Information Processing Apparatus and Information Processing Method

Abstract

To quickly extract lots using more energy than usual in the manufacturing process.SOLUTION: An information processing device 10 of the present disclosure includes a control unit 15 that acquires manufacturing performance data and utility data, generates cut-out data for each lot by cutting out the utility data in a range from a process start date and time to a process end date and time of the lot, classifies a plurality of the cut-out data by cluster analysis using a learning model generated based on previous manufacturing performance data and previous utility data, extracts notable lots by comparing the classification by cluster analysis with a classification by category codes included in the manufacturing performance data, and generates a notable lot list based on the notable lots.SELECTED DRAWING: Figure 10

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus and an information processing method.

Background Art

[0002] Conventionally, in the operation of a factory, it is important to detect abnormalities at an early stage. For example, Patent Document 1 discloses a technique for effectively utilizing a time-series data group related to a plant by graphically displaying specified time-series data among the time-series data group related to the plant, which is useful for early detection of abnormalities.

[0003] In recent years, as part of energy-saving measures for reducing greenhouse gas emissions, the importance of energy management in factories has been increasing.

[0004] From the perspective of energy conservation, in the manufacturing process of products in a factory, it is important to extract lots that use more energy than usual, investigate the causes, and take countermeasures.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] Extracting lots that use more energy than usual in the manufacturing process and taking countermeasures are currently being carried out as energy-saving measures. However, there has been a problem that it takes a long time to extract such lots.

[0007] Therefore, an object of the present disclosure is to provide an information processing apparatus and an information processing method capable of extracting, in a short time, a lot in which more energy than normal is used in a manufacturing process.

Means for Solving the Problems

[0008] An information processing apparatus according to some embodiments acquires manufacturing result data including information on lot number, category code, process start date and time, and process end date and time for each lot of a plurality of lots, and for each utility of a plurality of utilities, acquires utility data including time-series data of energy amounts, generates cut-out data by cutting out the utility data for each lot in the range from the process start date and time to the process end date and time of the lot, classifies and classifies a plurality of the cut-out data using a learning model generated based on previous manufacturing result data and previous utility data, compares the classification by cluster analysis with the classification by the category code included in the manufacturing result data, extracts a lot of interest, and generates a list of lots of interest based on the lot of interest, and includes a control unit. According to such an information processing apparatus, it is possible to extract, in a short time, a lot in which more energy than normal is used in a manufacturing process.

[0009] In an information processing apparatus according to an embodiment, the control unit may compare the classification by cluster analysis with the classification by the category code included in the manufacturing result data, and extract a lot whose classification results do not match as the lot of interest. Thereby, the lot of interest can be automatically extracted.

[0010] In an information processing apparatus according to an embodiment, the information processing apparatus further includes an output unit, and the control unit may cause the output unit to display the cut-out data of the lot of interest selected from the list of lots of interest. Thereby, the user can easily analyze the data of the utilities used in the manufacture of the lot of interest.

[0011] An information processing apparatus according to some embodiments acquires manufacturing result data including information on lot numbers, category codes, process start times, and process end times for each of a plurality of lots, acquires utility data including time-series data of energy amounts for each of a plurality of utilities, for each lot, extracts the utility data within a range from the process start time to the process end time of the lot to generate extracted data, performs cluster analysis on the plurality of extracted data for classification, compares the classification by the cluster analysis with the classification by the category code included in the manufacturing result data to determine the validity of the cluster analysis, and when the cluster analysis is valid, generates a learning model based on the parameters used in the cluster analysis, and includes a control unit. According to such an information processing apparatus, a learning model can be generated based on previous manufacturing result data and utility data.

[0012] In an information processing apparatus according to an embodiment, when the cluster analysis is not valid, the control unit may adjust the parameters used in the cluster analysis. Thereby, the parameters can be adjusted until an appropriate learning model can be generated.

[0013] An information processing method according to some embodiments is an information processing method in an information processing apparatus, including: obtaining manufacturing result data including information on lot number, category code, process start date and time, and process end date and time for each of a plurality of lots; obtaining utility data including time-series data of energy amount for each of a plurality of utilities; for each of the lots, extracting the utility data within a range from the process start date and time to the process end date and time of the lot to generate extracted data; classifying a plurality of the extracted data by performing cluster analysis; comparing the classification by cluster analysis with the classification by the category code included in the manufacturing result data to extract a target lot; and generating a target lot list based on the target lot. According to such an information processing method, it is possible to extract, in a short time, lots in which more energy than normal is used in the manufacturing process.

[0014] An information processing method according to some embodiments is an information processing method in an information processing apparatus, including: obtaining manufacturing result data including information on lot number, category code, process start date and time, and process end date and time for each of a plurality of lots; obtaining utility data including time-series data of energy amount for each of a plurality of utilities; for each of the lots, extracting the utility data within a range from the process start date and time to the process end date and time of the lot to generate extracted data; classifying a plurality of the extracted data by performing cluster analysis; comparing the classification by cluster analysis with the classification by the category code included in the manufacturing result data to determine the validity of the cluster analysis; and when the cluster analysis is valid, generating a learning model based on the parameters used in the cluster analysis. According to such an information processing method, a learning model can be generated based on previous manufacturing result data and utility data.

Advantages of the Invention

[0015] According to the present disclosure, it is possible to provide an information processing apparatus and an information processing method capable of extracting, in a short time, a lot in which more energy than usual is used in the manufacturing process.

Brief Description of the Drawings

[0016]

Figure 1

Figure 2

Figure 3

Figure 4A

Figure 4B

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Embodiments for Carrying Out the Invention

[0017] (Comparative Example) First, as a comparative example, an example of extracting a lot in the prior art in which more energy than normal is used from among a plurality of lots will be described.

[0018] Normally, products are manufactured in lots. The quantity of products included in one lot is, depending on the product, for example, on the order of several thousand to several hundred thousand.

[0019] When manufacturing products in a factory, data called production record data is generated. The production record data is data that associates a lot number, a category code, a process start date and time, a process end date and time, etc. for each lot.

[0020] The lot number is an identification number for identifying a lot. The category code is a code for classifying which category the lot belongs to. Here, a category is a grouping of lots with similar manufacturing processes. For example, lots in the range of about 100 to 250 may be classified into about 5 to 10 categories.

[0021] The process start date and time is the date and time when the process of manufacturing that lot starts. The process end date and time is the date and time when the process of manufacturing that lot ends.

[0022] Also, in a factory, utility data is generated. The utility data includes time-series data of the energy amounts of a plurality of utilities used in the manufacturing process of the product. Here, a utility is something necessary for the manufacturing process of the product, such as electricity, steam, water, compressed air, etc.

[0023] FIG. 1 shows an example of a case where time-series data of utilities is displayed as a graph in the prior art. In FIG. 1, the horizontal axis is the date and time. Also, the vertical axis is the energy amount. In FIG. 1, the plurality of graphs show a plurality of utilities.

[0024] In the prior art, utility data is cut out for each lot based on the process start date and time and the process end date and time included in the manufacturing performance data. In the example shown in FIG. 1, for example, the range indicated by reference numeral 101 corresponds to the data for one lot.

[0025] In the prior art, for each category, the cut-out utility data for one lot is arranged and displayed. FIG. 2 shows a state in which graphs of the cut-out utility data for lot A, lot B, and lot C for a certain category are arranged side by side.

[0026] Factory managers and the like look at the graphs arranged in the same category as shown in FIG. 2, and extract a lot with a specific feature such as a point where the energy amount is large compared to other lots as a target lot.

[0027] Managers and the like execute a process of comparing the graphs of the cut-out utility data for each category and extracting target lots.

[0028] The method for extracting a target lot as described above according to the prior art has the following problems. · Extracting a target lot with a specific feature by comparing the graphs of the cut-out utility data cannot be done by anyone, and can only be carried out by a skilled manager with knowledge of the manufacturing process. · When the number of categories and the number of lots are large, the work of extracting target lots takes a very long time. For example, it may take about several months. · After extracting the target lot, an investigation into the cause will be started for the utility with a higher energy amount than usual. However, since the work of extracting the target lot takes a very long time, the timing to start the investigation into the cause is delayed. Since product manufacturing continues until the cause is investigated and resolved, the period during which products are manufactured with a higher energy amount than usual becomes longer. ·If it takes a long time to extract the lot of interest and there is not enough data related to the factory operation remaining, it becomes difficult to investigate the cause.

[0029] (The information processing apparatus of the present disclosure) FIG. 3 is a diagram showing a schematic configuration of an information processing apparatus 10 according to an embodiment. The information processing apparatus 10 is an information processing apparatus aimed at solving the problems of the above-described conventional technology.

[0030] In a factory, the manufacturing process of products is carried out in lots. Usually, in lots with similar manufacturing processes, the amount of energy used to manufacture one product is about the same. However, due to some abnormality, there may be a lot in which more energy is used in the manufacturing process of the product than usual.

[0031] The information processing apparatus 10 is an information processing apparatus that extracts lots in which more energy may have been used than usual from a plurality of lots manufactured during a certain period. Hereinafter, a lot in which more energy may have been used than usual during the manufacturing process of a product may be referred to as a "lot of interest".

[0032] The information processing apparatus 10 may be installed in a factory or may be installed at a location separate from the factory. The information processing apparatus 10 may be a general-purpose computer such as a workstation or a personal computer, or may be a dedicated computer used for the purpose of this embodiment.

[0033] With reference to FIG. 3, the configuration of the information processing apparatus 10 will be described. The information processing apparatus 10 includes a communication unit 11, a storage unit 12, an input unit 13, an output unit 14, and a control unit 15.

[0034] The communication unit 11 includes at least one of a communication module corresponding to wired communication and a communication module corresponding to wireless communication. The information processing apparatus 10 can communicate with other devices via the communication unit 11.

[0035] The storage unit 12 is, for example, a semiconductor memory, a magnetic memory, an optical memory, etc., but is not limited thereto. The storage unit 12 may function as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 12 stores any information used for the operation of the information processing apparatus 10. For example, the storage unit 12 may store a system program, an application program, and various information received by the communication unit 11. A part of the storage unit 12 may be installed outside the information processing apparatus 10. In that case, a part of the storage unit 12 installed outside may be connected to the information processing apparatus 10 via an arbitrary interface.

[0036] The input unit 13 includes one or more input interfaces that detect user input and acquire input information based on the user's operation. For example, the input unit 13 includes, but is not limited to, physical keys, capacitive keys, a touch screen provided integrally with the display of the output unit 14, or a microphone that receives voice input.

[0037] The output unit 14 includes one or more output interfaces that output information and notify the user. For example, the output unit 14 includes, but is not limited to, a display that outputs information as an image, a speaker that outputs information as voice.

[0038] The control unit 15 includes at least one processor, at least one dedicated circuit, or a combination thereof. The processor is a general-purpose processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), or a dedicated processor specialized for specific processing. The dedicated circuit is, for example, an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). The control unit 15 executes processing related to the operation of the information processing apparatus 10 while controlling each part of the information processing apparatus 10.

[0039] The control unit 15 acquires utility data, production result data, and category master data. The control unit 15 may acquire the utility data, production result data, and category master data from another device via the communication unit 11. Alternatively, the control unit 15 may acquire the utility data, production result data, and category master data by an input operation of the user to the input unit 13. In the following description, it is assumed that the control unit 15 acquires the utility data, production result data, and category master data from another device via the communication unit 11.

[0040] FIGS. 4A and 4B are diagrams showing an example of utility data. Since the utility data shown in FIGS. 4A and 4B has many columns and it is difficult to show it in one diagram, it is shown over two diagrams.

[0041] The utility data includes time-series data of energy amounts for each of a plurality of utilities. Here, a utility is something necessary for manufacturing products in a factory, for example, electric power, steam, water, compressed air, etc. Note that in the utility data, the amounts of electric power, steam, water, and compressed air are stored as data converted into energy amounts.

[0042] In FIGS. 4A and 4B, the utility data is data in units of one hour, and one row shows data for one hour. Also, E-13102 to E-14004 show data of the electric power utility, and one column shows data of one electric power utility facility. For example, in the column of E-13103, the data at 0:00 on January 1, 2020 is "43", which indicates that the energy amount used between 0:00 and 1:00 on January 1, 2020 in the electric power utility facility E-13103 is "43".

[0043] In addition, S-0034 to S-0038 show the data of steam utilities, and each column shows the data of one steam utility facility. For example, the data at 0:00 on January 1, 2020 in the column of S-0034 is "0", which indicates that the amount of energy used between 0:00 and 1:00 on January 1, 2020 in the steam utility facility named S-0034 is "0".

[0044] In addition, W-0042 to W-0045 show the data of water utilities, and each column shows the data of one water utility facility. For example, the data at 0:00 on January 1, 2020 in the column of W-042 is "0.01", which indicates that the amount of energy used between 0:00 and 1:00 on January 1, 2020 in the water utility facility named W-042 is "0.01".

[0045] Figure 5 is a diagram showing an example of production result data. The production result data includes the production result data for a plurality of lots. In the example shown in Figure 5, the production result data includes, for each lot, lot number, lot size, unit, item name, category code, category group, process A start date and time, process A end date and time, process B start date and time, process B end date and time, process C start date and time, process C end date and time, item code, and category name data.

[0046] Note that the production result data shown in Figure 5 is only an example, and the types of data included in the production result data are not limited to the types of data shown in Figure 5. The production result data only needs to include, for each lot, at least the lot number, category code, process start date and time of one process, and process end date and time of one process.

[0047] The lot number is an identification number for specifying a lot. The lot size indicates the quantity of products included in that lot. The unit indicates the unit used for the products included in that lot. The item name indicates the name of the products included in that lot. The category code is a code for classifying which category the lot belongs to. The category group is the number assigned to that category and corresponds one-to-one with the category code. The start date and time of Process A is the date and time when Process A for manufacturing that lot started. The end date and time of Process A is the date and time when Process A for manufacturing that lot ended. The start date and time of Process B is the date and time when Process B for manufacturing that lot started. The end date and time of Process B is the date and time when Process B for manufacturing that lot ended. The start date and time of Process C is the date and time when Process C for manufacturing that lot started. The end date and time of Process C is the date and time when Process C for manufacturing that lot ended. The item code is a code that corresponds one-to-one with the item name. The category name is the name of the category and corresponds one-to-one with the category code and the category group.

[0048] Here, a category is a grouping of lots with similar manufacturing processes. For example, lots of about 100 to 250 can be classified into about 5 to 10 categories. In the manufacturing performance data, how to classify multiple lots into categories can be arbitrarily set by the factory management staff, etc.

[0049] Figure 6 is a diagram showing an example of category master data. The category master data is data that associates an item code with an item name, a category code, a category name, and a category group. The category master data is used to define the correspondence between items and categories.

[0050] The information processing device 10 executes two types of processing. One is the processing for generating a learning model used when extracting a target lot. The other is the processing for extracting a target lot using the learning model.

[0051] <Learning Model Generation Process> First, the process by which the information processing apparatus 10 generates a learning model will be described.

[0052] The control unit 15 of the information processing apparatus 10 acquires production result data and utility data via the communication unit 11. The production result data and utility data acquired by the control unit 15 are learning data used to generate a learning model, and may be data within a wide time range acquired during the production of previously manufactured products.

[0053] For each lot included in the production result data, the control unit 15 reads out the process start date and time and the process end date and time. For each lot, the control unit 15 extracts the utility data within the range from the process start date and time of the lot to the process end date and time, and generates extracted data. Hereinafter, the term "extracted data" may be used to mean data obtained by extracting utility data within the range from the process start date and time of a lot to the process end date and time.

[0054] The control unit 15 performs cluster analysis using the extracted data of all lots included in the production result data. By performing cluster analysis, the lots included in the production result data are classified into a plurality of clusters.

[0055] Before performing cluster analysis, the control unit 15 may adjust the parameters used for cluster analysis as preprocessing. Adjustment of the parameters used for cluster analysis may include, for example, adjustment of the weighting of the energy amount of each utility included in the extracted data, adjustment of the parameters set in the algorithm of cluster analysis, and the like.

[0056] The control unit 15 may perform adjustment of the parameters used for cluster analysis based on an input to the input unit 13 by a user such as a factory manager, or may perform it automatically.

[0057] The control unit 15 determines the validity of the cluster analysis by comparing the classification by cluster analysis with the classification by the category code included in the production result data. Here, the classification by the category code included in the production result data is a classification in which the lots included in the production result data are classified for each category code actually included in the production result data. That is, the classification by the category code included in the production result data can be said to be the correct answer data.

[0058] For example, the control unit 15 may determine that the cluster analysis is valid when the classification by cluster analysis and the classification by the category code included in the production result data match at a ratio equal to or higher than a predetermined threshold value. The predetermined threshold value may be, for example, 80%.

[0059] Alternatively, the user of the information processing apparatus 10 may compare the classification results to determine the validity of the cluster analysis by comparing the classification by cluster analysis with the classification by the category code included in the production result data. If a skilled user makes the determination, the validity of the cluster analysis can be appropriately determined based on the knowledge.

[0060] When the control unit 15 determines that the cluster analysis is valid, it generates a learning model based on the parameters used in the cluster analysis. The generated learning model is used in the extraction process of the target lot described later.

[0061] When the control unit 15 determines that the cluster analysis is not valid, it adjusts the parameters used in the cluster analysis and executes the cluster analysis again.

[0062] <Target Lot Extraction Process> Subsequently, the process in which the information processing apparatus 10 extracts the target lot will be described.

[0063] The control unit 15 of the information processing apparatus 10 acquires manufacturing result data and utility data via the communication unit 11. The manufacturing result data and utility data acquired by the control unit 15 are data that are targets for extracting the lot of interest.

[0064] For each lot included in the manufacturing result data, the control unit 15 reads out the process start date and time and the process end date and time. For each lot, the control unit 15 extracts the utility data within the range from the process start date and time to the process end date and time of the lot, and generates extracted data.

[0065] The control unit 15 performs cluster analysis and classification on a plurality of extracted data using the learning model generated by the above-described learning model generation process based on previous manufacturing result data and previous utility data.

[0066] The control unit 15 compares the classification by cluster analysis with the classification by the category code included in the manufacturing result data, and extracts the lots whose classification results do not match as the lots of interest. Here, since it can be said that the classification by the category code included in the manufacturing result data is the correct data, it can also be said that the control unit 15 extracts, as the lots of interest, the lots that do not match the correct data among the classifications by cluster analysis.

[0067] The control unit 15 generates a list of lots of interest based on the extracted lots of interest. An example of the list of lots of interest is shown in FIG. 7. Comparing the list of lots of interest shown in FIG. 7 with the manufacturing result data shown in FIG. 5, it can be seen that the list of lots of interest is a list from which a part of the manufacturing result data has been extracted.

[0068] The control unit 15 can thus automatically extract the target lot by cluster analysis using the learning model, and can extract in a short time a lot that may have used more energy than usual in the manufacturing process. The control unit 15 can automatically extract the target lot, for example, in about several hours. The user can immediately start to improve the manufacturing process based on the target lot extracted by the information processing apparatus 10.

[0069] In addition, the operation that the user performs when extracting the target lot is only to specify which manufacturing result data and which utility data are to be read into the information processing apparatus 10. Therefore, the user can easily extract the target lot with a simple operation.

[0070] The control unit 15 can cause the output unit 14 to display the data based on the target lot list. FIG. 8 shows an example in which the analysis data 200 based on the target lot list is displayed on the output unit 14.

[0071] In the example shown in FIG. 8, the analysis data 200 displayed on the output unit 14 includes a target lot list 201, a utility selection screen 202, and a graph 203.

[0072] The target lot list 201 displays the target lot list. The user can select which cut-out data of the target lot is to be displayed by selecting one of the target lot lists displayed in the target lot list 201.

[0073] The utility selection screen 202 is a screen for selecting the utility to be displayed in the graph 203. The cut-out data of the utility for which the “display” column is checked in the utility selection screen 202 is displayed in the graph 203.

[0074] Graph 203 displays the cut-out data of the utilities selected on the utility selection screen 202 for the selected lots of interest in the list of lots of interest 201.

[0075] Since the control unit 15 can cause the output unit 14 to display the analysis data 200 in this way, the user can easily analyze the data of the utilities used in the production of the lots of interest by looking at the analysis data 200.

[0076] With reference to the flowchart shown in FIG. 9, the procedure of the learning model generation process by the information processing apparatus 10 according to an embodiment will be described.

[0077] In step S101, the control unit 15 of the information processing apparatus 10 acquires production result data and utility data as learning data via the communication unit 11.

[0078] In step S102, for each lot included in the production result data, the control unit 15 reads out the process start date and time and the process end date and time, and for each lot, cuts out the utility data in the range from the process start date and time to the process end date and time of the lot.

[0079] In step S103, the control unit 15 adjusts the parameters used for the cluster analysis.

[0080] In step S104, the control unit 15 performs cluster analysis using the cut-out data of all the lots included in the production result data.

[0081] In step S105, the control unit 15 compares the classification by the cluster analysis with the classification by the category code included in the production result data, and determines the validity of the cluster analysis.

[0082] If it is determined in step S105 that the cluster analysis is not valid, the control unit 15 returns to step S103.

[0083] If it is determined in step S105 that the cluster analysis is appropriate, the control unit 15 proceeds to step S106.

[0084] In step S106, the control unit 15 generates a learning model based on the parameters used in the cluster analysis in step S104.

[0085] With reference to the flowchart shown in FIG. 10, the procedure of the target lot extraction process by the information processing apparatus 10 according to an embodiment will be described.

[0086] In step 201, the control unit 15 of the information processing apparatus 10 acquires production result data and utility data via the communication unit 11. The production result data and utility data acquired here are the data targeted for extracting the target lot.

[0087] In step S202, for each lot included in the production result data, the control unit 15 reads out the process start date and time and the process end date and time, and for each lot, cuts out the utility data in the range from the process start date and time to the process end date and time of the lot.

[0088] In step S203, the control unit 15 performs cluster analysis on the cut-out data of all lots included in the production result data using the learning model generated by the learning model generation process.

[0089] In step S204, the control unit 15 compares the classification by the cluster analysis with the classification by the category code included in the production result data, and extracts the lots whose classification results do not match as the target lots.

[0090] In step S205, the control unit 15 generates a target lot list based on the extracted target lots.

[0091] According to the information processing apparatus 10 according to the above-described embodiment, it is possible to extract, in a short time, a lot in which more energy than usual is used in the manufacturing process. More specifically, the information processing apparatus 10 acquires manufacturing result data and utility data, and for each lot, cuts out the utility data in the range from the process start date and time to the process end date and time of the lot to generate cut-out data, and uses a learning model generated based on previous manufacturing result data and previous utility data to perform cluster analysis on a plurality of cut-out data for classification, compares the classification by cluster analysis with the classification by the category code included in the manufacturing result data to extract a target lot, and generates a target lot list based on the target lot. Therefore, the information processing apparatus 10 can automatically extract a target lot by comparing the classification by cluster analysis with the classification by the category code included in the manufacturing result data, so that it is possible to extract, in a short time, a lot in which more energy than usual is used in the manufacturing process.

[0092] It is obvious to those skilled in the art that the present disclosure can be realized in other predetermined forms other than the above-described embodiments without departing from its spirit or its essential features. Therefore, the foregoing description is illustrative and not restrictive. The scope of the disclosure is defined by the appended claims rather than by the foregoing description. Some changes within the scope of equivalents of any change are included therein.

[0093] For example, the arrangement and number of each of the above-described components are not limited to the content shown in the above description and the drawings. The arrangement and number of each component may be arbitrarily configured as long as its function can be realized.

[0094] For example, in the above-described embodiment, the case where the information processing apparatus 10 executes both the learning model generation process and the target lot extraction process has been described, but the information processing apparatus that executes the learning model generation process and the information processing apparatus that executes the target lot extraction process may be different information processing apparatuses.

Explanation of Reference Numerals

[0095] 10 Information processing apparatus 11 Communication unit 12 Memory unit 13 Input unit 14 Output unit 15 Control unit

Claims

1. For each lot of a plurality of lots, obtain manufacturing performance data including information on lot number, category code, process start date and time, and process end date and time, For each utility of a plurality of utilities, obtain utility data including time-series data of energy amount, For each of the lots, cut out the utility data within the range from the process start date and time to the process end date and time of the lot to generate cut-out data, Using a learning model generated based on previous manufacturing performance data and previous utility data, perform cluster analysis on a plurality of the cut-out data for classification, Compare the classification by cluster analysis with the classification by the category code included in the manufacturing performance data, and extract a lot of interest, A control unit that generates a list of lots of interest based on the lots of interest, An information processing apparatus comprising:

2. In the information processing apparatus according to claim 1, The control unit compares the classification by cluster analysis with the classification by the category code included in the manufacturing performance data, and extracts a lot whose classification results do not match as the lot of interest. An information processing apparatus.

3. In the information processing apparatus according to claim 1, Further comprising an output unit, The control unit causes the output unit to display the cut-out data of the lot of interest selected from the list of lots of interest. An information processing apparatus.

4. For each lot of a plurality of lots, obtain manufacturing performance data including information on lot number, category code, process start date and time, and process end date and time, For each utility of a plurality of utilities, obtain utility data including time-series data of energy amount, For each lot, cut out the utility data within the range from the process start date and time to the process end date and time of the lot to generate cut-out data. Perform cluster analysis on and classify a plurality of the cut-out data. Compare the classification by cluster analysis with the classification by the category code included in the manufacturing result data to determine the validity of the cluster analysis. When the cluster analysis is valid, generate a learning model based on the parameters used in the cluster analysis, a control unit. An information processing apparatus comprising the same.

5. In the information processing apparatus according to claim 4, When the cluster analysis is not valid, the control unit adjusts the parameters used in the cluster analysis, an information processing apparatus.

6. An information processing method in an information processing apparatus, For each of a plurality of lots, a step of obtaining manufacturing result data including information on lot number, category code, process start date and time, and process end date and time; For each of a plurality of utilities, a step of obtaining utility data including time-series data of energy amount; For each lot, a step of cutting out the utility data within the range from the process start date and time to the process end date and time of the lot to generate cut-out data; A step of performing cluster analysis on and classifying a plurality of the cut-out data using a learning model generated based on previous manufacturing result data and previous utility data; A step of comparing the classification by cluster analysis with the classification by the category code included in the manufacturing result data to extract a target lot; A step of generating a target lot list based on the target lot; An information processing method including the same.

7. An information processing method in an information processing apparatus, For each lot of a plurality of lots, a step of obtaining production result data including information on lot number, category code, process start date and time, and process end date and time; For each utility of a plurality of utilities, a step of obtaining utility data including time-series data of energy amounts; For each of the lots, a step of extracting the utility data within a range from the process start date and time to the process end date and time of the lot to generate extracted data; A step of performing cluster analysis on a plurality of the extracted data and classifying them; A step of comparing the classification by cluster analysis with the classification by the category code included in the production result data to determine the validity of the cluster analysis; When the cluster analysis is valid, a step of generating a learning model based on the parameters used in the cluster analysis; An information processing method including the above.

Images