CN115668080A - Quality influence factor determination support device and method - Google Patents
Quality influence factor determination support device and method Download PDFInfo
- Publication number
- CN115668080A CN115668080A CN202180037000.7A CN202180037000A CN115668080A CN 115668080 A CN115668080 A CN 115668080A CN 202180037000 A CN202180037000 A CN 202180037000A CN 115668080 A CN115668080 A CN 115668080A
- Authority
- CN
- China
- Prior art keywords
- quality
- affecting
- factor
- process data
- candidate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 178
- 238000000605 extraction Methods 0.000 claims abstract description 54
- 238000010923 batch production Methods 0.000 claims abstract description 27
- 238000013480 data collection Methods 0.000 claims abstract description 7
- 238000010586 diagram Methods 0.000 claims description 28
- 238000009434 installation Methods 0.000 claims description 16
- 238000004519 manufacturing process Methods 0.000 claims description 14
- 230000007717 exclusion Effects 0.000 claims description 2
- 238000003860 storage Methods 0.000 abstract description 13
- 239000000284 extract Substances 0.000 abstract description 2
- 230000001364 causal effect Effects 0.000 description 4
- 238000007405 data analysis Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000010924 continuous production Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000000491 multivariate analysis Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000011410 subtraction method Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Automation & Control Theory (AREA)
- General Factory Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The quality influence factor determination support device includes: a data collection unit (1) for collecting product quality data and process data of a batch process; a storage unit (2) for storing product quality data and process data; a feature value extraction unit (3) that extracts a feature value of the process data; a feature amount classification unit (4) that groups feature amounts based on the similarity between the feature amounts; a representative feature value selection unit (5) that selects a representative feature value for each group based on the relationship with the product quality data for each group; a candidate specifying unit (6) that specifies, as a quality-affecting-factor candidate, a representative feature amount that contributes to a variation in product quality data from among the representative feature amounts; and a determination result notification unit (7) that notifies the user of the determination result.
Description
Technical Field
The present invention relates to a batch process (batch process) oriented data analysis technology, and more particularly, to a quality-influencing-factor determination support apparatus and method for supporting determination of quality influencing factors that influence product quality.
Background
Process production can be broadly divided into continuous processes and batch processes. The continuous process is generally a way of continuously producing the same species. On the other hand, the batch process is a method of repeating the input and processing of the raw material and the delivery of the product for each item, and a plurality of items are produced by the same facility, and the actual results thereof are managed in units of lots (lots). In recent years, batch processes suitable for variant variable production have been of increasing interest.
However, various characteristics (maximum, minimum, slope, etc.) of the process data from the start of the batch to the end point in time can have an impact on the product quality of the batch process. Fig. 11 is a diagram schematically showing the influence of process data such as the temperature, pressure, and liquid level of a raw material on the product quality of each batch in the batch process. In this way, the product quality of the batch process is influenced by the characteristics of the process data, but in order to identify the factors that affect the product quality, a high level of knowledge and a large number of man-hours are required, and even a skilled engineer is fatigued to extract the factors that affect the product quality and model the correlation between the factors and the product quality.
Conventionally, there has been proposed a method of: a quality estimation model is constructed based on a group of factors (quality-affecting factors) that affect the quality/characteristics of a product, thereby providing measures for improving the operation of a process. For example, patent document 1 discloses a technique of: a quality evaluation function is constructed based on a multivariate analysis technique from the manipulated variable data and the state variable data of the target process, and the values of the manipulated variables for quality improvement are optimized. In the method disclosed in patent document 1, in selecting the quality-affecting factor, a variable representing the quality itself is used as an explained variable (target variable), and a combination of the explained variables capable of expressing the target variable in a form with a minimum error by a regression model is used as the quality-affecting factor.
However, the quality-affecting factor determined by the method disclosed in patent document 1 is not necessarily a cause of quality variation of the product. In particular, when there is a high correlation (multiple collinearity) between candidates of the quality-affecting factor (an explanatory variable of the regression model), the quality-affecting factor cannot be determined appropriately. For example, a variable addition or subtraction method (step wise) as a method of making a regression model is pointed out that numerical calculation becomes unstable. Also, in Least Absolute value convergence and Selection Operator (LASSO) regression, which is another method, only one of influence factors having high correlation with each other can be selected.
As is clear from the above description, the quality-affecting factor specified by the method disclosed in patent document 1 is a quality-affecting factor automatically selected by a specific algorithm from among a group of mutually highly correlated quality-affecting factors, and is a quadratic factor propagated from a true cause, and therefore, the reason why the quality is affected is not well explained. In addition, in the method disclosed in patent document 1, there is a possibility that a variable that cannot be directly controlled is selected as a quality-affecting factor, and thus it may not be used for operational improvement of the process.
Documents of the prior art
Patent document
Patent document 1: japanese patent laid-open No. 2006-323523
Disclosure of Invention
Problems to be solved by the invention
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a quality influence factor determination support apparatus and method for supporting the determination of appropriate feature quantities with high explanatory performance of the reason for the influence on the quality and high control realizability from a feature quantity group that may influence the product quality.
Means for solving the problems
The quality influence factor determination support device of the present invention is characterized by comprising: a data collection unit configured to collect product quality data of a batch process and process data of the batch process; a feature value extraction unit configured to extract a feature value of the process data; a feature amount classification unit configured to group all the feature amounts extracted by the feature amount extraction unit based on a similarity between the feature amounts; a representative feature amount selection unit configured to select a representative feature amount for each group generated by the feature amount classification unit based on a relationship with the product quality data; a candidate specification unit configured to specify, as a quality-affecting-factor candidate, a representative feature amount contributing to a variation in the product quality data from the representative feature amounts selected by the representative feature amount selection unit; and a determination result notification unit configured to notify a user of the determination result of the quality-affecting-factor candidate.
In the quality-affecting-factor determination support apparatus according to the aspect of the invention, the determination result notification unit displays a piping installation diagram of a production facility of a product, and in the piping installation diagram, the name of the quality-affecting-factor candidate and the name of the process data of the extraction source of the quality-affecting-factor candidate are displayed in a position where the process data is measured while being superimposed on each other, and the position where the product quality data is measured is displayed in the piping installation diagram.
In addition, in the quality-affecting-factor determination support apparatus according to the present invention, the determination result notification unit displays a name of the quality-affecting-factor candidate, a name of the process data that is an extraction source of the quality-affecting-factor candidate, and a type of the process data.
In the quality-related-factor determination support device according to the present invention, the determination result notification unit displays, when instructed by a user, names of other feature amounts in a group including the quality-related-factor candidates.
In the quality-influencing-factor determination support device according to the present invention, the determination result notification unit displays a first graph of process data as an extraction source of the quality influencing factor candidate, displays time information at which the quality influencing factor candidate is extracted on the first graph, displays a second graph of process data as an extraction source of another feature amount included in the group including the quality influencing factor candidate, and displays time information at which the feature amount is extracted on the second graph.
In the quality-affecting-factor determination support device according to the present invention, the determination result notification unit displays a first graph of process data as an extraction source of the quality affecting factor candidate, displays time information at which the quality affecting factor candidate is extracted on the first graph, displays a second graph of process data as an extraction source of another feature amount included in a group including the quality affecting factor candidate, displays time information at which the feature amount is extracted on the second graph, and displays the first graph, the second graph, and the time information for each batch of the batch process.
Further, an embodiment of the quality-affecting-factor determination support apparatus according to the present invention is characterized by further comprising: the quality estimation model generation unit is configured to generate a first quality estimation model for modeling a relationship between the quality-affecting factor candidate and the product quality, and a second quality estimation model for modeling a relationship between another feature amount included in the group including the quality-affecting factor candidate and the product quality, and the determination result notification unit displays a first graph representing the relationship between the quality-affecting factor candidate and the product quality based on the first quality estimation model, and displays a second graph representing the relationship between another feature amount included in the group including the quality-affecting factor candidate and the product quality based on the second quality estimation model.
Further, an example of the structure of the quality-affecting-factor determination support apparatus according to the present invention further includes: a process model generation unit configured to generate a process model with respect to process data of an extraction source as the quality-affecting-factor candidate and process data of an extraction source as another feature amount included in a group including the quality-affecting-factor candidate; and a process data estimation display unit configured to display a third graph of the process data as an extraction source of the quality-affecting-factor candidate when a change amount of the quality-affecting-factor candidate is designated by a user, display time information of the extracted quality-affecting-factor candidate on the third graph, display an estimated time series of the process data when the quality-affecting-factor candidate is changed by the change amount designated by the user on the third graph in an overlapping manner based on the process model, display a fourth graph of the process data as an extraction source of the feature amount when a change amount of another feature amount included in a group including the quality-affecting-factor candidate is designated by the user, display time information of the extracted feature amount on the fourth graph, and display an estimated time series of the process data when the feature amount is changed by the change amount designated by the user on the fourth graph in an overlapping manner based on the process model.
Further, an example of the structure of the quality-affecting-factor determination support apparatus according to the present invention further includes: the determination result changing unit is configured to, when a replacement instruction of a quality-affecting-factor candidate is issued from a user, replace the quality-affecting-factor candidate with another feature amount selected by the user in a group including the quality-affecting-factor candidate, and, when an exclusion instruction of a group is issued from the user, exclude the designated group from the determination result of the quality-affecting-factor candidate.
Further, the quality-influencing-factor determination support method of the present invention is characterized by comprising: the method comprises the following steps of firstly, collecting product quality data of a batch process and process data of the batch process; a second step of extracting a feature amount of the process data; a third step of grouping all the feature quantities extracted in the second step based on the similarity of the feature quantities to each other; a fourth step of selecting, for each group generated in the third step, a representative feature amount for each group based on a relationship with the product quality data; a fifth step of determining, as quality-affecting-factor candidates, representative feature quantities that contribute to variation in the product quality data from among the representative feature quantities selected in the fourth step; and a sixth step of notifying a user of the determination result of the quality-affecting factor candidate.
In addition, according to a configuration example of the quality-related factor determination support method according to the present invention, the sixth step includes a step of notifying, simultaneously with the determination result of the quality-related factor candidate or in accordance with an instruction from a user, another feature amount in the group including the quality-related factor candidate.
ADVANTAGEOUS EFFECTS OF INVENTION
According to the present invention, in order to identify a proper feature having high explanatory properties and high realizability as a quality factor from a feature group having a possibility of affecting the product quality, the user is notified of a quality factor candidate, whereby the user can be supported in analysis. As a result, the user can specify the quality-affecting factor with high adequacy, and the relationship between the quality-affecting factor and the product quality can be easily modeled.
In addition, in the present invention, the piping installation diagram of the production facility of the product is displayed, and the name of the quality-affecting factor candidate and the name of the process data of the extraction source as the quality-affecting factor candidate are displayed in a position where the process data is measured in the piping installation diagram in a superimposed manner, and the position where the product quality data is measured is displayed in the piping installation diagram, whereby the user can exclude the feature quantity which is not physically related to the product quality data or the feature quantity whose causal relationship is reversed from the quality-affecting factor candidate based on the process knowledge.
In addition, in the present invention, the name of the quality-affecting-factor candidate, the name of the process data as the extraction source of the quality-affecting-factor candidate, and the type of the process data are displayed, whereby the user can select the quality-affecting-factor candidate extracted from the controllable process data.
In the present invention, the first chart of the process data of the extraction source as the quality-affecting-factor candidate is displayed, the time information of the extracted quality-affecting-factor candidate is displayed on the first chart, the second chart of the process data of the extraction source including the other feature amounts included in the group including the quality-affecting-factor candidate is displayed, and the time information of the extracted feature amount is displayed on the second chart, whereby the user can grasp the upstream-side factor which is the true factor for the quality variation, and the explanatory performance can be improved.
In addition, in the present invention, the first chart and the second chart and the time information are displayed for each batch of the batch process, so that the user can exclude the quality-affecting-factor candidates that are accidentally selected. Further, even if the user replaces the quality-affecting factor candidates with similar other feature amounts, it is possible to predict whether or not the product quality will not deteriorate.
In the present invention, by displaying the first graph showing the relationship between the quality-affecting factor candidates and the product quality and displaying the second graph showing the relationship between the other feature amounts included in the group including the quality-affecting factor candidates and the product quality, the operation for improving the product quality can be simulated to evaluate the realizability.
In the present invention, the estimated time series of the process data when the quality-affecting-factor candidates are changed by the change amount designated by the user is displayed in a superimposed manner on the third graph, and the estimated time series of the process data when the other feature amounts included in the group including the quality-affecting-factor candidates are changed by the change amount designated by the user is displayed in a superimposed manner on the fourth graph, whereby the user can recognize the trajectory (time series graph) necessary for improving the product quality with respect to the process data which becomes the source of the quality-affecting-factor candidates or feature amounts, and can determine the feasibility or risk with respect to the operation of the target process necessary for improving the product quality.
In the present invention, by providing the determination result changing unit, the user can replace the quality-affecting-factor candidates determined to be inappropriate with other feature amounts, or exclude the groups determined to be inappropriate from the determination result.
Drawings
Fig. 1 is a block diagram showing the configuration of a quality-affecting-factor determination support apparatus according to a first embodiment of the present invention.
Fig. 2 is a flowchart illustrating the operation of the quality-affecting-factor determination support apparatus according to the first embodiment of the present invention.
Fig. 3 is a diagram showing an example of a screen displayed by the determination result notification unit of the quality-affecting-factor determination support apparatus according to the second embodiment of the present invention.
Fig. 4 is a diagram showing an example of a screen displayed by the determination result notification unit of the quality-affecting-factor determination support apparatus according to the third embodiment of the present invention.
Fig. 5 is a diagram showing an example of a screen displayed by the determination result notification unit of the quality-affecting-factor determination support apparatus according to the fourth embodiment of the present invention.
Fig. 6 is a diagram showing an example of a screen displayed by the determination result notification unit of the quality-affecting-factor determination supporting apparatus according to the fifth embodiment of the present invention.
Fig. 7 is a flowchart illustrating the operation of the quality-affecting-factor determination support apparatus according to the sixth embodiment of the present invention.
Fig. 8 is a diagram showing an example of a screen displayed by the determination result notification unit of the quality-affecting-factor determination support apparatus according to the sixth embodiment of the present invention.
Fig. 9 is a diagram showing an example of a display of a process data estimation time series chart according to a sixth embodiment of the present invention.
Fig. 10 is a block diagram showing a configuration example of a computer that realizes the quality-affecting-factor determination support apparatus according to the first to sixth embodiments of the present invention.
Fig. 11 is a diagram schematically showing the influence of process data on the quality of a product of each lot during a lot process.
Detailed Description
[ Focus of the invention ]
In the case of producing a product of a device for specifying a quality-affecting factor, the device is operated under the responsibility of a manufacturing person in charge of a user company, and therefore, if the quality-affecting factor specified by the device is not approved by the manufacturing person in charge, the device cannot be effectively used in actual operation. Therefore, how the intention of the person in charge of manufacturing (the explanatory effect of the reason why the quality-affecting factor affects the product quality, the realizability of the control of the quality-affecting factor, and the like) can be reflected in the step of selecting the quality-affecting factor becomes a key point.
Generally, the adequacy of the selection of the quality-affecting factors cannot be judged by only data analysis, and therefore, the user's process insight (or logic for regularizing the process insight) must be involved. Thus, the inventors thought: if the quality-affecting-factor candidates are presented by the automatic data analysis, the user can determine the appropriate quality-affecting factors considering the explanatory property or the control realizability by effectively using his/her own process knowledge.
[ first embodiment ]
Hereinafter, embodiments of the present invention will be described with reference to the drawings. Fig. 1 is a block diagram showing a configuration of a quality-affecting-factor determination supporting apparatus according to a first embodiment of the present invention. The quality influence factor determination support device includes: a data collection unit 1 for collecting product quality data of a batch process and process data of the batch process; a storage part 2 for storing product quality data and process data; a feature value extraction unit 3 for extracting a feature value of the process data; a feature amount classification unit 4 that groups all the feature amounts extracted by the feature amount extraction unit 3 based on the similarity between the feature amounts; a representative feature amount selection unit 5 that selects a representative feature amount for each group generated by the feature amount classification unit 4 based on a relationship with the product quality data; a candidate specifying unit 6 that specifies, as a quality-affecting-factor candidate, a representative feature amount that contributes to a variation in product quality data from the representative feature amounts selected by the representative feature amount selecting unit 5; and a determination result notification unit 7 for notifying the user of the determination result of the quality-affecting-factor candidate.
Further, the quality influence factor determination support device includes: a determination result changer 8 for, when there is an instruction from the user to replace the quality factor candidate, replacing the quality factor candidate with another feature selected by the user in a group including the quality factor candidate, and, when there is an instruction to exclude the user from the determination result of the quality factor candidate, excluding the indicated group; a determination result output unit 9 for outputting a final determination result of the quality-affecting factor candidates; a quality estimation model generation unit 10 that generates a quality estimation model for modeling the relationship between the quality-affecting factor candidates and the product quality, and a quality estimation model for modeling the relationship between the other feature values included in the group including the quality-affecting factor candidates and the product quality; a process model generation unit 11 that generates a process model for process data of an extraction source that is a quality-affecting factor candidate and process data of an extraction source that is another feature amount included in a group including the quality-affecting factor candidate; and a process data estimation display unit 12 that displays, based on the process model, an estimated time series of the process data when the quality-affecting factor candidates are changed by the change amount specified by the user, or an estimated time series of the process data when other feature amounts included in the group including the quality-affecting factor candidates are changed by the change amount specified by the user.
Fig. 2 is a flowchart illustrating the operation of the quality-affecting-factor determination support apparatus according to the present embodiment.
The data collection unit 1 collects product quality data (for example, quality data measured by a measuring instrument or quality data input by a person in charge at a manufacturing site) and process data (for example, temperature, pressure, flow rate, liquid level height, elapsed time, and the like) of a batch process from a management device or the like of a production facility of a target product (step S100 in fig. 2). The collected process data and quality data are accumulated in the storage section 2.
Next, the feature extraction unit 3 extracts features (for example, an average value, a maximum value, a minimum value, a slope, a variance, and the like) for each segment of the process data for a plurality of types of process data of the lot corresponding to the product quality data of the lot process (step S101 in fig. 2). Here, the term "segment" refers to a range in the time direction in which the change point is separated from the change point in the time-series data of the process data.
The feature amount extraction unit 3 may automatically separate time-series data of the process data into segment units, or may separate the time-series data based on information acquired from the outside. Segmentation methods for automatically separating time series data are described, for example, in the documents "e.keogh, s.chu, d.hart, m.pazzani," time series segmentation: a survey and new method, "data mining in time series databases, published in 2004". As a method of partitioning based on information acquired from the outside, for example, the following methods are considered: information on the progress of the batch process control ("in-temperature", etc.) is acquired from the outside to separate the time-series data.
The feature amount classification unit 4 groups all the feature amounts extracted by the feature amount extraction unit 3 based on the similarity between the feature amounts (step S102 in fig. 2). Examples of the grouping method include hierarchical clustering (hierarchical clustering) and spectral clustering (spectral clustering).
Next, the representative feature amount selection unit 5 selects a representative feature amount for each group generated by the feature amount classification unit 4 based on the relationship with the product quality data (step S103 in fig. 2).
Each group includes various feature quantities, each having four attributes of a group, a batch of a batch process, a segment (for example, a transient state period, a steady state period, and the like), and process data as an extraction source. The representative feature selection unit 5 sets the feature data string x = { x } for the same type of fragment and the same type of process data as the extraction source and different feature data in batches 1 、x 2 Data string y = (y) with product quality data of each lot 1 、y 2 8230and calculating the similarity. Then, the representative feature amount selection unit 5 selects, as the representative feature amount, the feature amount having the greatest similarity to the product quality data among the feature amounts included in each group. The similarity of the data strings may utilize known correlation coefficients. A method of calculating the correlation coefficient (specimen correlation coefficient) is disclosed in, for example, the literature "shizaki science, chinese and western widi, shikakoff," the term "kukoku" in practical statistics, and the company Ohm, 2004. In addition, the similarity here is not limited toTo the correlation coefficient.
The candidate specification unit 6 specifies, as the quality-affecting-factor candidates, representative feature amounts contributing to the variation of the product quality data from the representative feature amounts selected by the representative feature amount selection unit 5 (step S104 in fig. 2). The candidate identification unit 6 may identify the quality-affecting-factor candidates by a stepwise method, LASSO regression, or the like, using the product quality data as an interpreted variable (target variable) and the representative feature quantity as an interpreted variable.
Next, the determination result notification unit 7 notifies the user of the determination result of the quality-affecting factor candidate together with other feature amounts in the group including the quality-affecting factor candidate (step S105 in fig. 2). However, as described later, the quality-affecting factor candidates and the other feature amounts in the group including the quality-affecting factor candidates may not be notified at the same time. The notification method to the user will be described later.
The user who knows the determination result of the quality factor candidate determines whether the determination of the quality factor candidate is appropriate or not, considering whether the quality factor candidate is likely to have a causal relationship (explanatory) with respect to the product quality in principle and whether the quality factor candidate is a controllable feature quantity (control realizability). When the user determines that the quality-affecting-factor candidate is not properly determined, the user selects a feature that can be determined to be suitable as a quality affecting factor from the other feature in the group including the quality-affecting-factor candidate that has been notified, and instructs the quality-affecting-factor determination support apparatus to replace the selected feature with the quality-affecting-factor candidate.
When the user instructs replacement of the quality-affecting-factor candidate (yes in step S107 in fig. 2), the determination result changing unit 8 changes the instructed quality-affecting-factor candidate to the feature amount selected by the user in the group including the quality-affecting-factor candidate (step S108 in fig. 2).
Then, when all the feature quantities included in the quality-affecting-factor candidates and the group including the quality-affecting-factor candidates are judged to be inappropriate, the user instructs the quality-affecting-factor determination support device to exclude the group including the quality-affecting-factor candidates from the determination result.
When the user has an instruction to exclude a group (yes in step S109 in fig. 2), the determination result changing unit 8 excludes the indicated group from the determination results of the quality-affecting-factor candidates (step S110 in fig. 2).
The user instructs the quality-affecting-factor determination support apparatus to end the determination process when the determination result of the quality-affecting-factor candidate first notified from the quality-affecting-factor determination support apparatus is judged to be appropriate, or when the determination of the quality-affecting-factor candidate finally judged to be appropriate is excluded by replacing the quality-affecting-factor candidate.
When the user instructs to end the determination process (yes in step S106 in fig. 2), the determination result output unit 9 determines the current quality-affecting-factor candidate as the quality-affecting factor, and displays the quality-affecting factor (step S111 in fig. 2).
As described above, in the present embodiment, in order to determine a proper feature quantity with high explanatory power and high realizability as a quality factor from a feature quantity group that may affect the product quality, the user is notified of a quality factor candidate and a feature quantity similar thereto, thereby enabling the user to support the analysis. As a result, the user can specify the quality-affecting factor with high adequacy, and the relationship between the quality-affecting factor and the product quality can be easily modeled.
[ second embodiment ]
The second embodiment of the present invention is explained next. The present embodiment is an example of a notification method for effectively using a piping installation diagram as a quality-affecting-factor candidate, and is an example of effectively using a user's knowledge of a cause-and-effect relationship relating to movement of a substance or energy in order to identify a quality-affecting factor. In this embodiment, the configuration of the quality-affecting-factor determination support apparatus is also the same as that in the first embodiment, and therefore the description will be given using reference symbols in fig. 1.
Fig. 3 shows an example of a screen displayed by the determination result notification unit 7 of the quality-affecting-factor determination support apparatus. In the example of fig. 3, the determination result notification unit 7 displays the piping installation diagram of the production facility of the target product on the screen 70.
The determination result notification unit 7 displays a mark 700 indicating the name of the quality-affecting-factor candidate and the name of the process data that is the extraction source of the quality-affecting-factor candidate at a position where the process data is measured, in the pipe installation diagram, in a superimposed manner. Further, the determination result notification unit 7 superimposes and displays a mark 701 indicating the product quality data on the position of the pipe installation diagram where the product quality data is measured.
As can be seen from the screen 70 of fig. 3, for example, the process data of the flow rate a and the quality-affecting factor candidate of the maximum value of the flow rate a are selected. The image data of the piping installation diagram, the product quality data collected by the data collection unit 1, and the coordinate data of the process data on the piping installation diagram are registered in the storage unit 2 in advance.
In this way, in the present embodiment, by displaying the location of the process data of the quality-affecting-factor candidates and the extraction source of the quality-affecting-factor candidates on the piping installation diagram, it is possible to eliminate the quality-affecting-factor candidates which are difficult to interpret and violate the process insights of the user. For example, if there are process data from a to F as extraction sources of quality-affecting-factor candidates, process data of the flow rate D, the temperature E, and the flow rate F acquired from a device downstream of the location where the product quality data is measured should not be selected as the quality-affecting-factor candidates.
Therefore, the user instructs the quality-affecting-factor determination support apparatus to replace the three quality-affecting-factor candidates, which are the slope of the flow rate D, the variance of the temperature E, and the minimum value of the flow rate F, with other feature amounts in the group, respectively, or instructs the quality-affecting-factor determination support apparatus to exclude the quality-affecting-factor candidates from the determination result.
As described above, in the present embodiment, the feature amount that is not physically related to the product quality data or the feature amount whose causal relationship is reversed can be excluded from the quality-affecting-factor candidates based on the process knowledge.
In the example of fig. 3, other feature quantities in the group including the quality-affecting factor candidates are not shown. Therefore, when the user designates the mark 700 on the screen 70 using a pointing device such as a mouse, for example, the determination result notification unit 7 may display other feature amounts in the group including the quality-affecting-factor candidates indicated by the designated mark 700. Of course, the quality-affecting factor candidates and other feature quantities in the group including the quality-affecting factor candidates may be displayed simultaneously.
[ third embodiment ]
Next, a third embodiment of the present invention will be explained. The present embodiment is an example of effectively utilizing the kind of process data in order to determine the quality-affecting factor, and is an example of effectively utilizing the knowledge about the logical causal relationship.
Fig. 4 shows an example of a screen displayed by the determination result notification unit 7 of the quality-affecting-factor determination support apparatus. In the example of fig. 4, the determination result notification unit 7 displays the name of the quality-affecting-factor candidate, the name of the process data that is the extraction source of the quality-affecting-factor candidate, and the type of the process data (the operation amount, the measurement value, the setting value, and the like) in a table format on the screen 70. The type of the process data is defined in advance in a control system of a production facility of a product, and is registered in advance in the storage section 2.
In this way, in the present embodiment, by displaying the type of the process data as the extraction source of the quality-affecting-factor candidates, the user can select the quality-affecting-factor candidates extracted from the controllable process data.
The operation amount indicates an amount applied to the control target, the measured value indicates a value measured by a sensor or the like, and the set value indicates a value to be a target of the control amount. Generally, the manipulated variable and the set value can be directly changed from the control system, but the measured value cannot be directly changed. For example, in the example of fig. 4, it is difficult to directly change the outside air temperature or the flow rate of the mixer a in order to improve the product quality. On the other hand, there is a possibility that the flow rate setting value of the mixer a or the flow rate of the pipe B can be changed.
Therefore, the user instructs the quality-affecting-factor determination support device to replace the two quality-affecting-factor candidates of the outside air temperature and the flow rate of the mixer a with the other feature values in the group, or instructs the quality-affecting-factor determination support device to exclude the quality-affecting-factor candidates from the determination result.
In the example of fig. 4, other feature quantities in the group including the quality-affecting factor candidates are not shown. Therefore, for example, when the user designates a specific field of the quality-affecting-factor candidates on the screen 70 using a pointing device such as a mouse, the determination result notification unit 7 may display other feature amounts in the group including the designated quality-affecting-factor candidate. Of course, the quality-affecting factor candidates and other feature quantities in the group including the quality-affecting factor candidates may be displayed simultaneously.
[ fourth embodiment ]
Next, a fourth embodiment of the present invention will be explained. The present embodiment is an example of effectively utilizing the feature amount acquisition timing for determining the quality-affecting factor, and is an example of effectively utilizing the time information associated with the feature amount.
In the example of fig. 5, the determination result notification unit 7 displays, on the screen 70, a time-series graph 702 of the process data a as the extraction source of the quality-affecting-factor candidate A1 and a time-series graph 703 of the process data B as the extraction source of the other feature amount B1 included in the group including the quality-affecting-factor candidate A1. As described above, the time-series data of the process data is accumulated in the storage section 2. Then, the determination result notification unit 7 displays the timing (time information) at which the quality-affecting factor candidate A1 is extracted on the time-series graph 702, and displays the timing (time information) at which the feature amount B1 is extracted on the time-series graph 703.
Definitions of the respective feature quantities including the quality-affecting factor candidates are stored in the storage unit 2 in advance. Therefore, the determination result notification unit 7 may display the timing of extracting the quality-affecting factor candidate A1 and the timing of extracting the feature amount B1 based on the definition stored in the storage unit 2. In the case of a feature extracted from process data at a specific timing, as in the example of the quality-affecting factor candidate A1 (maximum value) in fig. 5, the timing at which the feature is extracted is indicated by vertical arrow 7000. Further, in the case of a feature amount extracted over a period of a plurality of pieces of process data as in the example of the feature amount B1 (elapsed time in the steady state), the timing of extracting the feature amount is displayed by the lateral arrow 7001.
In this way, in the present embodiment, by displaying the timing of extracting the feature amount together with the process data as the extraction source, the upstream side factor which is the true cause of the fluctuation in the product quality can be grasped, and the explanatory performance can be improved. For example, in the case where the quality-affecting-factor candidate A1 and the feature B1 are extracted as similar feature amounts as in the example of fig. 5, the possibility that the quality-affecting-factor candidate A1 extracted at a time point immediately after the start of the lot is the root cause of the quality variation with respect to the product is higher than the feature amount B1 extracted for the first time at the latter half of the lot elapsed time. Therefore, the user determines that the quality-affecting-factor candidate A1 is properly identified.
[ fifth embodiment ]
Next, a fifth embodiment of the present invention will be explained. The present embodiment is an example of effectively utilizing process data of each lot for the purpose of determining quality-affecting factors.
Fig. 6 shows an example of a screen displayed by the determination result notification unit 7 of the quality-affecting-factor determination support apparatus. As is clear from the description of the first embodiment, since the same kind of feature quantity is extracted for each batch of the batch process, the feature quantity of each batch can be displayed. In the example of fig. 6, as in fig. 5, on the screen 70, for each of the lots 101 to 103: time series graphs 704-101 to 704-103 of the process data a as the extraction source of the quality factor candidate A1, timings (vertical arrows) on the graphs 704-101 to 704-103 at which the quality factor candidate A1 is extracted, time series graphs 705-101 to 705-103 of the process data C as the extraction source of the other feature C1 included in the group including the quality factor candidate A1, and timings (vertical arrows) on the graphs 705-101 to 705-103 at which the feature C1 is extracted.
In this way, in the present embodiment, which feature of the process data is extracted is displayed for each batch, whereby the user can exclude the quality-affecting-factor candidates that are accidentally selected. Further, even if the user replaces the quality-affecting factor candidates with similar other feature amounts, it is possible to predict whether or not the product quality will not be deteriorated. For example, in the case where the quality-affecting-factor candidate A1 and the feature amount C1 are extracted as similar feature amounts as in the example of fig. 6, the quality-affecting-factor candidate A1 is more preferable as the quality affecting factor than the feature amount C1 having low reproducibility among lots. Therefore, the user determines that the quality-affecting-factor candidate A1 is properly specified.
In the fourth and fifth embodiments, time-series graphs of the process data are shown, but graphs of frequency changes of the process data may be shown.
[ sixth embodiment ]
Next, a sixth embodiment of the present invention will be explained. The present embodiment is an example of quantifying the influence of the feature quantity on the product quality in order to determine the quality-affecting factor.
Fig. 7 is a flowchart for explaining the operation of the quality-affecting-factor determination support apparatus according to the present embodiment, and fig. 8 is a diagram showing an example of a screen displayed by the determination result notification unit 7 of the quality-affecting-factor determination support apparatus.
The processing of steps S100 to S111 in fig. 7 is as described in the first embodiment. The quality estimation model generation unit 10 of the quality factor determination support device generates a quality estimation model for modeling the relationship between the quality factor candidates and the product quality and a quality estimation model for modeling the relationship between the other feature values included in the group including the quality factor candidates and the product quality at the time point when the candidate determination unit 6 completes the determination of the quality factor candidates (step S112 in fig. 7).
Examples of the method for generating the quality estimation model include a method such as multivariate regression, support vector regression, LASSO regression, ridge (Ridge) regression, and random forest regression (documents "t. Hastti, r. Ti bestimpani, j. Freudman (t. Hastie, r. Tibshirani, j. Friedman)", basic data mining/inference/prediction of statistical learning ", 2014.
When the user wants to confirm the influence of the feature quantity on the product quality, the user inputs a value of a desired product quality to the quality influence factor determination support device.
When the product quality desired by the user is specified (YES in step S113 in fig. 7), the determination result notifying unit 7 displays a quality estimation graph 706 indicating the relationship between the quality-affecting factor candidate A1 and the product quality and a quality estimation graph 707 indicating the relationship between the other feature quantity D1 included in the group including the quality-affecting factor candidate A1 and the product quality on the basis of the quality estimation model generated by the quality estimation model generating unit 10 as shown in fig. 8 (step S114 in fig. 7).
The determination result notification unit 7 displays the latest product quality value Q1 and the desired product quality value Q2 input by the user on the graph 706 and the graph 707, and further displays a mark 708 and a mark 709, where the mark 708 and the mark 709 indicate the quality factor candidates A1 and the change amounts of the feature amounts D1 required to improve the latest product quality Q1 to the desired product quality Q2.
In this way, in the present embodiment, by quantifying the sensitivity of the explanatory variable (characteristic amount) with respect to the product quality, the operation for improving the product quality can be simulated, and the realizability can be evaluated. For example, when the quality-affecting-factor candidate A1 and the feature D1 are extracted as similar features as in the example of fig. 8, the amount of change of the quality-affecting-factor candidate A1 is smaller when the amount of change required to improve the quality Q1 of the product manufactured recently to the desired value Q2 is compared. Therefore, the user determines that: the quality-affecting factor candidate A1 can be handled more easily, and the determination of the quality-affecting factor candidate A1 is appropriate.
When the amount of change of the quality-affecting-factor candidate or the amount of change of the feature value is specified from the user (step S115 in fig. 7), the process model generation unit 11 of the quality-affecting-factor determination support apparatus generates a process model for the process data that is the extraction source of the specified quality-affecting-factor candidate or feature value (step S116 in fig. 7). The process model may be made using known modeling techniques. When a regression model, which is a representative modeling technique, is used, a process model is generated in which the lot elapsed time is used as an explanatory variable and the value of the process data is used as a target variable, and the regression coefficient of the process model is varied to estimate virtual time-series data when the influencing factor candidates or the feature quantity is changed. The modeling technique is not limited to the regression model.
The process data estimation display unit 12 displays a time-series graph 710 of the process data as the extraction source of the quality-affecting factor candidates or the specified feature amounts specified by the user, and displays the timing (time information) at which the quality-affecting factor candidates or the feature amounts are extracted on the time-series graph 710. Further, the process data estimation display unit 12 displays the estimated time series graph 711 of the process data obtained when the quality-affecting factor candidates or the feature amount is changed by the change amount specified by the user, on the basis of the process model so as to overlap the time series graph 710 (step S117 in fig. 7).
In the example of fig. 9, a time-series graph 710 of the process data a as the extraction source of the quality-affecting-factor candidate A1 and the timing (vertical arrow mark) on the graph 710 at which the quality-affecting-factor candidate A1 is extracted are displayed, and an estimated time-series graph 711 of the process data a when the value of the quality-affecting-factor candidate A1 is changed is displayed so as to overlap with the graph 710.
In this way, the user can recognize the trajectory (time series graph) required for improving the product quality with respect to the process data serving as the quality-affecting factor candidate or the root of the feature quantity, and can determine the realizability or the risk with respect to the operation of the target process required for improving the product quality.
In the example of fig. 9, it is assumed that the determination results of the quality-affecting-factor candidates described in the first to fifth embodiments are notified (step S105), and after the quality-affecting-factor candidates are replaced and eliminated (step S107 to step S110), the processes of step S113 to step S117 are performed, but the processes of step S105 and step S107 to step S110 may or may not be performed.
The quality-influencing factor determination support device described in the first to sixth embodiments may be realized by a computer including a Central Processing Unit (CPU), a storage device, and an interface, and a program that controls these hardware resources. Fig. 10 shows a configuration example of the computer.
The computer includes a CPU300, a storage device 301, and an Interface (I/F) device 302. The I/F302 is connected to a management device of a production facility of a product. In such a computer, a program for implementing the quality-influencing factor determination support method of the present invention is stored in the storage device 301. The CPU300 executes the processes described in the first to sixth embodiments in accordance with the programs stored in the storage device 301.
Possibility of industrial utilization
The invention can be applied to the data analysis technology facing the batch process.
Description of the symbols
1: data collection unit
2: storage unit
3: feature value extraction unit
4: feature amount classification unit
5: representative feature value selection unit
6: candidate identification unit
7: determination result notification unit
8: determination result changing unit
9: determination result output unit
10: quality estimation model generation unit
11: process model generation unit
12: a process data estimation display unit.
Claims (11)
1. A quality influence factor determination support apparatus, characterized by comprising:
a data collection unit configured to collect product quality data of a batch process and process data of the batch process;
a feature value extraction unit configured to extract a feature value of the process data;
a feature amount classification unit configured to group all the feature amounts extracted by the feature amount extraction unit based on a similarity between the feature amounts;
a representative feature amount selection unit configured to select a representative feature amount for each group generated by the feature amount classification unit based on a relationship with the product quality data;
a candidate specification unit configured to specify, as a quality-affecting-factor candidate, a representative feature amount contributing to a variation in the product quality data from the representative feature amounts selected by the representative feature amount selection unit; and
and a determination result notification unit configured to notify a user of the determination result of the quality-affecting-factor candidate.
2. The quality influence factor determination support apparatus according to claim 1,
the determination result notification unit displays a piping installation diagram of production equipment for a product, displays a name of the quality-affecting-factor candidate and a name of process data that is an extraction source of the quality-affecting-factor candidate in a position where the process data is measured in the piping installation diagram in a superimposed manner, and displays a position where the product quality data is measured in the piping installation diagram.
3. Quality-influencing-factor determination support apparatus according to claim 1,
the determination result notification unit displays the name of the quality-affecting factor candidate, the name of the process data that is the extraction source of the quality-affecting factor candidate, and the type of the process data.
4. Quality-influencing-factor determination support apparatus according to claim 2 or 3,
the determination result notification unit displays, when instructed by a user, names of other feature quantities in the group including the quality-affecting-factor candidates.
5. Quality-influencing-factor determination support apparatus according to claim 1,
the determination result notification unit displays a first graph of process data that is an extraction source of the quality-affecting-factor candidates, displays time information that indicates the extraction of the quality-affecting-factor candidates on the first graph, displays a second graph of process data that is an extraction source of another feature included in the group including the quality-affecting-factor candidates, and displays time information that indicates the extraction of the feature on the second graph.
6. Quality-influencing-factor determination support apparatus according to claim 1,
the determination result notification unit displays a first graph of process data that is an extraction source of the quality-affecting-factor candidates, displays time information that indicates the extracted quality-affecting-factor candidates on the first graph, displays a second graph of process data that is an extraction source of another feature included in a group including the quality-affecting-factor candidates, displays time information that indicates the extracted feature on the second graph, and displays the first graph, the second graph, and the time information for each batch of a batch process.
7. The quality-influencing factor determination support apparatus according to claim 1, characterized by further comprising:
a quality estimation model generation unit configured to generate a first quality estimation model for modeling a relationship between the quality-affecting factor candidate and the product quality, and a second quality estimation model for modeling a relationship between the product quality and another feature amount included in the group including the quality-affecting factor candidate,
the determination result notification unit displays a first graph indicating a relationship between the quality-affecting factor candidate and the product quality based on the first quality estimation model, and displays a second graph indicating a relationship between another feature amount included in the group including the quality-affecting factor candidate and the product quality based on the second quality estimation model.
8. The quality-influencing factor determination support apparatus according to claim 7, characterized by further comprising:
a process model generation unit configured to generate a process model with respect to process data of an extraction source as the quality-affecting-factor candidate and process data of an extraction source as another feature amount included in a group including the quality-affecting-factor candidate; and
a process data estimation display unit configured to display a third graph of process data as an extraction source of the quality-affecting-factor candidate when a change amount of the quality-affecting-factor candidate is designated by a user, display time information of the extracted quality-affecting-factor candidate on the third graph, display an estimated time series of the process data when the change amount designated by the user is changed for the quality-affecting-factor candidate on the third graph in a superimposed manner based on the process model, display a fourth graph of the process data as an extraction source of the feature amount when a change amount of another feature amount included in a group including the quality-affecting-factor candidate is designated by the user, display time information of the extracted feature amount on the fourth graph, and display an estimated time series of the process data when the change amount designated by the user is changed based on the process model on the fourth graph in a superimposed manner.
9. The quality influence factor determination support apparatus according to any one of claims 1 to 8, characterized by further comprising:
the determination result changing unit is configured to, when a replacement instruction of a quality-affecting-factor candidate is issued from a user, replace the quality-affecting-factor candidate with another feature amount selected by the user in a group including the quality-affecting-factor candidate, and, when an exclusion instruction of a group is issued from the user, exclude the designated group from the determination result of the quality-affecting-factor candidate.
10. A quality influence factor determination support method characterized by comprising:
the method comprises the following steps of firstly, collecting product quality data of a batch process and process data of the batch process;
a second step of extracting a feature amount of the process data;
a third step of grouping all the feature quantities extracted in the second step based on the similarity of the feature quantities to each other;
a fourth step of selecting, for each group generated in the third step, a representative feature amount for each group based on a relationship with the product quality data;
a fifth step of determining, as quality influencing factor candidates, representative feature quantities that contribute to a variation in the product quality data from among the representative feature quantities selected in the fourth step; and
a sixth step of notifying a user of the determination result of the quality-affecting factor candidate.
11. The quality-influencing factor determination support method according to claim 10,
the sixth step includes a step of notifying, simultaneously with the determination result of the quality-affecting-factor candidate or in accordance with an instruction of a user, other feature amounts within the group including the quality-affecting-factor candidate.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2020-098263 | 2020-06-05 | ||
| JP2020098263A JP2021192137A (en) | 2020-06-05 | 2020-06-05 | Quality influencing factor identification support device and method |
| PCT/JP2021/014866 WO2021246051A1 (en) | 2020-06-05 | 2021-04-08 | Quality influence factor identification assistance device and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115668080A true CN115668080A (en) | 2023-01-31 |
Family
ID=78830834
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202180037000.7A Pending CN115668080A (en) | 2020-06-05 | 2021-04-08 | Quality influence factor determination support device and method |
Country Status (3)
| Country | Link |
|---|---|
| JP (1) | JP2021192137A (en) |
| CN (1) | CN115668080A (en) |
| WO (1) | WO2021246051A1 (en) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3161921B2 (en) * | 1994-10-27 | 2001-04-25 | ファナック株式会社 | Product quality influence factor analysis method and equipment, molding condition adjustment method, product quality judgment item selection method |
| JP2010146137A (en) * | 2008-12-17 | 2010-07-01 | Yokogawa Electric Corp | Parameter adjustment support device |
| JP7144676B2 (en) * | 2018-05-07 | 2022-09-30 | 富士通株式会社 | Information processing device, quality-related expression generation method, and quality-related expression generation program |
-
2020
- 2020-06-05 JP JP2020098263A patent/JP2021192137A/en active Pending
-
2021
- 2021-04-08 CN CN202180037000.7A patent/CN115668080A/en active Pending
- 2021-04-08 WO PCT/JP2021/014866 patent/WO2021246051A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| JP2021192137A (en) | 2021-12-16 |
| WO2021246051A1 (en) | 2021-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108875784A (en) | The method and system of the optimization based on data for the performance indicator in industry | |
| US11030167B2 (en) | Systems and methods for providing data quality management | |
| JP6670418B2 (en) | State prediction apparatus and state prediction control method | |
| JP2007052744A (en) | Classification rule creation supporting method | |
| JP2020064367A (en) | Abnormality detection device and abnormality detection method | |
| JP6835098B2 (en) | Factor analysis method, factor analyzer and factor analysis program | |
| WO2022038804A1 (en) | Diagnostic device and parameter adjustment method | |
| JP7030072B2 (en) | Time-series data monitoring system and time-series data monitoring method | |
| CN108780313A (en) | Increment related data digging technology for assembly line | |
| JP2017084224A (en) | Design support device, program, and design support method | |
| JP6879749B2 (en) | Support device and support method | |
| JP5571528B2 (en) | Production information management apparatus and production information management method | |
| KR102054500B1 (en) | Method for providing design drawing | |
| CN117556366B (en) | Data abnormality detection system and method based on data screening | |
| WO2006047464A2 (en) | Training a multi-dimensional, expert behavior-emulation system | |
| JP7283105B2 (en) | Analysis device and analysis method | |
| CN115668080A (en) | Quality influence factor determination support device and method | |
| CN117094599A (en) | Production quality management method and device based on machine learning and computer equipment | |
| CN116579842A (en) | Credit data analysis method and system based on user behavior data | |
| WO2020066124A1 (en) | Process state analysis device and process state display method | |
| WO2020225959A1 (en) | Design assistance system, design assistance method, and design assistance program | |
| JP4351578B2 (en) | Production planning support system and computer program | |
| TW202121221A (en) | Transferability determination apparatus, transferability determination method, and recording medium | |
| JP2006091937A (en) | Data-analyzing device, method therefor, and program | |
| JP2021140510A (en) | Design support apparatus, design support method, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |