt is made to cope with a neural network or the like, there is a case where the use has to be forgotten in terms of reliability unless the relationship between the factors is accurately grasped. In other words, when a new control is introduced into the manufacturing process, a factor analysis underlying the control needs to be finally tracked and confirmed by a human, even if the machine performs all the operations automatically.

However, in the conventional data processing apparatus, a predetermined processing is automatically performed on the data of the item representing the characteristic of the product, and only the processing result is recorded. Before, during and after processing, numerical data after processing, and visualized graph data were not recorded and stored, so it took a long time for humans to judge the correctness of the result of automatic processing. age,
As a result, there is a problem that there is a risk of erroneous judgment.

[0006]

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems of the prior art and to make it easy for human eyes to understand the progress of processing when automatically processing multivariate data. It is an object of the present invention to provide a data processing apparatus and a processing method that can easily determine correct / incorrect processing contents in a short time.

[0007]

In order to achieve the above object, the present invention provides a data collection unit for acquiring data of operating conditions in a manufacturing process of a product and data of items representing characteristics of the product in a time series. A data processing unit that performs predetermined processing on the data acquired by the data collection unit based on a processing condition set in advance, and a data storage unit that records a process performed by the data processing unit. And a data processing device characterized by the following.

Further, the present invention provides a method for obtaining, in a time series, data on operating conditions of a product manufacturing process and data on items representing characteristics of the product by a measuring means, and Setting a processing condition, and, based on the processing condition, performing predetermined processing on the acquired data by a control unit,
Recording the processing steps of the processing processing in a storage means.

Here, the processing is realized by a software program operating on a computer, and the control means uses the input means among the data of the items representing the characteristics registered in the spreadsheet. It is preferably applied to data included in a predetermined area to be processed set in advance.

In the processing, abnormal data is detected and deleted from data included in the area set using the input means, and statistical processing is performed using the data sheet after the deletion of the abnormal data. It is preferred that

The statistical processing is preferably any one of calculation of basic statistics, correlation analysis, and multivariate analysis.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a data processing apparatus and a data processing method according to the present invention will be described in detail based on preferred embodiments shown in the accompanying drawings.

FIG. 1 is a conceptual view showing the configuration of an embodiment of a data processing apparatus according to the present invention. Data processing device 10 of the present invention
According to the data processing method of the present invention, collect data such as operating conditions of the product manufacturing process and items representing the characteristics of the product, perform predetermined processing on the collected data,
As shown in FIG. 1, it records the processing steps, and includes a data collection unit 12, a data processing unit 14, and a data storage unit 16.

In the data processing apparatus 10 shown in the figure, first, the data collecting unit 12 uses the measuring means 18 such as various sensors and measuring devices to measure the data of the operating conditions in the product manufacturing process and the characteristics of the product. This is a part for acquiring data of an item to be represented (hereinafter, referred to as a characteristic item) in time series. There is no restriction on the time period for acquiring data in chronological order.
For example, data for a necessary period, such as one lot, one day, one week, one month, and one year, may be obtained.

Here, the operating conditions of the product manufacturing process are as follows.
For example, the operating conditions (operating conditions) of each part of the manufacturing apparatus in the manufacturing process of the product, such as the moving speed of a belt conveyor that conveys the product, the rotation speed of various motors, and the like. The product characteristic item is an item that represents the characteristic of the product, such as the temperature, weight, and length of the product. It goes without saying that both the operating conditions of the product manufacturing process and the characteristic items of the product may include various items other than the exemplified items.

Subsequently, the data processing unit 14 is acquired by the data collection unit 12 based on processing conditions set by the user of the data processing apparatus 10 using the input means 20 such as a keyboard and a mouse. The data is subjected to predetermined processing by the control means 22 such as a CPU (Central Processing Unit). The processing by the data processing unit 14 is performed by a display means 2 such as a CRT (CRT) or an LCD (Liquid Crystal Display).
4 is displayed.

Here, the processing includes, for example, deleting abnormal data (abnormal values, outliers, etc.), and performing various statistical processes. Statistical processing includes calculation of basic statistics (average, σ (standard deviation), R (correlation coefficient), min (minimum), max (maximum), histogram, etc.), correlation analysis, multivariate analysis Processing. It goes without saying that various items other than the exemplified items may be included in the processing and statistical processing.

The processing conditions described above include, for example, the setting of the division of the manufacturing process (for example, for each size, for each lot, for the first / second half of each lot, or for each work step), and the setting of a signal indicating the start of a lot. , Setting of a method of calculating abnormal data (for example, data in a range exceeding 3σ is regarded as abnormal data), setting of processing (for example, deletion, contents of statistical processing, and the like). Similarly, it goes without saying that the processing conditions may include various items other than the exemplified items.

Finally, the data storage section 16 is a section for recording the processing process by the data processing section 14 in a storage means 26 such as a hard disk. In other words, the data storage unit 16 processes the data by the data processing unit 14,
That is, the data before, during, and after the processing (processing result) may be in any predetermined format specified by the processing conditions, such as a file in a numerical data format, a table format, a graph format, an image format, a video format, or the like. Is saved as a file.

Here, the processing steps by the data processing unit 14 include the processing by the data processing unit 14,
For example, it includes processing steps for deleting abnormal data and performing various statistical processes. It goes without saying that the processing process may include various items other than the illustrated items. The files to be stored are recorded, for example, by day, size, work step, and analysis order, and can be used (referred) at any time after processing.

Hereinafter, the data processing method of the present invention will be described together with the operation of the data processing apparatus 10 of the present invention.

In the data processing apparatus 10 of the present invention, according to the data processing method of the present invention, first, in the data collecting section 12, the measuring means 18 converts the data of the operating conditions of the product manufacturing process and the data of the characteristic items of the product in time series. Will be collected.
Further, the user of the data processing apparatus 10 sets a processing condition for the data acquired by the data collection unit 12 using the input unit 20. Note that the processing condition may be set at any time before the processing by the data processing unit 14 is performed.

Subsequently, in the data processing unit 14, the control unit 22 performs a predetermined processing such as deletion of abnormal data, various statistical processes, etc. on the data acquired by the data collection unit 12 based on the processing conditions described above. Is performed, and the process is displayed on the display means 24. Further, the data storage unit 16 records the process of the data processing unit 14, ie, the data before, during, and after the process, in the storage unit 26 as a file of a predetermined format such as a table, a graph, an image, and a video. Is done.

After the data collection by the data collection unit 12 is completely completed, the data processing by the data processing unit 14 may be started, or the data processing may be performed sequentially while collecting the data. Good. After the data processing by the data processing unit 14 is completely completed, the recording of the processing by the data storage unit 16 may be started, or the processing may be sequentially recorded while processing the data. Good.

The operation of the data processing apparatus 10 can be performed interactively or automatically by batch processing. Here, the direct data is collected in chronological order using the measuring means 18 and the direct processing conditions are set using the input means 20, but from a file in which the data collected in advance using the measuring means 18 is recorded. The data may be read, or the processing conditions may be read from a file in which the processing conditions are set in advance using the input unit 20.

In the following, an example in which the data processing unit 14 is realized by a software program operating on a computer will be described with reference to the display screens of FIGS.
This will be described more specifically. In the following example, as processing, after removing abnormal data from data included in an area set by the user of the data processing apparatus 10 for data registered in a spreadsheet, Assume that correlation analysis is performed as an example.

The software for realizing the data processing unit 14 automatically processes data in a predetermined area registered in a spreadsheet in accordance with processing conditions set by a user of the data processing apparatus 10, A table, a graph, or the like corresponding to the data registered in the sheet is displayed.
2. It operates on a computer provided with a display means 24, a storage means 26 and the like.

Here, the spreadsheet is also called spreadsheet software. In addition to a function of registering data such as numbers, characters, and calculation formulas in a table format to create a database, the spreadsheet performs various processing on the registered data. And a function of creating and displaying various graphs based on the registered data. If this spreadsheet is used, the registered data can be displayed as a graph and the data can be visually confirmed, so that it is generally used in the field of data analysis, for example.

In the software for realizing the data processing unit 14 of the present invention (hereinafter simply referred to as software or software according to the present invention), in the present embodiment, first, the software is set by a user of the software and registered in a spreadsheet. A graph including an arbitrary characteristic item selected from the plurality of characteristic items is automatically displayed on the screen of the display means 24 of the computer based on (linked to) the data of the selected characteristic item. Is displayed.

Here, the characteristic items and graphs can be freely set by the user of the data processing apparatus 10. For example, as the characteristic item, one or more arbitrary characteristic items can be selected at a time, and a plurality of graphs of any type can be selected.
It is preferable that two or more images can be displayed simultaneously. In addition, it is preferable that the display size of the characteristic item and the graph can be arbitrarily set, and when two or more graphs are displayed at the same time, the display order can be specified.

It should be noted that the selection of the characteristic item is not an essential matter. For example, when the characteristic item is not selected, it can be set so that all the characteristic items registered in the spreadsheet are selected. In addition, when the setting of the number of graphs to be displayed, the display size thereof, and other setting items other than these are omitted by the user of the data processing apparatus 10, the default values at the time of the omission are used. It is preferable that the setting is made such that:

FIG. 2 is a schematic diagram of one embodiment of a display screen according to the present invention, and based on data registered in a spreadsheet, a first graph including two characteristic items TREADL and CUTL. At the top of the screen
TMT, ATMCOOLT, TREADT, COOL
A second graph including six characteristic items of T, CTEMP, and STV is displayed at the bottom of the screen. That is, at least the spreadsheet shown in FIG.
Data for one characteristic field has been created.

Here, the characteristic items of TREADL, CUTL, ATMT,... Shown in the graph shown in FIG. This is data that automatically collects the status in a time series. The horizontal axes of the first and second graphs shown in FIG. (Number). The vertical axes of the first and second graphs represent the values of TREADL, CUTL, ATMT,... Data registered in the spreadsheet.

The number of characteristic items registered in the spreadsheet and the number of data included in each characteristic item are not limited. Therefore, the spreadsheet may include a plurality of other characteristic items other than those described above, which are not set (selected) by the user of the data processing apparatus 10.

The data registered in the spreadsheet may be either a real number or a standardized numerical value.
It is preferable that both data can be freely selected. As the standardized numerical value, for example, 3σ / average value, (measured value−average value) / σ (σ is standard deviation) and the like can be exemplified, but data according to other standardization rules can be freely used. It is preferable to be able to set to.

After the graph including the designated characteristic item is displayed on the screen of the display means 24, as shown in FIG. 3, an arbitrary area to be processed set by the user of the software is displayed on the display means 24. 24 is drawn on the graph displayed on the graph 24. Further, the control means 22 performs predetermined processing similarly set by the user of the software on the data included in the set area among the data of the designated characteristic items.

It is preferable that a plurality of areas can be specified at the same time. Also, the processing is not performed directly on the original data registered in the spreadsheet, but is performed on the original data (data before processing).
In order to facilitate comparison with the processed data, create a work spreadsheet that duplicates the contents of the data registered in the original data. Processing is performed.

In the above-described processing, first, when abnormal data is deleted, as shown in FIG. 3, for each characteristic item, an area set by the user of the software, that is, a region around the abnormal data to be deleted. (First area) is drawn on the graph. Further, out of the data registered in the work spreadsheet, abnormal data included in the designated area is detected (calculated), and the abnormal data is deleted from the work spreadsheet.

More specifically, the detection of abnormal data is based on processing conditions set by the user of the software, for example, processing conditions such as a range exceeding 3σ, and an area near the corresponding abnormal data is displayed on the screen. The data corresponding to the above-described processing conditions is calculated from the data included in the above-described area among the data registered in the work data sheet while being drawn on the graph, and is calculated from the work spreadsheet. Abnormal data is deleted.

After the completion of the deletion process, the data registered in the work spreadsheet is updated based on the calculation result. As shown in FIG. 4, the processed graph is displayed again on the display means 24 in accordance with the update of the data. In addition,
The graph after the processing is recorded (saved) in a predetermined format in the storage unit 26 as a separate file together with the graphs before and during the processing shown in FIGS. 2 and 3 so that these graphs are referred to after the processing. It is possible to

According to the processing conditions, the operation up to this point is performed in units of one lot (second area). When the above-described deletion processing is completed for the data of one lot, as shown in FIG. As processing, an average value, σ (standard deviation), and n number (parameter) are calculated for each characteristic item, and displayed on the screen in a table format. Further, the above-described operation is performed for each lot, and FIG.
As shown in (1), an average value for each characteristic item is calculated for all lots. These tables are also recorded in the storage means 26 as a separate file in a predetermined format.

Subsequently, when performing the correlation analysis, the data registered in the work spreadsheet is converted into a data sheet of a predetermined format and stored in the storage means 26 as a temporary file.

By storing the data sheet in an existing general-purpose format, statistical processing such as correlation analysis can be performed using commercially available statistical processing software. Thereby, versatility of data can be improved.

Subsequently, the correlation analysis is performed using the data included in the data sheet stored as the above-mentioned temporary file, that is, using all the data registered in the data sheet as one area. Although a detailed description of the analysis result is omitted, it is displayed on the screen in a table format, for example, as shown in FIG. 7, for example. Similarly, the analysis result is recorded in the storage means 26 as another file in a predetermined format.

In the above embodiment, before processing, during processing,
The processed data is recorded and saved in the form of numerical data registered in the spreadsheet, as well as in the form of tables and graphs, but is not limited to this. For example, image formats,
Tables, graphs, images, and the like displayed on the screen during the processing can be recorded and stored in a digital video format.
Further, the data to be recorded and stored and the format thereof may be set by the user. For example, only a part of the processing contents may be recorded, or all of the processing contents may be recorded one by one.

As described above, the user of the data processing apparatus 10 may, after completion of the processing, determine the contents of the processing, that is, the data before, during and after the processing, for example, in a table format, a graph format, an image format. Alternatively, it is possible to reproduce and display an image displayed on the screen during processing in a video format. Therefore, the user of the data processing device 10 can accurately and visually confirm the processing process while viewing the contents of the processing process.

In the above embodiment, the processing is
Although the correlation analysis is performed after the abnormal data is deleted, the present invention is not limited to this, and the deletion and various kinds of statistical processing may be repeatedly performed. Further, the case where the data processing unit 14 is realized by a software program is exemplified, but the present invention is not limited thereto, and the data processing unit 14 may be realized by using any conventionally known means. The same applies to the realization means of the data collection unit 12 and the data storage unit 16.

In the above embodiment, the data processing unit 14
Is displayed on the display means 24, but the present invention is not limited to this. For example, a screen during data processing by the data processing unit 14 may not be displayed on the display means 24. Further, in the above embodiment, the area (second area) where the correlation analysis is performed is the data of all lots. However, the present invention is not limited to this, and the user of the data processing apparatus 10 may perform the correlation analysis using the input unit. The region to be performed may be set as the processing condition.

The data processing apparatus and processing method of the present invention are basically as described above. As described above, the data processing device and the processing method of the present invention have been described in detail. However, the present invention is not limited to the above embodiments, and various improvements and modifications may be made without departing from the gist of the present invention. Of course.

[0050]

As described above in detail, according to the present invention, the data of the operating conditions of the product manufacturing process and the data of the items representing the characteristics of the product are acquired in time series using the measuring means.
Based on the processing conditions set by using the input means, the control means performs predetermined processing on the acquired data, and records the processing in the storage means. As a result, according to the present invention, when analyzing multivariate data, such as operation analysis of a manufacturing process, automatic processing is performed. Thus, the process can be easily verified while visually confirming it with a graph or an image file, so that the accuracy of analysis and the efficiency of analysis can be greatly improved.

[Brief description of the drawings]

FIG. 1 is a conceptual diagram illustrating a configuration of a data processing apparatus according to an embodiment of the present invention.

FIG. 2 is a schematic view of an embodiment of a display screen according to the present invention.

FIG. 3 is a schematic view of an embodiment of a display screen according to the present invention.

FIG. 4 is a schematic view of an embodiment of a display screen according to the present invention.

FIG. 5 is a schematic diagram of an embodiment of a display screen according to the present invention.

FIG. 6 is a schematic diagram of one embodiment of a display screen according to the present invention.

FIG. 7 is a schematic diagram of one embodiment of a display screen according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Data processing apparatus 12 Data collection part 14 Data processing part 16 Data storage part 18 Measurement means 20 Input means 22 Control means 24 Display means 26 Storage means

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G05B 23/02 301 G05B 23/02 301V 302 302T G06F 17/60 106 G06F 17/60 106 (72) Inventor Kazuo Miyasaka 2-1 Oiwake, Hiratsuka-shi, Kanagawa Prefecture Inside the Hiratsuka Factory of Yokohama Rubber Co., Ltd. (72) Inventor Masaru Kida 2-1 Oiwake, Hiratsuka-shi, Kanagawa Prefecture Inside the Hiratsuka Factory of Yokohama Rubber Co., Ltd. No.2-1 Oiwake, Hiratsuka-shi, Kanagawa Prefecture Inside the Hiratsuka Factory of Yokohama Rubber Co., Ltd. (72) Inventor Masamichi Koyama No.2-1 Oiwake, Hiratsuka-shi, Kanagawa Prefecture F-term in Hiratsuka Factory of Yokohama Rubber Co., Ltd. 2F070 AA01 CC06 CC11 DD01 DD14 FF04 FF12 2F076 BA01 BE04 BE05 BE08 BE09 BE13 5H223 EE06 EE15 FF06

Claims (5)

[Claims] 1. A data collection unit for acquiring data of operating conditions of a manufacturing process of a product and data of an item representing characteristics of the product in a time series, and the data collection unit based on a preset processing condition. A data processing device, comprising: a data processing unit that performs a predetermined processing process on acquired data; and a data storage unit that records a process performed by the data processing unit. 2. A step of obtaining data of operating conditions of a manufacturing process of a product and data of items representing characteristics of the product in time series by a measuring means, and setting processing conditions for the data by using an input means. Performing a predetermined processing on the acquired data by the control unit based on the processing condition; and recording a processing step of the processing in a storage unit. Characteristic data processing method. 3. The processing is realized by a software program operating on a computer, and the control means uses the input means of data of items representing the characteristics registered in a spreadsheet. 3. The data processing method according to claim 2, wherein the method is performed on data included in a predetermined area to be set. 4. The processing process detects and deletes abnormal data from data included in an area set using the input means, and performs statistical processing using the data sheet after the deletion of the abnormal data. 4. The data processing method according to claim 3, wherein: 5. The statistical processing is any one of calculation of basic statistics, correlation analysis, and multivariate analysis.
The data processing method described in 1.