JP2024046789A - Apparatus, method and program for analyzing data relating to a production system - Google Patents

Abstract

[Problem] To provide information useful for identifying the cause of defects in product quality. [Solution] The method comprises the steps of acquiring time series data of variables and assigning numbers to multiple data elements of the time series data. The assigning step includes, for each processing unit that is repeatedly performed, a) assigning a reference number to a data element at the start timing of the processing unit, b) assigning consecutive numbers in chronological order to one or more data elements corresponding to the processing unit so that the consecutive numbers are consecutive to the reference number, and c) assigning consecutive numbers in chronological order to one or more data elements corresponding to the processing unit preceding the processing unit so that the consecutive numbers are consecutive to the reference number. [Selected Figure] Figure 1

Description

This disclosure relates to an apparatus, method, and program for analyzing data related to a production system.

In general, in a production system, multiple products are produced by repeatedly performing a predetermined processing unit. Anomalies that occur in a processing unit can affect the quality of the product. Therefore, if the quality of a product is poor, the status of the processing unit performed to produce that product is analyzed to find the cause.

JP 2021-196950 A (Patent Document 1) discloses a technology that divides time-series numerical values generated by a sensor installed in a production system into groups for each processing unit period, and provides display information showing the time change of feature values obtained from the numerical values of each group.

JP 2021-196950 A

There may be cases where the cause of a product's quality defect cannot be identified using only the display information provided by the technology disclosed in Patent Document 1.

This disclosure has been made in consideration of the above problems, and its purpose is to realize an apparatus, method, and program that can provide information useful for identifying the cause of defects in product quality.

According to one example of the present disclosure, an apparatus for analyzing data related to a production system includes an acquisition unit that acquires time-series data of variables measured in the production system, and an assignment unit that assigns numbers to each of a plurality of data elements included in the time-series data. For each processing unit that is repeatedly performed in the production system, the assignment unit assigns a reference number to a first data element at the start timing of the processing unit among the plurality of data elements. Furthermore, for each processing unit, the assignment unit assigns consecutive numbers in chronological order to one or more second data elements measured during the period in which the processing unit is performed among the plurality of data elements so as to be continuous with the reference number. For each processing unit, the assignment unit assigns consecutive numbers in chronological order to one or more third data elements measured during the period in which the processing unit preceding the processing unit is performed among the plurality of data elements so as to be continuous with the reference number.

According to this disclosure, the user can easily check the data set including one or more third data elements, a first data element, and one or more second data elements, which are assigned consecutive numbers centered on a reference number, for each processing unit. The data set includes not only the first data element and the second data element during the period in which the corresponding processing unit is performed, but also the third data element during the period in which the processing unit preceding the corresponding processing unit is performed. Therefore, for example, when a defect occurs in a certain product, the user can notice an abnormality not only in the processing unit but also in the processing unit preceding the processing unit by checking the data set corresponding to the processing unit performed to produce the product. As a result, it becomes easier to identify the cause of the defect. In this way, the above-mentioned device can provide information useful for finding the cause of the defect in the quality of the product.

In the above disclosure, the reference number is 0. The number assigned to each of the one or more second data elements is positive. The number assigned to each of the one or more third data elements is negative.

According to the above disclosure, by checking whether the number of the data element indicating an anomaly is positive or negative, the user can easily determine whether the anomaly occurred during the target processing unit or the processing unit immediately prior to that.

In the above disclosure, the processing unit includes a first process for maintaining an equipment installed in the production system in an on state, and a second process for maintaining the equipment in an off state after the first process. The start timing is the timing at which the equipment changes from an off state to an on state. For each processing unit, the assignment unit determines data elements measured during the period when the first process included in the processing unit is being performed as one or more second data elements. Furthermore, for each processing unit, the assignment unit determines data elements measured during the period when the second process included in the processing unit preceding the processing unit is being performed as one or more third data elements.

According to the above disclosure, the first to third data elements can be easily determined based on the state of the device.

In the above disclosure, the assignment unit determines, for each processing unit, data elements measured during a period from the start timing of the processing unit to the start timing of the processing unit following the processing unit as one or more second data elements. Furthermore, for each processing unit, the assignment unit determines, for each processing unit, data elements measured during a period from the start timing of the processing unit preceding the processing unit to the start timing of the processing unit itself as one or more third data elements.

According to the above disclosure, the first to third data elements are easily determined based on the start timing of each processing unit.

In the above disclosure, the assignment unit determines, for each processing unit, a predetermined number of data elements preceding the first data element as one or more third data elements.

According to the above disclosure, it is possible to assign consecutive numbers to a predetermined number of data elements before the start of a processing unit so that the consecutive numbers are consecutive with the reference numbers assigned to the data elements at the start timing of the processing unit.

In the above disclosure, the reference number is the same as the value of the predetermined number. As a result, the oldest third data element among one or more third data elements to which consecutive numbers are assigned is assigned the number "0". As a result, a user can easily identify the oldest data element among a block of consecutively numbered data elements.

In the above disclosure, the device further includes a providing unit that extracts, for each processing unit, a data set including one or more third data elements, a first data element, and one or more second data elements, which are assigned consecutive numbers according to the processing unit, from the time-series data, and provides provided data generated based on the data set. The provided data indicates a time change in a variable or a feature calculated from the variable.

According to the above disclosure, users can investigate the cause of product quality defects by checking the provided data.

In the above disclosure, the device further includes a determination unit that extracts, for each processing unit, from the time series data, a data set including one or more third data elements, a first data element, and one or more second data elements, each of which is assigned a consecutive number according to the processing unit, and determines whether the data set includes a data element indicating an abnormality in the production system.

According to the above disclosure, depending on the judgment result by the judgment unit, the user can take measures early on to prevent the production of defective products.

According to another example of the present disclosure, a method for analyzing data related to a production system includes a step of acquiring time series data of variables measured in the production system, and a step of assigning numbers to each of a plurality of data elements included in the time series data. The assigning step includes, for each processing unit repeatedly performed in the production system, a) a step of assigning a reference number to a first data element at the start timing of the processing unit among the plurality of data elements, b) a step of assigning a consecutive number in chronological order to one or more second data elements measured during the period during which the processing unit is performed among the plurality of data elements so as to be continuous with the reference number, and c) a step of assigning a consecutive number in chronological order to one or more third data elements measured during the period during which the processing unit preceding the processing unit is performed among the plurality of data elements so as to be continuous with the reference number.

According to yet another example of the present disclosure, a program causes a computer to execute the above method.

These disclosures can also provide useful information for identifying the causes of product quality defects.

This disclosure can provide useful information for identifying the causes of product quality defects.

FIG. 2 is a schematic diagram illustrating a method according to an embodiment. FIG. 1 is a diagram showing an overview of a production system according to an embodiment. 2 is a block diagram showing an example of a hardware configuration of a controller according to an embodiment; FIG. 2 is a block diagram illustrating an example of a hardware configuration of a computer. FIG. 2 is a diagram illustrating an example of a functional configuration of a controller according to the present embodiment. FIG. 2 is a diagram showing the transition of processes executed in each processing unit. 13 is a flowchart showing the process flow of a first numbering method. FIG. 13 is a diagram showing an example of numbering according to a first method. FIG. 9 is a diagram showing a data set for each processing unit extracted from the time-series data shown in FIG. 8 . 13 is a flowchart showing the process flow of a second numbering method. FIG. 13 is a diagram showing an example of numbering according to a second method. FIG. 12 is a diagram showing a data set for each processing unit extracted from the time series data shown in FIG. 11 . FIG. 13 is a diagram showing an example of a screen generated using provided data. FIG. 13 is a diagram showing a functional configuration of a computer according to a fourth modified example.

The embodiment of the present invention will be described in detail with reference to the drawings. Note that the same or corresponding parts in the drawings will be given the same reference numerals and their description will not be repeated.

§1 Application Example An example of a situation in which the present invention is applied will be described with reference to Fig. 1. Fig. 1 is a schematic diagram showing a method according to an embodiment. Fig. 1 shows a method for analyzing data related to a production system.

As shown in FIG. 1, the method includes step S100 of acquiring time series data 2 of variables measured in a production system, and step S200 of assigning numbers to each of a plurality of data elements 3 included in the time series data 2. The time series data 2 associates, for each time, a time with the value of the variable measured at that time. The data elements indicate the value of the variable at each time.

In a production system, a predetermined processing unit (also called a "batch") is repeatedly performed to produce a plurality of products. In Fig. 1, a period _Tk is a period during which the kth processing unit is performed.

Step S200 includes steps S201 to S203 for each processing unit. Figure 1 shows steps S201 to S203 performed for the kth processing unit.

Step S201 is a step of assigning a reference number to the first data element 3a at the start timing of the target processing unit among the multiple data elements 3. In the example shown in Fig. 1, the reference number "0" is assigned to the first data element 3a at the start timing tk of _{the kth} processing unit.

Step S202 is a step of assigning consecutive numbers in chronological order to be consecutive to a reference number to one or more second data elements 3b measured during a period in which a target processing unit is being performed among the multiple data elements 3. In the example shown in Fig. 1, numbers "1", "2", ..., "N" are assigned in chronological order to N second data elements 3b measured during a period _Tk in which the kth processing unit is being performed.

Step S203 is a step of assigning consecutive numbers in chronological order to be continuous with the reference number to one or more third data elements 3c measured during the period in which the processing unit preceding the target processing unit is being performed among the multiple data elements 3. In the example shown in Fig. 1, the numbers "-M", ..., "-1" are assigned in chronological order to M third data elements 3c measured during the period T _k- 1 in which the k-1th processing unit preceding the kth processing unit is being performed.

In the example shown in FIG. 1, the time series data 2 shows an abnormal value just before the period T _k (see the frame line 8). The abnormal value shown by the frame line 8 may affect the quality of the product produced by the next processing unit, not the product produced by the processing unit performed in the period T _k-1 including the timing at which the abnormal value was measured. For example, when the time series data 2 indicates the temperature of a heating device, the abnormal value shown by the frame line 8 corresponds to the cooling timing after the heating process in the period T _{k -1, and does not affect the quality of the product produced by the processing unit performed in the period T k-1} . However, the abnormal value shown by the frame line 8 may affect the temperature rise process in the next period T _k . As a result, the quality of the product produced by the processing unit performed in the period T _k may deteriorate. Since an inexperienced worker does not have such knowledge, when a product produced by a processing unit performed in the period T _k is defective, the worker only checks the status of the processing unit in the period T _k and is unable to identify the cause of the defect.

According to the method of this embodiment, the user can easily check the data set including one or more third data elements 3c, first data elements 3a, and one or more second data elements 3b, which are assigned consecutive numbers centered on a reference number, for each processing unit. The data set includes not only the first data elements 3a and second data elements 3b during the period in which the corresponding processing unit is performed, but also the third data elements 3c during the period in which the processing unit before the corresponding processing unit is performed. Therefore, for example, when a defect occurs in a certain product, the user can notice an abnormality not only in the processing unit but also in the processing unit before the processing unit by checking the data set corresponding to the processing unit performed to produce the product. As a result, it becomes easier to identify the cause of the defect. In this way, the method of this embodiment can provide information useful for finding the cause of the defect in the quality of the product.

§2 Specific example <Production system>
2 is a diagram showing an overview of a production system according to an embodiment of the present invention. As shown in FIG. 2, the production system 1 includes a controller 100, a field device group 200, and a computer 300.

The controller 100 corresponds to an industrial controller that controls control targets such as various facilities and devices. The controller 100 is a type of computer that executes control calculations, and may typically be embodied as a PLC (programmable logic controller). The controller 100 is connected to the field device group 200 via a field bus 5, and exchanges data with the field device group 200 according to a predetermined cycle.

The field device group 200 is installed at a production site (hereinafter also referred to as "field") such as a production line, and includes multiple field devices related to production work. The field device group 200 includes, for example, devices that exert some kind of physical effect on the field, and input/output devices such as sensors that measure the state of the field. In the example shown in FIG. 2, the field device group 200 includes a heating device 200a and a sensor 200b that measures the temperature of an area to be heated by the heating device 200a.

The computer 300 is a general-purpose computer and operates as a user interface. Specifically, the computer 300 displays information received from the controller 100 and transmits information input by the user to the controller 100.

<Hardware Configuration>
Fig. 3 is a block diagram showing an example of a hardware configuration of a controller according to an embodiment. As shown in Fig. 3, the controller 100 includes a processor 101 such as a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit), a chipset 102, a main memory 103, a storage 104, a higher-level network controller 105, a field bus controller 106, a USB controller 107, and a memory card interface 108.

The processor 101 reads various programs stored in the storage 104, expands them in the main memory 103, and executes them to realize control calculations for controlling the control target. The chipset 102 controls data transmission between the processor 101 and each component.

Storage 104 stores a system program 110 for implementing basic processing and a user program 111 for implementing control calculations.

The upper network controller 105 controls data exchange with external devices (including, for example, MES (Manufacturing Execution System) servers) via the upper network.

The field bus controller 106 controls data exchange with the field device group 200 via the field bus 5.

The USB controller 107 controls data exchange with an external device (e.g., computer 300) via a USB connection.

The memory card interface 108 is configured to allow the memory card 120 to be attached and detached, and is capable of writing data to the memory card 120 and reading various data (such as user programs) from the memory card 120.

Although FIG. 3 shows an example of a configuration in which the processor 101 executes a program to provide the necessary processing, some or all of the provided processing may be implemented using a dedicated hardware circuit (e.g., an ASIC or FPGA). Alternatively, the main part of the controller 100 may be realized using hardware that follows a general-purpose architecture (e.g., an industrial PC based on a general-purpose PC). In this case, virtualization technology may be used to run multiple OSs with different uses in parallel, and the necessary applications may be executed on each OS.

Figure 4 is a block diagram showing an example of the hardware configuration of a computer. Computer 300 has a structure that conforms to a general-purpose computer architecture. As shown in Figure 4, computer 300 includes processor 301, such as a CPU or MPU, memory 302, storage 303, communication interface 304, display device 305, and input device 306. Each of these components is connected to each other via a bus so that data can be communicated.

The processor 301 deploys various programs stored in the storage 303 in the memory 302 and executes them to realize various processes according to this embodiment.

Memory 302 is typically a volatile storage device such as a dynamic random access memory (DRAM), and stores programs and the like read from storage 303. Storage 303 is typically a non-volatile magnetic storage device such as a hard disk drive.

The communication interface 304 mediates data transmission between the processor 301 and external devices (including the controller 100). The communication interface 304 typically includes Ethernet (registered trademark) and USB (Universal Serial Bus).

The display device 305 is, for example, a liquid crystal display. The input device 306 includes, for example, a mouse, a keyboard, and a touch panel.

<Functional configuration>
Fig. 5 is a diagram showing an example of a functional configuration of a controller according to the present embodiment. As shown in Fig. 5, the controller 100 includes a memory unit 10, an acquisition unit 11, an assignment unit 12, a provision unit 13, a determination unit 14, and a control unit 15. The memory unit 10 is realized by a main memory 103 and a storage 104. The acquisition unit 11 is realized by a field bus controller 106 and a processor 101 that executes a user program 111. The assignment unit 12, the provision unit 13, the determination unit 14, and the control unit 15 are realized by the processor 101 executing the user program 111.

The acquisition unit 11 acquires time series data 2 of variables measured in the production system 1. The acquisition unit 11 stores the acquired time series data 2 in the memory unit 10. For example, the acquisition unit 11 acquires time series data 2 of temperature from the sensor 200b (see FIG. 2). Specifically, the acquisition unit 11 acquires data elements 3 indicating temperature from the sensor 200b according to a predetermined cycle. The acquisition unit 11 adds the data elements 3 associated with the measurement time to the time series data 2 stored in the memory unit 10.

The assigning unit 12 assigns numbers to each of the multiple data elements 3 included in the time series data 2. For each processing unit that is repeatedly performed in the production system 1, the assigning unit 12 assigns a reference number to the first data element 3a at the start timing of the processing unit among the multiple data elements 3. The reference number is, for example, "0". Furthermore, for each processing unit, the assigning unit 12 assigns consecutive numbers in chronological order to the reference number to one or more second data elements 3b among the multiple data elements 3 measured during the period in which the processing unit is performed. When the reference number is "0", a positive number is assigned to one or more second data elements 3b. For each processing unit, the assigning unit 12 assigns consecutive numbers in chronological order to the reference number to one or more third data elements 3c among the multiple data elements 3 measured during the period in which the processing unit before the processing unit is performed. When the reference number is "0", a negative number is assigned to one or more third data elements 3c.

The providing unit 13 extracts, for each processing unit, from the time series data 2, a data set including one or more third data elements 3c, a first data element 3a, and one or more second data elements 3b, which are assigned consecutive numbers according to the processing unit. In other words, the providing unit 13 extracts, from the time series data 2, a block of data elements 3 assigned consecutive numbers centered around a reference number as one data set.

The providing unit 13 provides provided data generated based on the extracted data set. The provided data indicates the change over time of variables or features calculated from the variables. The provided data is used, for example, to generate a screen including a graph indicating the change over time of variables or features. The providing unit 13 provides the provided data to the computer 300. This allows the user to operate the computer 300 to check the change over time of variables or features for each processing unit from a timing prior to the start timing of the processing unit.

The providing unit 13 may calculate the difference value disclosed in Patent Document 1 as a feature amount. Specifically, the providing unit 13 extracts multiple data sets from the time series data 2. For each number, the providing unit 13 extracts data elements 3 to which the number is assigned from the multiple data sets, and calculates the statistical value of the value indicated by the extracted data elements 3. For example, the providing unit 13 extracts data elements 3 to which the number "m" is assigned from the multiple data sets, and calculates the statistical value of the value indicated by the extracted data elements 3 as the "statistical value of number "m". The statistical value is, for example, the mean, median, standard deviation, maximum value, minimum value, kurtosis, skewness, etc. The providing unit 13 calculates the difference value between the value indicated by each data element 3 included in the data set and the statistical value of the number assigned to the data element 3 as a statistical amount.

Similar to the providing unit 13, the determining unit 14 extracts, for each processing unit, from the time series data 2, a data set including one or more third data elements 3c, a first data element 3a, and one or more second data elements 3b, which are assigned consecutive numbers according to the processing unit. Then, the determining unit 14 determines whether or not the extracted data set includes a data element 3 indicating an abnormality in the production system 1.

For example, the determination unit 14 calculates one or more features from each data set and determines the presence or absence of an abnormality based on the results of comparing each calculated feature with a threshold value.

As described above, the difference value disclosed in Patent Document 1 may be calculated as the feature amount. If the difference value is large, it is estimated that some abnormality has occurred in the production system 1. Therefore, if the data element 3 indicates a difference value from the statistical value that exceeds a threshold value, the determination unit 14 determines that the data set includes a data element 3 indicating an abnormality in the production system 1.

The control unit 15 performs control according to the judgment result of the judgment unit 14. For example, when the control unit 15 judges that the data set contains a data element 3 indicating an abnormality, it outputs an instruction to an alarm device included in the field device group 200 to issue an alarm. Alternatively, when the control unit 15 judges that the data set contains a data element 3 indicating an abnormality, it may output an instruction to one or more field devices included in the field device group 200 to stop operating.

<First method of numbering>
A first method of numbering the data elements 3 will now be described with reference to FIGS.

6 is a diagram showing the progress of processing executed in each processing unit. Each processing unit includes a first processing for maintaining the heating device 200a installed in the production system 1 in an ON state, and a second processing for maintaining the heating device 200a in an OFF state after the first processing. The period during which the processing unit is executed (for example, the period T _k-3 to T _k+1 shown in FIG. 6 ) includes a period Ta during which the heating device 200a is maintained in an ON state, and a period Tb during which the heating device 200a is maintained in an OFF state.

The start timing of each processing unit coincides with the start timing of the first processing. Therefore, the processor 101 operating as the assignment unit 12 determines the timing at which the heating device 200a changes from an OFF state to an ON state as the start timing of the processing unit. The processor 101 then determines the data element 3 of that start timing as the first data element 3a, and assigns a reference number to the first data element 3a.

Furthermore, the processor 101 operating as the assigning unit 12 determines, for each processing unit, each data element 3 measured during the period Ta during which the first processing included in the processing unit is performed as the second data element 3b. In addition, the processor 101 operating as the assigning unit 12 determines, for each processing unit, each data element 3 measured during the period Tb during which the second processing included in the processing unit preceding the processing unit is performed as the third data element 3c.

Figure 7 is a flowchart showing the process flow of the first numbering method. Figure 8 is a diagram showing an example of numbering according to the first method.

As shown in FIG. 8(a), the time series data 2 includes data elements 3 indicating the temperature measured by the sensor 200b and the state of the heating device 200a for each time. The state of the heating device 200a is indicated by data acquired from the heating device 200a. The time series data 2 shown in FIG. 8 has a table format. In the time series data 2, the lower the record, the more recent the temperature measurement time.

In step S1 shown in FIG. 7, the processor 101 determines whether or not there is an unselected data element 3 in the time series data 2. If there is no unselected data element 3 (NO in step S1), the process ends.

If there are unselected data elements 3 (YES in step S1), the processor 101 selects the oldest data element 3 among the unselected data elements 3 from the time series data 2 (step S2). Hereinafter, the data element 3 selected in step S2 is referred to as the "selected data element."

Next, the processor 101 determines whether the selected data element indicates an on state (step S3).

If the selected data element indicates the OFF state (NO in step S3), the process returns to step S1. As shown in FIG. 8(a), in the time series data 2, the data elements 3 in the first to fourth rows all indicate the OFF state. Therefore, steps S1 to S3 are repeated four times, and the data elements 3 in the first to fourth rows are selected without being assigned numbers.

If the selected data element indicates an ON state (YES in step S3), the processor 101 determines whether the data element 3 immediately preceding the selected data element indicates an OFF state (step S4).

If the previous data element 3 indicates an off state (YES in step S4), the processor 101 determines the selected data element as the first data element 3a and assigns a reference number (e.g., "0") to the selected data element (step S5).

As shown in FIG. 8B, the data element 3 in the fifth row indicates an ON state and the data element 3 in the fourth row indicates an OFF state, so when the data element 3 in the fifth row is selected in step S2, it is determined to be the first data element 3a. As a result, the reference number "0" is assigned to the data element 3 in the fifth row.

Next, the processor 101 sets i to 1 (step S6). The processor 101 determines the data element i before the selected data element as the third data element 3c, and assigns the number "reference number - i" to the data element i before the selected data element (step S7). For example, if the reference number is "0", the processor 101 assigns the number "-i" to the data element i before the selected data element.

Next, the processor 101 adds 1 to i (step S8). Next, the processor 101 determines whether the data element 3 i before the selected data element indicates an OFF state (step S9). If the data element 3 i before the selected data element indicates an OFF state (YES in step S9), the process returns to step S7. If the data element 3 i before the selected data element indicates an ON state (NO in step S9), the process returns to step S1.

As shown in FIG. 8(c), in the time series data 2, the data elements 3 in the first to fourth rows are measured during a period Tb during which the second process of the processing unit immediately preceding the processing unit being performed when the data element 3 in the fifth row was measured is being performed. During the period Tb, the heating device 200a is maintained in an off state. Therefore, when the data element 3 in the fifth row is selected in step S2, steps S5 to S8 are repeated four times, and the data elements 3 in the first to fourth rows are determined as the third data element 3c. The third data elements 3c in the first to fourth rows are then assigned consecutive numbers from "-4" to "-1" in chronological order so as to be consecutive to the reference number "0".

On the other hand, if the answer is NO in step S4 (i.e., the data element 3 immediately preceding the selected data element indicates an ON state), the processor 101 determines the selected data element as the second data element 3b, and assigns to the selected data element a number that is the number assigned to the immediately preceding data element 3 plus 1 (step S10). After step S10, the process returns to step S1.

If there are consecutive data elements 3 indicating the ON state, step S9 is repeated. Therefore, as shown in FIG. 8(d), the data elements 3 in the 6th to 10th rows measured during the period Ta in which the first process is being performed are determined to be the second data elements 3b. The data elements 3 in the 6th to 10th rows are then assigned consecutive numbers "1" to "5" in chronological order so as to be consecutive to the reference number "0".

Steps S1 to S10 are repeated until there are no unselected data elements 3. Therefore, as shown in FIG. 8(e), the data element 3 in the 14th row measured at the start timing of the first process of the next processing unit is determined as the first data element 3a, and the data element 3 in the 14th row is assigned the reference number "0". Thereafter, as shown in FIG. 8(f), the data elements 3 in the 11th to 13th rows measured during the period Tb in which the immediately preceding second process is being performed are determined as the third data element 3c. As a result, the data elements 3 in the 11th to 13th rows are assigned consecutive numbers so as to be continuous with the reference number assigned to the data element 3 in the 14th row. Furthermore, as shown in FIG. 8(g), the data elements 3 in the 15th to 17th rows measured during the period Ta in which the first process is being performed are determined as the second data element 3b. As a result, the data elements 3 in the 15th to 17th rows are assigned consecutive numbers so as to be continuous with the reference number assigned to the data element 3 in the 14th row.

Figure 9 is a diagram showing a data set for each processing unit extracted from the time series data shown in Figure 8. The data elements included in the data set indicate the temperature measured by sensor 200b. Figure 9 shows data set 4a corresponding to the first processing unit (batch) and data set 4b corresponding to the second processing unit (batch). Data set 4a includes 10 data elements numbered consecutively from "-4" to "5". Data set 4b includes 7 data elements numbered consecutively from "-3" to "3".

The length of the period during which a processing unit is performed is generally not constant. Therefore, the processor 101 may add data elements indicating missing values to the dataset according to the number of second data elements 3b and third data elements 3c so that the number of data elements included in each dataset is constant. In FIG. 9, missing values are represented as "NA".

<Second method of numbering>
A second method of numbering the data elements 3 will now be described with reference to FIGS.

The processor 101 operating as the assigning unit 12 determines, for each processing unit, the data elements 3 measured during the period from the start timing of the processing unit to the start timing of the processing unit following the processing unit as one or more second data elements 3b. Furthermore, for each processing unit, the processor 101 determines, for the processing unit, the data elements 3 measured during the period from the start timing of the processing unit preceding the processing unit to the start timing of the processing unit itself as one or more third data elements 3c.

The start timing of each processing unit is, for example, the timing of receiving a trigger indicating the start of the processing unit. For example, the controller 100 may receive the above trigger from a higher-level controller (not shown). Alternatively, the controller 100 may receive the above trigger from a field device included in the field device group 200. Alternatively, the trigger may be generated based on data received from the field device group 200. For example, the trigger is generated at the timing when the state indicated by the data received from the heating device 200a changes from an on state to an off state.

Figure 10 is a flowchart showing the process flow of the second numbering method. Figure 11 is a diagram showing an example of numbering according to the second method.

As shown in FIG. 11(a), the time series data 2 includes data elements 3 indicating the temperature measured by the sensor 200b and the batch ID that identifies the processing unit being performed at that timing, for each time. Each time the processor 101 acquires data indicating the temperature from the sensor 200b, it generates a data element 3 that combines the data with the batch ID set at the timing when the data was acquired, and adds the generated data element 3 to the time series data 2. The processor 101 updates the batch ID at the timing when it receives a trigger indicating the start of a processing unit. As a result, data elements 3 measured during the period when the same processing unit is being performed indicate the same batch ID. The time series data 2 shown in FIG. 11 has a table format. In the time series data 2, the lower the record, the newer the temperature measurement time.

In step S11 shown in FIG. 10, the processor 101 determines whether or not there is an unselected data element 3 in the time series data 2. If there is no unselected data element 3 (NO in step S11), the process ends.

If there are unselected data elements 3 (YES in step S11), the processor 101 selects the oldest data element 3 from the time series data 2 among the unselected data elements 3 (step S12). Hereinafter, the data element 3 selected in step S12 is referred to as the "selected data element."

In the example shown in FIG. 11(a), data elements 3 in rows 1 to 10 have already been selected, so in step S12, data element 3 in row 11 is selected. Note that numbers "0" to "9" have already been assigned to data elements 3 in rows 1 to 10, respectively.

Next, the processor 101 determines whether or not a trigger indicating the start of a processing unit was received at the time when the selected data element was measured (step S13). The processor 101 may determine that a trigger was received at the time when the selected data element was measured based on the difference between the batch ID indicated by the selected data element and the batch ID indicated by the data element 3 immediately preceding the selected data element.

If a trigger is received at the timing when the selected data element is measured (YES in step S13), the processor 101 determines the selected data element as the first data element 3a and assigns a reference number (e.g., "0") to the selected data element (step S14).

As shown in FIG. 11(b), the batch ID "2" indicated by the data element 3 on the 11th line is different from the batch ID "1" indicated by the data element 3 on the 10th line, so it is determined that a trigger was received at the timing when the data element 3 on the 11th line was measured. As a result, the data element 3 on the 11th line is determined to be the first data element 3a, and the reference number "0" is assigned to the data element 3 on the 11th line.

Next, the processor 101 sets i to 1 (step S15). The processor 101 determines the data element 3 that is i elements before the selected data element as the third data element 3c, and assigns the number "reference number - i" to the data element 3 that is i elements before the selected data element (step S16). For example, if the reference number is "0", the processor 101 assigns the number "-i" to the data element 3 that is i elements before the selected data element.

Next, the processor 101 adds 1 to i (step S17). Next, the processor 101 determines whether the batch ID indicated by the data element 3 that is i before the selected data element is the same as the batch ID indicated by the data element 3 that is one before the selected data element (step S18). If the batch IDs are the same (YES in step S18), the process returns to step S16. If the batch IDs are not the same (NO in step S18), the process returns to step S11.

As shown in FIG. 11(c), in the time series data 2, the data elements 3 in the first to tenth rows all indicate a batch ID of "1". Therefore, when the data element 3 in the eleventh row is selected in step S2, steps S15 to S18 are executed, and the data elements 3 in the first to tenth rows are determined as the third data element 3c. The data elements 3 in the first to tenth rows are then assigned consecutive numbers from "-10" to "-1" in chronological order so as to be continuous with the reference number "0". As mentioned above, the data elements 3 in the first to tenth rows have already been assigned the numbers "0" to "9", respectively. Therefore, two numbers (0 or a positive number and a negative number) are assigned to the data elements 3 in the first to tenth rows.

On the other hand, if the answer is NO in step S13 (i.e., the selected data element has not received a trigger at the timing when it was measured), the processor 101 determines the selected data element as the second data element 3b, and assigns to the selected data element a number that is the number assigned to the data element 3 immediately preceding the selected data element plus 1 (step S19). After step S19, the process returns to step S11.

If there are consecutive data elements 3 indicating the same batch ID, step S19 is repeated. Therefore, as shown in FIG. 11(d), when the data elements 3 in lines 12 to 19 indicating batch ID "2" are selected in order in step S12, the data elements 3 are determined as the second data element 3b in step S19. The data elements 3 in lines 12 to 19 are then assigned consecutive numbers "1" to "8" in chronological order so as to be consecutive to the reference number "0" assigned to the data element 3 in line 11.

Steps S11 to S19 are repeated until there are no unselected data elements 3. Therefore, as shown in FIG. 11(e), the data element 3 in the 20th row measured at the start timing of the next processing unit is determined as the first data element 3a, and the data element 3 in the 20th row is assigned the reference number "0". Thereafter, as shown in FIG. 11(f), the data elements 3 in the 11th to 19th rows are determined as the third data element 3c. As a result, the data elements 3 in the 11th to 19th rows are assigned consecutive numbers so as to be continuous with the reference number assigned to the data element 3 in the 20th row. Furthermore, as shown in FIG. 11(g), the data elements 3 in the 21st to 25th rows, which indicate the same batch ID as the data element 3 in the 20th row, are determined as the second data element 3b. As a result, the data elements 3 in the 21st to 25th rows are assigned consecutive numbers so as to be continuous with the reference number assigned to the data element 3 in the 20th row.

Figure 12 is a diagram showing a data set for each processing unit extracted from the time series data shown in Figure 11. The data elements included in the data set indicate the temperature measured by sensor 200b. Figure 12 shows data set 4a corresponding to the first processing unit (batch), data set 4b corresponding to the second processing unit (batch), and data set 4c corresponding to the third processing unit (batch). Data set 4a includes 10 data elements numbered consecutively from "0" to "9". Data set 4b includes 19 data elements numbered consecutively from "-10" to "8". Data set 4c includes 15 data elements numbered consecutively from "-9" to "5".

According to the second method, as shown in FIG. 11, two types of numbers can be assigned to each data element 3: a number equal to or greater than 0, and a negative number. Therefore, some of the data elements 3 included in dataset 4a are also included in dataset 4b. For example, the data elements 3 assigned the numbers "0" to "9" in dataset 4a are identical to the data elements 3 assigned the numbers "-10" to "-1" in dataset 4b, respectively.

The length of the period during which a processing unit is performed is generally not constant. Therefore, the processor 101 may add data elements indicating missing values to the dataset according to the number of second data elements 3b and third data elements 3c so that the number of data elements included in each dataset is constant. In FIG. 12, missing values are represented as "NA".

<Screen example>
Fig. 13 is a diagram showing an example of a screen generated using the provided data. The screen 6 shown in Fig. 13 is generated based on the provided data provided from the controller 100 to the computer 300, for example, and is displayed on the display device 305 of the computer 300.

Screen 6 includes graph 7 showing changes in feature amounts over time, generated based on data sets corresponding to each of the first to fifth processing units (batches). The horizontal axis of graph 7 represents numbers assigned to data elements 3 in chronological order. The vertical axis of graph 7 represents feature amounts. The feature amounts are difference values as disclosed in Patent Document 1.

On screen 6, the difference values of numbers "83" to "85" on graph 7 corresponding to the third processing unit and the difference values of numbers "-3" to "-1" on graph 7 corresponding to the fourth processing unit are significantly different from the difference values of the remaining numbers. Therefore, the user can notice an abnormality in production system 1 at the measured timing of data element 3 assigned numbers "-3" to "-1" in the data set corresponding to the fourth processing unit as a possible cause of the defect in the product produced by the fourth processing unit.

By assigning "0" as the reference number, the user can easily determine whether the abnormality occurred during the target processing unit or the previous processing unit by checking whether the feature number indicating the abnormality is positive or negative.

<Modification 1>
In the above description, the processor 101 includes data elements indicating missing values in the dataset according to the number of second data elements 3b and third data elements 3c so that the number of data elements included in each dataset is constant (processing (A)).

However, the processor 101 may perform the following process (B) or process (C) instead of process (A) on the data sets extracted for each processing unit so that the number of data elements contained in each data set is constant.

Process (B): The processor 101 identifies the minimum number M1 of second data elements 3b contained in multiple data sets extracted from the time series data 2. The processor 101 deletes from each data set the second data elements 3b numbered "M1+1" or greater. Similarly, the processor 101 identifies the minimum number M2 of third data elements 3c contained in multiple data sets extracted from the time series data 2. The processor 101 deletes from each data set the third data elements 3c numbered "-(M2+1)" or less.

Process (C): The processor 101 deletes from each data set any second data elements 3b that are assigned numbers equal to or greater than a predetermined number M3 (positive number). Similarly, the processor 101 deletes from each data set any third data elements 3c that are assigned numbers equal to or less than a predetermined number M4 (negative number). The numbers M3 and M4 are predetermined according to the time period that affects the quality of the product during the period during which the processing unit is performed.

Alternatively, the processor 101 may perform one of processes (A) to (C) that is specified in advance by the user.

<Modification 2>
The processor 101 operating as the assigning unit 12 may assign numbers by a method other than the above-mentioned first method and second method. For example, the processor 101 may determine a predetermined number (hereinafter referred to as the "designated number") of data elements 3 immediately preceding the first data element 3a as the third data element 3c. The designated number is, for example, a value (Tc/P) obtained by dividing a time Tc of a period immediately preceding the processing unit, which may affect the status of the production system 1 in the processing unit, by an acquisition period P of the variable value.

<Modification 3>
In the above description, an example in which "0" is assigned as the reference number has been described. However, the reference number is not limited to "0". For example, a positive value may be set in advance as the reference number.

For example, variant 2 and variant 3 may be combined, and the reference number may be set to the same value as the specified number. This causes the oldest third data element 3c of the one or more third data elements 3c to be assigned the number "0". As a result, the user can check the data set of the target processing unit and grasp the status of the production system 1 from the time Tc before the start timing of the processing unit.

<Modification 4>
In the above description, the controller 100 assigns numbers to each data element 3 of the time series data 2. However, the assignment of numbers to each data element 3 of the time series data 2 may be performed by the computer 300.

FIG. 14 is a diagram showing the functional configuration of a computer according to Modification 4. As shown in FIG. 14, computer 300A according to Modification 4 includes a memory unit 30, an acquisition unit 31, an assignment unit 32, a provision unit 33, and a determination unit 34. Computer 300A has the same hardware configuration as computer 300 shown in FIG. 4. Memory unit 30 is realized by memory 302 and storage 303. Acquisition unit 31 is realized by communication interface 304 and processor 301 that executes a program stored in storage 303. Assignment unit 32, provision unit 33, and determination unit 34 are realized by processor 301 executing a program stored in storage 303.

The acquisition unit 31 acquires the time series data 2 from the controller 100 and stores the acquired time series data 2 in the storage unit 30. The acquisition unit 31 may periodically acquire the time series data 2 from the controller 100. Alternatively, the acquisition unit 31 may acquire the time series data 2 from the controller 100 in response to a user operation.

The assignment unit 32 has the same function as the assignment unit 12, and assigns a number to each of the multiple data elements 3 contained in the time series data 2 stored in the memory unit 30.

The providing unit 33 has the same functions as the providing unit 13, extracts a data set from the time series data 2 for each processing unit, and provides provided data generated based on the data set. Specifically, the providing unit 33 generates a screen showing the time change of a variable or feature based on the provided data, and displays the generated screen on the display device 305.

The determination unit 34 has the same functions as the determination unit 14, and extracts a data set from the time series data 2 for each processing unit, and determines whether the extracted data set includes a data element 3 indicating an abnormality in the production system 1. Furthermore, the determination unit 34 may display the determination result on the display device 305.

§3 Supplementary Note As described above, the present embodiment includes the following disclosure.

(Configuration 1)
An apparatus (100, 300) for analyzing data relating to a production system (1), comprising:
an acquisition unit (11, 31, 101, 301) that acquires time series data (2) of variables measured in the production system (1);
an assigning unit (12, 32, 101, 301) that assigns a number to each of a plurality of data elements (3) included in the time series data (2);
The assignment unit (12, 32, 101, 301) performs, for each processing unit repeatedly performed in the production system (1),
assigning a reference number to a first data element (3a) at a start timing of the processing unit among the plurality of data elements (3);
assigning consecutive numbers in a time series to one or more second data elements (3b) among the plurality of data elements (3) measured during the period in which the processing unit is being performed, so as to be consecutive to the reference number;
The device (100, 300) assigns consecutive numbers in chronological order to the reference number for one or more third data elements (3c) among the plurality of data elements (3) measured during the period in which the processing unit preceding the processing unit in question is being performed.

(Configuration 2)
The reference number is 0,
the number assigned to each of the one or more second data elements (3b) is positive;
2. The apparatus (100, 300) of configuration 1, wherein the number assigned to each of the one or more third data elements (3c) is negative.

(Configuration 3)
The processing unit includes a first process for maintaining an equipment (200a) installed in the production system (1) in an ON state, and a second process for maintaining the equipment (200a) in an OFF state after the first process,
the start timing is a timing at which the device changes from an off state to an on state,
The assignment unit (12, 32, 101, 301) performs, for each processing unit,
determining a data element measured during a period during which the first process included in the processing unit is being performed as the one or more second data elements (3b);
The apparatus (100, 300) according to configuration 1 or 2, wherein a data element measured during a period during which the second processing included in a processing unit preceding the processing unit in question is performed is determined as the one or more third data elements (3c).

(Configuration 4)
The assignment unit (12, 32, 101, 301) performs, for each processing unit,
determining a data element measured during a period from the start timing of the processing unit to the start timing of the next processing unit of the processing unit as the one or more second data elements (3b);
The apparatus according to configuration 1 or 2, wherein a data element measured during a period from a start timing of a processing unit preceding the processing unit in question to a start timing of the processing unit in question is determined as the one or more third data elements (3c).

(Configuration 5)
The apparatus (100, 300) according to configuration 1 or 2, wherein the assignment unit (12, 32, 101, 301) determines, for each processing unit, a predetermined number of data elements preceding the first data element as the one or more third data elements (3c).

(Configuration 6)
6. The apparatus (100, 300) of configuration 5, wherein the reference number is the same as the predetermined number of values.

(Configuration 7)
a providing unit (13, 33, 101, 301) that extracts, for each processing unit, from the time-series data (2), a data set (4a-4c) including the one or more third data elements (3c), the first data element (3a), and the one or more second data elements (3b), to which a consecutive number is assigned according to the processing unit, and provides provided data generated based on the data sets (4a-4c);
The device (100, 300) according to any one of configurations 1 to 6, wherein the provided data indicates a time change of the variable or a feature calculated from the variable.

(Configuration 8)
The apparatus (100, 300) according to any one of configurations 1 to 7, further comprising a determination unit (14, 34, 101, 301) that extracts, for each processing unit, from the time-series data (2), a data set (4a to 4c) including the one or more third data elements (3c), the first data element (3a), and the one or more second data elements (3b), which are assigned consecutive numbers according to the processing unit, and determines whether or not the data set (4a to 4c) includes a data element indicating an abnormality in the production system (1).

(Configuration 9)
A method for analyzing data relating to a production system (1), comprising the steps of:
A step (S100) of acquiring time series data (2) of variables measured in the production system (1);
and a step (S200) of assigning a number to each of a plurality of data elements (3) included in the time series data (2),
The step of assigning (S200) is performed for each processing unit that is repeatedly performed in the production system.
A step (S201) of assigning a reference number to a first data element (3a) at a start timing of the processing unit among the plurality of data elements (3);
a step (S202) of assigning consecutive numbers in a time series to one or more second data elements (3b) among the plurality of data elements (3) measured during the period in which the processing unit is being performed, so as to be consecutive to the reference number;
and a step (S203) of assigning consecutive numbers in chronological order to the reference number for one or more third data elements (3c) among the plurality of data elements (3) measured during a period in which a processing unit preceding the processing unit in question is being performed.

(Configuration 10)
A program for analyzing data relating to a production system (1), comprising:
A step (S100) of acquiring time series data (2) of variables measured in the production system (1);
and a step (S200) of assigning a number to each of a plurality of data elements (3) included in the time series data (2).
The step of assigning (S200) is performed for each processing unit that is repeatedly performed in the production system.
A step (S201) of assigning a reference number to a first data element (3a) at a start timing of the processing unit among the plurality of data elements (3);
a step (S202) of assigning consecutive numbers in a time series to one or more second data elements (3b) among the plurality of data elements (3) measured during the period in which the processing unit is being performed, so as to be consecutive to the reference number;
and a step (S203) of assigning consecutive numbers in a chronological order to the reference number for one or more third data elements (3c) among the plurality of data elements (3) measured during a period in which a processing unit preceding the processing unit in question is being performed.

Although the embodiments of the present invention have been described, the embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present invention is defined by the claims, and it is intended to include all modifications within the meaning and scope of the claims.

1 Production system, 2 Time series data, 3 Data element, 3a First data element, 3b Second data element, 3c Third data element, 4a to 4c Data set, 5 Field bus, 6 Screen, 7 Graph, 10, 30 Memory unit, 11, 31 Acquisition unit, 12, 32 Assignment unit, 13, 33 Provision unit, 14, 34 Determination unit, 15 Control unit, 100 Controller, 101, 301 Processor, 102 Chip set, 103 Main memory, 104, 303 Storage, 105 Upper network controller, 106 Field bus controller, 107 USB controller, 108 Memory card interface, 110 System program, 111 User program, 120 Memory card, 200 Field device group, 200a Heating device, 200b Sensor, 300, 300A Computer, 302 Memory, 304 Communication interface, 305 Display device, 306 input device.

Claims (10)

An apparatus for analyzing data relating to a production system, comprising:
an acquisition unit that acquires time series data of variables measured in the production system;
an assigning unit that assigns a number to each of a plurality of data elements included in the time series data;
The application unit, for each processing unit that is repeatedly performed in the production system,
assigning a reference number to a first data element at a start timing of the processing unit among the plurality of data elements;
assigning consecutive numbers to one or more second data elements among the plurality of data elements measured during the period in which the processing unit is performed, in a time series consecutive to the reference number;
an apparatus for assigning sequential numbers in a chronological order to one or more third data elements among the plurality of data elements, the sequential numbers being consecutive with the reference number, the third data elements being measured during a period in which a processing unit preceding the processing unit in question is being performed. The reference number is 0,
the number assigned to each of the one or more second data elements is positive;
2. The apparatus of claim 1, wherein the number assigned to each of the one or more third data elements is negative. the processing unit includes a first process for maintaining an equipment installed in the production system in an ON state, and a second process for maintaining the equipment in an OFF state after the first process,
the start timing is a timing at which the device changes from an off state to an on state,
The assignment unit, for each processing unit,
determining a data element measured during a period during which the first process included in the processing unit is being performed as the one or more second data elements;
The apparatus according to claim 1 , further comprising: determining, as the one or more third data elements, a data element measured during a period during which the second process included in a processing unit preceding the current processing unit is being performed. The assignment unit, for each processing unit,
determining, as the one or more second data elements, data elements measured during a period from a start timing of the processing unit to a start timing of a processing unit following the processing unit;
The apparatus according to claim 1 , further comprising: determining, as the one or more third data elements, a data element measured during a period from a start timing of a processing unit preceding the processing unit to a start timing of the processing unit itself. The device according to claim 1, wherein the assigning unit determines, for each processing unit, a predetermined number of data elements preceding the first data element as the one or more third data elements. The device of claim 5, wherein the reference number is equal to the predetermined number of values. a providing unit that extracts, for each processing unit, a data set including the one or more third data elements, the first data element, and the one or more second data elements, to which a consecutive number is assigned according to the processing unit, from the time-series data, and provides provided data generated based on the data set;
The device according to claim 1 , wherein the provided data indicates a time change of the variable or a feature amount calculated from the variable. The device according to any one of claims 1 to 6, further comprising a determination unit that extracts, for each processing unit, from the time-series data, a data set including the one or more third data elements, the first data element, and the one or more second data elements, which are assigned consecutive numbers according to the processing unit, and determines whether the data set includes a data element indicating an abnormality in the production system. 1. A method for analyzing data relating to a production system, comprising:
acquiring time series data of variables measured in the production system;
and assigning a number to each of a plurality of data elements included in the time series data;
The step of assigning includes the steps of:
assigning a reference number to a first data element at a start timing of the processing unit among the plurality of data elements;
assigning consecutive numbers to one or more second data elements among the plurality of data elements measured during a period during which the processing unit is being performed, in a time series manner so as to be consecutive to the reference number;
and assigning sequential numbers in chronological order to one or more third data elements among the plurality of data elements, the sequential numbers being consecutive with the reference number, the third data elements being measured during a period in which a processing unit preceding the processing unit in question is being performed. A program for analyzing data relating to a production system, comprising:
acquiring time series data of variables measured in the production system;
and assigning a number to each of a plurality of data elements included in the time series data.
The step of assigning includes the steps of:
assigning a reference number to a first data element at a start timing of the processing unit among the plurality of data elements;
assigning consecutive numbers to one or more second data elements among the plurality of data elements measured during a period during which the processing unit is being performed, in a time series manner so as to be consecutive to the reference number;
and assigning consecutive numbers in a chronological order to one or more third data elements among the plurality of data elements, the consecutive numbers being consecutive with the reference number, the third data elements being measured during a period in which a processing unit preceding the processing unit in question is being performed.

Images