WO2020261364A1 - Program and information processing method - Google Patents

Abstract

The present invention assists in narrowing down to a measured value group that indicates a relationship that is highly reproducible.　An item selection reception unit 102 receives selection of an analysis item group and a narrowing item group. From data sets for a plurality of periods, a data set extraction unit 103 extracts a data set group for periods that include a limited range of measured values for one or more items from the narrowing item group. From a set of measured values from the analysis item group, a relational expression calculation unit 105 calculates a relational expression for measured values for each item. On the basis of the calculated relational expression, a condition satisfaction determination unit 104 determines data sets for which the number of post-extraction data sets and the statistical variability of measured values from the analysis item group included in the post-extraction data sets satisfy narrowing conditions. A completion determination unit 106 determines whether completion conditions for repeated extraction processing and narrowing conditions determination processing have been met.

Description

Program and information processing method

The present invention relates to a technique for supporting the analysis of the tendency indicated by the measured value group.

As a technique for supporting the analysis of the tendency indicated by the measured value group, Patent Document 1 describes the propulsion of the ship under a predetermined meteorological environment based on the data group recorded when the ship is operated in the actual sea area. A technique for calculating an estimated value of performance and correcting the estimated value calculated based on a measured value of sea weather is disclosed.

JP-A-2018-34585

In order to evaluate the performance of vehicles and machines, the tendency shown by the measured value group is analyzed from the relationship shown by the measured values of each item. For example, fuel efficiency is analyzed from the relationship between ship speed and fuel consumption. The measured value group used at that time is a mixture of a plurality of measured value groups (for example, a measured value group at the time of strong wind and a measured value group at the time of light wind) showing different relationships because the situation at the time of measurement is different. In some cases. In that case, as a relationship with higher practicality, it is desirable to narrow down to a group of measured values showing a relationship with higher reproducibility before analysis.
Therefore, an object of the present invention is to support narrowing down to a group of measured values showing a relationship with higher reproducibility.

In order to achieve the above object, the present invention has a plurality of periods when the computer uses one data set of the measured value groups of the first item group and the second item group at the same time among the plurality of items. Provide a program for executing a process of extracting a data set at a time including a limited range of measured values of one or more items of the second item group from the data set of.

In addition, the computer is subjected to a process of determining whether or not the number of the data sets after extraction and the statistical variation of the measured values of the first item group included in the data set group after extraction satisfy a predetermined condition. You may let it run.

Further, the condition may be satisfied when the ratio of the reduction amount of the statistical variation to the reduction amount of the number of the data sets before and after the extraction is larger than a predetermined standard.

Further, the computer may be made to execute the extraction process and the determination process for the extracted data set group determined to satisfy the above conditions.

Further, the computer may change the second item group before and after extraction.

Further, until at least one of the statistical variability, the number of the data sets, the reduction amount of the statistical variability before and after extraction, or the reduction amount of the number of data sets before and after extraction satisfies the termination condition on the computer. The extraction process and the determination process may be executed.

Further, the computer is made to execute a process of calculating the relational expression between the measured values of each item from the set of the measured values of the first item group, and the statistical expression represented by the difference from the calculated relational expression is obtained. A process of determining whether or not the above conditions are satisfied may be executed based on the variation.

Further, the computer is made to execute a process of calculating the relational expression between the measured values of each item from the set of the measured values of the second item group, and the statistical variation represented by the difference from the calculated relational expression. May execute a process of extracting a data set group at a time including a measured value that satisfies a predetermined condition.

Further, the computer may be made to execute the process of performing the extraction by limiting the range so that the amount of decrease in the number of the data sets becomes a predetermined amount before and after the extraction.

Further, the computer may be made to execute the process of performing the extraction by limiting the range so that the amount of reduction of the statistical variation becomes a predetermined amount before and after the extraction.

Further, the computer may be made to execute the extraction process with the predetermined amount as the amount corresponding to the number of extractions up to the previous time.

Further, in the present invention, when the information processing apparatus uses a data set of the first item group and the second item group of the first item group and the second item group at the same time as one data set, the data set group of a plurality of times is used. To provide an information processing method comprising a step of extracting a data set group at a time including a limited range of measured values of one or more items of the second item group.

According to the present invention, it is possible to support narrowing down to a group of measured values showing a relationship with higher reproducibility.

The figure which shows an example of the hardware configuration of the performance analysis apparatus which concerns on Example. Diagram showing the functional configuration realized by the performance analyzer Diagram showing an example of accumulated measured values Diagram showing an example of the item selection screen Diagram showing an example of stored datasets Diagram showing an example of the calculated relational expression The figure which shows an example of the operation procedure of each apparatus in the extraction process Diagram showing an example of statistical variability and changes in the number of datasets Diagram showing an example of utilization in data set performance analysis Diagram showing an example of utilization in data set performance analysis Diagram showing an example of performance analysis for comparison Diagram showing an example of statistical variability of variants and changes in the number of datasets Diagram showing an example of statistical variability of variants and changes in the number of datasets

[1] Example FIG. 1 shows an example of the hardware configuration of the performance analysis device 10 according to the embodiment. The performance analysis device 10 is a device that performs various processes for analyzing the performance of a vehicle, a machine, or the like. In this embodiment, the performance analysis device 10 performs various processes for analyzing the performance of the ship. The performance analysis device 10 is a computer including a processor 11, a memory 12, a storage 13, a communication device 14, an input device 15, an output device 16, and a bus 17.

The processor 11 operates, for example, an operating system to control the entire computer. The processor 11 includes a central processing unit (CPU: Central Processing Unit) including an interface with peripheral devices, a control device, an arithmetic unit, and registers. The processor 11 reads a program, data, and the like from at least one of the storage 13 and the communication device 14 into the memory 12, and executes various processes according to the read program and the like.

The memory 12 is a computer-readable recording medium such as a ROM (Read Only Memory) and a RAM (Random Access Memory). The storage 13 is a recording medium that can be read by a computer such as a hard disk drive. The communication device 14 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network.

The input device 15 is an input device (for example, a switch, a button, a sensor, etc.) that receives an input from the outside. The input device 15 includes an imaging device (such as a digital camera) for taking a picture. The output device 16 is an output device (for example, a display, a speaker, an LED lamp, etc.) that outputs to the outside.

In the performance analysis device 10, the input device 15 and the output device 16 include an integrated configuration (a touch screen in which a display and a touch panel are integrated). Each device such as the processor 11 and the memory 12 is connected by a bus 17 for communicating information. The bus 17 may be configured by using a single bus, or may be configured by using a different bus for each device.

For each function in the performance analysis device 10, the processor 11 performs calculations by loading predetermined software (program) on hardware such as the processor 11 and the memory 12, and controls the communication by the communication device 14 or the memory. It is realized by controlling at least one of reading and writing of data in the storage 13 and the storage 13.

FIG. 2 shows the functional configuration realized by the performance analysis device 10. The performance analysis device 10 includes a measurement value storage unit 101, an item selection reception unit 102, a data set extraction unit 103, a condition satisfaction determination unit 104, a relational expression calculation unit 105, and an end determination unit 106. The measured value accumulating unit 101 accumulates measured values in a plurality of items related to the performance analysis target. In this embodiment, the measured value accumulating unit 101 accumulates the measured values measured at the time of operation of the ship to be analyzed.

Specifically, the measured value accumulating unit 101 includes ship speed (water speed, ground speed), fuel consumption, rotation speed (engine rotation speed), displacement, trim, change in ship speed, wind speed, wind direction, wave height, tidal current, Accumulate the measured values of each item such as pitch, roll, and steering angle. Each item is measured by a sensor provided on the ship, a crew member of the ship, or an external system (weather forecast system, etc.) and input to the performance analysis device 10. The input of the measured value to the performance analysis device 10 may be performed mechanically via the communication device 14 or may be performed manually via the input device 15.

The measured value accumulating unit 101 accumulates measured values for a plurality of periods of each item.
FIG. 3 shows an example of the accumulated measured values. In the example of FIG. 3, the measured value accumulating unit 101 accumulates the measured values of each measurement item at the measurement time t1, t2, t3, t4, t5, ... In association with the name of the ship on which the measurement was performed. doing. In the example of FIG. 3, "-t1" or the like added to the end of the measured value indicates the time of measurement.

As the plurality of times, for example, the measurement time of the item measured at the shortest time interval (hereinafter referred to as "shortest measurement interval") among each item is used. For example, if the shortest measurement interval is every 1 minute, the measured value accumulating unit 101 accumulates the measured values of each item every 1 minute. The measured value accumulating unit 101 interpolates (so-called interpolation. Used when there are measured values before and after) or extrapolated (used when there are measured values before and after) for items that are not measured at the shortest measurement interval. (Used when there are measured values only in front or behind) and accumulate.

The measured value storage unit 101 interpolates, for example, the average value of the measured values before and after the unmeasured measurement time as the measured value of the measurement time. Further, the measured value accumulating unit 101 calculates an approximate expression indicating the time-series change of the measured value, and even if the calculated approximate expression indicates the measured value for each measurement period as the measured value of each measurement period, it may be interpolated or extrapolated. Good. In addition, some items are measured only once a day (for example, displacement) because they rarely fluctuate during navigation. In that case, the measured value storage unit 101 extrapolates the measured value as the measured value at each measurement time on the measurement day.

Further, for example, when the shortest measurement interval is every 10 minutes and the measurement interval of a certain two items is the same every 10 minutes but the measurement time is shifted by several minutes, the measured value accumulating unit 101 is shifted by several minutes. The measured values may be accumulated even at the same measurement time as the measurement at the same time. That is, the same time period does not have to be exactly the same measurement time, and a slight deviation may be allowed. The measured value accumulating unit 101 does not allow the deviation of the measured time, and for example, even if the measured value at the same measurement time is obtained by using the above-mentioned interpolation or extrapolation method and accumulated as the measured value at the same time. Good.

The item selection reception unit 102 accepts selection of an analysis item group and a narrowing item group from a plurality of items related to the performance analysis target. The analysis item group is two or more items, and is an item used for performance analysis by finding the relationship between the measured values of each item at the same time. The narrowing item group is two or more items and is an item used to narrow down the measured values of the analysis item group to the measured values that appropriately represent the relationship between the measured values described above. The analysis item group is an example of the "first item group" of the present invention, and the narrowing item group is an example of the "second item group" of the present invention.

The item selection reception unit 102 displays an item selection screen for accepting an operation for selecting an analysis item group and a narrowing item group.
FIG. 4 shows an example of the item selection screen. In the example of FIG. 4, the item selection reception unit 102 inputs an input field C1 for inputting a ship to be subjected to performance analysis, an input field C2 for inputting an item to be included in the analysis item group, and an item to be included in the narrowing item group. An item selection screen A1 including an input field C3 for inputting, an input field C4 for inputting a period in which the measured value used for analysis is measured, and an analysis start button B1 is displayed.

Input to each input field is performed by a user of the performance analysis device 10 (for example, a person in charge of performance analysis). For input to each input field, any of a method of typing text, a method of selecting from a pull-down list, a method of dragging from a list of items and the like may be used. In addition, a method of displaying a list of target ships and candidate items and checking the selection may be used. The item selection reception unit 102 selects each input item (ship name, analysis item group, narrowing item group, and period) by pressing the analysis start button B1 with the necessary items input in each input field. Accept as an operation to do.

When the item selection reception unit 102 receives the above selection operation, the item selection reception unit 102 notifies the data set extraction unit 103 of each selected item. The data set extraction unit 103 selects one or more of the narrowed-down item groups from the data sets of a plurality of time periods when the selected analysis item group and the narrowed-down item group at the same time are set as one data set. Extract a time data set that contains measurements in a limited range of items.

First, the data set extraction unit 103 is notified by the measurement value storage unit 101 that the measured values associated with the notified ship names are the measured values of the notified analysis item group and the narrowed-down item group. Read out the measured value that includes the measurement time in the period. The data set extraction unit 103 temporarily stores the read measured values in the memory 12 with the measured value groups at the same time as one data set and the data sets at a plurality of measurement times.

FIG. 5 shows an example of a stored data set group. In the example of FIG. 5, the measurement value group (“LOG-t1” and “FOC-t1”) of the analysis item group (water speed and fuel consumption) measured at the measurement time t1 and the narrowing item group (wind speed). And the data set DS-t1 including the measured value group (“Wind-t1”, “WD-t1”, etc.) of (wind direction, etc.) is stored in the memory 12 as a data set of the measurement time t1.

Similarly, data sets such as measurement timings t2, t3, t4 and t5 DS-t2, DS-t3, DS-t4 and DS-t5 data sets are stored in the memory 12. The area in which the data set group shown in FIG. 5 is stored is hereinafter referred to as a “pre-extraction area”. Further, the data set group first stored in the pre-extraction area (the data set group first extracted by the data set extraction unit 103) is referred to as an "initial data set group".

The data set extraction unit 103 extracts a data set group included in a range in which the measured values of the narrowed-down item group are limited from the initial data set group stored in the pre-extraction area, and is separated from the pre-extraction area. It is stored in the "extracted area" which is the area of. Further, in the present embodiment, the data set extraction unit 103 performs extraction by limiting the range so that the amount of decrease in the number of data sets becomes a predetermined amount before and after the extraction.

In the following, the case where the range of the wind speed is limited will be described by taking as an example the case where the amount of reduction when the number of data sets is reduced to 90% of the number of data sets before extraction is set as a predetermined amount. First, the data set extraction unit 103 limits the measured value of the wind speed to a range less than a predetermined ratio of the maximum wind speed. For example, when limited to the range of less than 90% of the maximum wind speed, the data set extraction unit 103 excludes the data set including the measured value of the wind speed of 90% or more of the maximum wind speed, and measures the wind speed of less than 90% of the maximum wind speed. Temporarily extract a set of data containing values.

In the data set extraction unit 103, if the amount of decrease in the extracted data set does not reach a predetermined amount (the number of excluded data sets is the amount to be excluded (in the above example, 10% of the number of data sets before extraction). If the number of data sets) is not reached, the data set group excluding the data set containing the measured value of 80% or more of the maximum wind velocity is tentatively extracted. The data set extraction unit 103 excludes the data set and tentatively extracts the data set. The extraction is repeated, and when the amount of decrease in the number of data sets reaches a predetermined amount (when the amount decreases by 10% or more), the tentatively extracted data set group is regarded as the officially extracted data set group, and the extraction is completed.

The method of limiting the range of measured values of the narrowed down items is not limited to the above method. For example, the data set extraction unit 103 may repeat limiting the range at an arbitrary ratio such as 1% instead of 10% increments. Further, the data set extraction unit 103 re-excludes at 85% and reduces the amount of decrease in the number of data sets by 10, for example, when the number of data sets greatly exceeds 10% when the exclusion is performed at 80%. You may try to get closer to%.

Further, the data set extraction unit 103 may use a predetermined amount represented by a numerical value (for example, the number of N (N is a natural number) data set) instead of a ratio. The data set extraction unit 103 extracts the data set group by limiting the range of the measured values of the narrowed-down item group other than the wind speed as in the case of the wind speed. The data set extraction unit 103 prepares a post-extraction area for each item of the narrowed-down item group, and stores the data set group extracted for each item.

The condition satisfaction determination unit 104 determines whether or not the statistical variation in the number of data sets after extraction and the measured values of the analysis item group included in the data set group after extraction satisfies the narrowing condition. The narrowing-down condition is a condition defined for narrowing down the data set group, and is an example of the "predetermined condition" of the present invention. Statistical variation is a measure of variation in a data group (measured value group in the present invention), and the smaller the variation, the clearer the tendency of the data group, which makes it easier to use for performance analysis. The details of the statistical variability will be explained later.

In this embodiment, the condition satisfaction determination unit 104 determines that the ratio of the reduction amount of statistical variation to the reduction amount of the number of data sets before and after extraction (hereinafter referred to as “reduction amount ratio”) is larger than a predetermined standard. In addition, it is judged that the narrowing conditions are satisfied. The condition satisfaction determination unit 104 first narrows down the value obtained by subtracting the number of data sets stored in the post-extraction area from the number of data sets stored in the pre-extraction area as the amount of decrease in the number of data sets. Calculate for each item in the included item group.

In addition, the condition satisfaction determination unit 104 calculates the amount of reduction in statistical variation for each item in the narrowed-down item group, so that the relational expression between the measured values of the analysis item group based on the data set group before and after extraction is used. The calculation is instructed to the relational expression calculation unit 105. Upon receiving the instruction, the relational expression calculation unit 105 refers to the pre-extraction area and the post-extraction area of each item of the narrowed-down item group, and sets each item from the set of measured values of the analysis item group stored in each area. Calculate the relational expression between the measured values.

The relational expression calculation unit 105 calculates the relational expression by using, for example, a curve fitting (curve fitting) method. The curve fitting method is, for example, regression analysis such as nonlinear least squares method or nonlinear minimum absolute value method (either normal regression, principal component regression, geometric mean reversion, etc.). As the curve used for curve fitting, for example, y = ax ^ 3, which is a curve based on the square-cube law, or y = ax ^ b, which is a more general-purpose curve, may be used. May be good.

Further, as the curve used for curve fitting, a polynomial may be used, or a curve having an intercept may be used. Further, the user may be able to select the curve to be used for curve fitting. In that case, an input field for selecting a curve to be used for curve fitting may be provided on the item selection screen shown in FIG. In the example of FIG. 5, the relational expression calculation unit 105 calculates a relational expression representing the relationship between the measured value of the water speed and the measured value of the fuel consumption.

FIG. 6 shows an example of the calculated relational expression. In FIG. 6, the horizontal axis represents the speed against water, and the vertical axis represents the graph G1 showing the fuel consumption. The graph G1 shows a point corresponding to the measured value of the water speed and the measured value of the fuel consumption of each data set shown in FIG. 5, and a curve D1 showing the relational expression calculated by the relational expression calculation unit 105. ing. The relational expression calculation unit 105 calculates the relational expression for each item of the narrowed-down item group for the data set before extraction and the data set after extraction, and notifies the condition satisfaction determination unit 104 of the calculated relational expression.

The condition satisfaction determination unit 104 calculates the amount of reduction in the statistical variation of the measured values of the analysis item group before and after the extraction using the notified relational expression. Specifically, the condition satisfaction determination unit 104 determines the difference in the Y-axis direction (the difference in the X-axis direction, the difference in the shortest distance, etc.) between the measured value of the analysis item group and the curve indicated by the calculated relational expression. The variance (which may be the mean value or the standard deviation) is calculated before and after the extraction as an index of the statistical variation, and the value obtained by subtracting the index after the extraction from the index before the extraction is calculated as the amount of decrease in the statistical variation.

The condition satisfaction determination unit 104 calculates the value obtained by dividing the reduction amount of the statistical variation calculated as described above by the reduction amount of the number of data sets as the above-mentioned reduction amount ratio. The condition satisfaction determination unit 104 similarly calculates the reduction amount rate for each item in the narrowed-down item group. Then, the condition satisfaction determination unit 104 determines that the narrowing-down condition is satisfied when the calculated reduction amount ratio is larger than the above-mentioned predetermined standard.

In this embodiment, the condition satisfaction determination unit 104 reduces the reduction rate of the data set group of the item having the largest calculated reduction rate among each item of the narrowed-down item group, which is larger than the predetermined standard described above. Assuming that it is a quantity ratio, it is judged that the data set group satisfies the narrowing condition. That is, in this embodiment, the maximum reduction amount ratio among the reduction amount ratios calculated for each item of the narrowed-down item group excluding the own item is treated as a predetermined standard.

As described above, in this embodiment, the condition satisfaction determination unit 104 determines whether or not the narrowing condition is satisfied based on the statistical variation represented by the difference from the relational expression calculated by the relational expression calculation unit 105. To do. When the condition satisfaction determination unit 104 determines a data set group that satisfies the narrowing down condition, the condition satisfaction determination unit 104 notifies the data set extraction unit 103 of the extracted area in which the data set group is stored.

The data set extraction unit 103 performs the above-mentioned extraction process with respect to the data set group stored in the notified post-extraction area, that is, the post-extraction data set group determined to satisfy the narrowing condition. The data set extraction unit 103 treats the notified post-extraction area as a new pre-extraction data set group by changing it to a pre-extraction area, for example.

The data set extraction unit 103 may release the original pre-extraction area and the post-extraction area in which the data set group that does not satisfy the narrowing conditions is stored, and delete the data set group. , A flag indicating that the area is not referenced in the new extraction process may be stored so as not to be erroneously referenced. When the data set extraction unit 103 performs the extraction process using the new data set group before extraction, the condition satisfaction determination unit 104 determines the satisfaction of the narrowing conditions for the re-extracted data set group.

As described above, the extraction process and the narrowing-down condition judgment process are repeated, so that multiple data sets are gradually reduced. The end determination unit 106 determines whether or not the end condition of the repeated extraction process and the determination process of the narrowing condition is satisfied. In this embodiment, the condition satisfaction determination unit 104 notifies the end determination unit 106 of the number of data sets determined to satisfy the narrowing condition.

Then, the end determination unit 106 determines that the end condition is satisfied when the number of notified data sets becomes less than the threshold value. When the end determination unit 106 determines that the end condition is satisfied, it notifies the condition satisfaction determination unit 104 to that effect (end notification). Upon receiving the end notification, the condition satisfaction determination unit 104 stops the notification of the extracted area to the data set extraction unit 103. As described above, the data set extraction unit 103 and the condition satisfaction determination unit 104 repeat the extraction process and the narrowing-down condition determination process until the end condition is satisfied.

The performance analysis device 10 performs an extraction process for extracting a data set group based on the above configuration.
FIG. 7 shows an example of the operation procedure of each device in the extraction process. The operation procedure of FIG. 7 is started, for example, when the person in charge of performance analysis displays the item selection screen shown in FIG. First, the performance analysis device 10 (item selection reception unit 102) receives selection of the analysis item group and the narrowing item group (step S11).

Next, the performance analysis device 10 (data set extraction unit 103) extracts a data set group of a period including a limited range of measured values of one or more items of the narrowing item group from the data sets of a plurality of periods. (Step S12). Subsequently, the performance analysis device 10 determines whether or not there is an item that is not used (unused) for extracting the data set group in the items of the narrowing item group (step S13).

If it is determined in step S13 that there is an unused item (YES), the performance analysis device 10 returns to step S12 and repeats the operation, and if it is determined that there is no unused item (NO), the performance analysis device 10 proceeds to step S14. .. The performance analysis device 10 (condition satisfaction determination unit 104) determines a data set in which the number of data sets after extraction and the statistical variation of the measured values of the analysis item group included in the data set after extraction satisfy the narrowing condition. (Step S14).

Subsequently, the performance analysis device 10 (end determination unit 106) determines whether or not the end conditions of the repeated extraction process and the narrowing-down condition determination process are satisfied (step S15). If it is determined in step S15 that the performance analysis device 10 is not satisfied (NO), the performance analysis device 10 returns to step S12 and repeats the operation, and if it is determined that the operation is satisfied (YES), the performance analysis device 10 ends this operation procedure.

FIG. 8 shows an example of statistical variability and transition of the number of data sets. In FIG. 8, the vertical axis represents the statistical variation and the horizontal axis represents the number of data sets. FIG. 8 shows a point E0 showing the relationship between the statistical variation of the initial data set group and the number of data sets. In addition, points E1, E2, E3, and E4 indicating the above-mentioned relationship of the four data set groups extracted in the first extraction process are represented.

Points E1, E2, E3, and E4 represent, for example, trim, wave height, tidal current, and wind speed of the narrowing down item group. In this embodiment, the number of data sets is reduced by ΔN each time the extraction process is performed. In the first extraction, the amount of decrease in statistical variation is in the order of trim <wave height <tide flow <wind speed. Therefore, the data set group extracted by limiting the range of the wind speed satisfies the narrowing condition and is used as the data set group before the next extraction.

In the second extraction, the amount of decrease in statistical variation is in the order of wave height <trim <wind speed <tide flow. Therefore, the data set group extracted by limiting the range of the tidal current satisfies the narrowing conditions and is used as the data set group before the next extraction. In the third extraction, the amount of decrease in statistical variation is in the order of tidal current <wind speed <trim <wave height. Therefore, the data set group extracted by limiting the range of wave height satisfies the narrowing condition.

FIG. 8 shows the threshold value Th1 used as the end condition. When the third extraction is performed, the number of data sets becomes less than the threshold Th1. Therefore, the extraction of the data set group is completed, and the data set group extracted by limiting the wave height range in the third extraction is the final. It is stored in the extracted area as a group of data sets extracted. The data set extraction unit 103 outputs the finally extracted data set group to a predetermined output destination.

The predetermined output destination is, for example, the display of the performance analysis device 10, the storage 13, an external device, an external recording medium, or the like. The output data set group is used for the performance analysis of the ship in this embodiment.
9 and 10 show examples of utilization in the performance analysis of the data set. In FIGS. 9 and 10, the vertical axis represents the speed against water and the horizontal axis represents the moon.

In the examples of FIGS. 9 and 10, the values randomly selected from the measured values of the water speed measured at the same time as the predetermined fuel consumption in the data set group of each month are shown. The faster the water speed, the higher the fuel efficiency. If the fuel consumption is limited to a predetermined fuel consumption, and the measured value of the water speed measured at the same time becomes too small, the water consumption measured at the same time as the fuel consumption in the predetermined "range". The measured value of speed may be represented in a graph.

Further, in order to increase the measured value of the water speed shown in the graph, a set of the measured value of the fuel consumption and the measured value of the water speed measured at the same time is used as a predetermined fuel by using the calculated relational expression. It may be corrected to the consumption amount and the measured value of the water speed which is a set thereof. FIG. 9 shows the water speed in the initial data set group F1, and FIG. 10 shows the water speed in the data set group F2 after the extraction process is performed.

In the examples of FIGS. 9 and 10, maintenance was performed in April, and the water speed improved before and after that. This indicates that maintenance has reduced resistance during propulsion and improved fuel efficiency. In the example of FIG. 9, since the variation in the speed against water is large, the tendency of the fuel consumption to decrease cannot be determined for about one year after the maintenance, and the tendency of the fuel consumption to decrease slightly appears in the second year.

On the other hand, in the example of FIG. 10, since the data set group F2 in which the statistical variation is small is extracted, the tendency shown by the data group (measured value group) becomes clearer, and the tendency of decreasing fuel consumption can be discriminated from the first year. It has become. As described above, in this embodiment, since the data set group in which the statistical variation is small is extracted, the data group narrowed down (measurement of the analysis item group) is compared with the case where the extraction is performed regardless of the statistical variation. The tendency indicated by the value group) can be clarified more.

Statistical variability tends to be smaller as the number of datasets is reduced, but if the number of datasets is reduced too much, another problem may occur.
FIG. 11 shows an example of performance analysis for comparison. In FIG. 11, a graph showing the water speed when a smaller number of data set groups F4 than the data set group F2 shown in FIG. 10 are extracted from the initial data set group F1 shown in FIG. 9 is shown.

In the example of FIG. 11, it is assumed that the number of data sets is not reduced little by little, but is reduced to the number shown in FIG. 11 by one extraction. In the example of FIG. 11, since there was a data set group F4 in which the statistical variation accidentally included in the initial data set group F1 was extremely small, those data set groups F4 were extracted. The water speed of the data set group F4 tends to increase with the passage of time, that is, the fuel efficiency performance tends to be unreliable.

In this way, as the number of data sets decreases, the possibility of extracting a data set group whose statistical variation becomes smaller by chance increases, and it becomes difficult to utilize the tendency shown by the data group. In this embodiment, not only the amount of decrease in statistical variation is large, but also the amount of decrease in the number of data sets is as small as possible. Data sets are extracted when the range of items is limited. As a result, the decrease in the number of data sets can be suppressed, and the possibility of extracting a data set group in which the statistical variation as described above is accidentally reduced can be reduced.

Also, for example, if it is possible to keep other conditions constant, the measured values of water speed and fuel consumption will show a substantially constant relationship with each other. However, the accumulated measured values will be mixed with measured values showing different relationships due to the influence of various situations that change during operation. When performing performance analysis, more situations appear during flight than in other situations (hereinafter """, rather than extracting datasets showing relationships in special situations that only occasionally appeared during flight. It is desirable to extract a set of data sets that show the relationship in "frequent situations").

The relationship indicated by the water velocity and fuel consumption measured in the frequent situation shows a more reproducible relationship than the other relationships among the relationships indicated by the accumulated water velocity and fuel consumption. It can be said that there is. Then, when the data set group showing the relationship in a special situation is excluded, the statistical variation is compared with the case where the data set group showing the relationship in a frequent situation is excluded for the decrease in the number of data sets. The amount of decrease (that is, the rate of decrease) becomes large.

Therefore, in this embodiment, by determining the narrowing conditions as described above using the reduction rate, the data set group including the measured values in a special situation is excluded as compared with the case where the reduction rate is not used. To make it easier, it is easier to extract a data set group including measurement values in a frequent situation, that is, a data set group including measurement value groups of analysis item groups showing a relationship with higher reproducibility than other relationships. I have to.

In addition, the items of the narrowing item group whose range is limited when excluding the data set group including the measured values in a special situation have an influence on the relationship between the water speed and the fuel consumption such as wind speed and tidal current. It does not have to be limited to clear items. To put it the other way around, even if the change in the situation that affects the relationship between water speed and fuel consumption and the change in the measured value are accidentally linked items, if the reduction rate satisfies the narrowing conditions, As described above, it contributes to the extraction of the data set group including the measured value group of the analysis item group showing the relationship with higher reproducibility.

Further, in this embodiment, the number of data sets was not reduced at once, but was gradually reduced as shown in FIG. For example, when the data set group F4 shown in FIG. 11 contains measured values in a special situation, the data set group F4 is compared with the case where the number of data sets is reduced at one time by gradually reducing the number of data sets. Since the measured values in the special situation included in the above can be easily excluded, it is possible to facilitate the extraction of the data set group including the measured value group of the analysis item group showing a highly reproducible relationship as described above.

Further, in this embodiment, the satisfaction of the narrowing-down condition and the satisfaction of the end condition are determined by the function of the performance analysis device 10. As a result, a certain data set group can be stably extracted without depending on the ability and experience of the worker who performs the performance analysis. Further, in this embodiment, the number of data sets that can be reduced by one extraction is limited. This makes it possible to prevent the number of data sets from becoming too small.

Further, in this embodiment, statistical variation was expressed by the difference between the measured values of each item in the analysis item group and the relational expression. By extracting a data set group that reduces the statistical variation, compared to the case where statistical variation that is not related to the relational expression between the measured values is used, the measured values of one item in the analysis item group are compared with other items. It is possible to improve the accuracy when estimating the measured value of.

[2] Modifications The above-mentioned examples are merely examples of the implementation of the present invention, and may be modified as follows. Further, the examples and the modified examples may be combined and implemented as necessary.

[2-1] End condition In the embodiment, the end condition determination unit 106 determines the end when the number of data sets satisfies the end condition, but an end condition different from that may be used. The end determination unit 106 may determine the end when, for example, the statistical variation of the measured values of the analysis item group in the extracted data set group satisfies the end condition. Specifically, the end determination unit 106 determines that the end condition is satisfied when the statistical variation becomes less than the threshold value.

Further, the end determination unit 106 may determine the end when the amount of reduction in statistical variation before and after the extraction satisfies the end condition. Specifically, the end determination unit 106 determines that the end condition is satisfied when the amount of decrease in statistical variation becomes equal to or greater than the threshold value. Further, the end determination unit 106 may determine the end when the amount of decrease in the number of data sets before and after the extraction satisfies the end condition. Specifically, the end determination unit 106 determines that the end condition is satisfied when the amount of decrease in the number of data sets becomes equal to or greater than the threshold value.

When the number of datasets or the amount of decrease in the number of datasets is used as the end condition, the extraction process can be terminated before the number of datasets becomes too small. However, in those cases, the extraction process may be completed before the statistical variation becomes so small. On the other hand, when the statistical variability or the reduction amount of the statistical variability is used as the end condition, the statistical variability can be surely reduced to a certain level.

In addition, the end determination unit 106 determines the statistical variation in the data set group after extraction, the number of data sets in the data set group after extraction, the amount of decrease in statistical variation before and after extraction, or the decrease in the number of data sets before and after extraction. Satisfaction of the termination condition may be determined by combining two or more of the four determination materials of quantity. For example, the end determination unit 106 determines that the end condition is satisfied when the statistical variation is equal to or greater than the first threshold value and the number of data sets is equal to or greater than the second threshold value.

Further, even if the end determination unit 106 calculates a function using two or more of the above four determination materials and determines that the end condition is satisfied when the obtained value satisfies a predetermined condition. Good. For example, the end determination unit 106 ends when the value obtained by multiplying the amount of decrease in statistical variation by a predetermined coefficient and the value obtained by multiplying the amount of decrease in the number of data sets by a predetermined coefficient is equal to or greater than the threshold value. Judge that the conditions are met.

Further, depending on how the above threshold value is set, for example, the number of data sets may be reduced too much, and when the data sets are extracted, the reduction amount ratio or the statistical variation may be rather increased. Therefore, in addition to the above-mentioned end condition, the end determination unit 106 satisfies the second end condition when the reduction rate or the statistical variation increases in a certain number of data set groups after extraction. You may judge that. In that case, the data set extraction unit 103 may use the last extracted data set group as a formally extracted data set group, or may use the previously extracted data set group as a formally extracted data set group. ..

In this modified example, as in the embodiment, a certain data set can be stably extracted without depending on the worker. In addition, by combining the four judgment factors, it is possible to extract the data set group while balancing the prevention of excessive decrease in the number of data sets and the improvement of the certainty of the decrease of statistical variability to a certain level. it can.

[2-2] Fixing the reduction amount of statistical variation In the embodiment, the reduction amount of the number of data sets in the extraction process was fixed to a predetermined amount, but the reduction amount of the statistical variation may be fixed. .. In this modification, the data set extraction unit 103 extracts by limiting the range so that the amount of reduction in statistical variation becomes a predetermined amount before and after the extraction.

When the data set group is tentatively extracted as in the embodiment, the data set extraction unit 103 calculates the relational expression between the measured values of the data set group before extraction and the analysis item group based on the tentatively extracted data set group. Instruct the calculation unit 105. The data set extraction unit 103 calculates the amount of reduction in statistical variation based on the calculated relational expression in the same manner as the condition satisfaction determination unit 104 in the embodiment, and the calculated amount of reduction in statistical variation becomes a predetermined amount. Repeat the extraction until.

FIG. 12 shows an example of the statistical variation of this modified example and the transition of the number of data sets. In FIG. 12, as in the example of FIG. 8, the point E0 showing the relationship between the statistical variation of the initial data set group and the number of data sets and the above relationship of the four data set groups extracted in the first extraction process are described. Points E1, E2, E3, and E4 (trim, wave height, tidal current, wind speed) indicating the above are represented. In this modification, the statistical variation decreases by Δσ each time the extraction process is performed.

In the first extraction, the amount of decrease in the number of data sets is in the order of trim> wave height> tidal current> wind speed. In this case, the rate of decrease in the wind velocity data set group (the ratio of the amount of decrease in statistical variation to the amount of decrease in the number of datasets before and after extraction) is the largest, so the narrowing conditions are satisfied and before the next extraction. It is used as a data set group. Similarly, in the second and third extractions, the data set group with the smallest decrease in the number of data sets is extracted.

In this modification, the end determination unit 106 determines that the end condition is satisfied when the statistical variation becomes less than the threshold value. FIG. 12 shows the threshold Th2 used as the end condition. When the third extraction is performed, the statistical variation becomes less than the threshold Th2, so that the extraction of the data set group is completed, and the data set group extracted by limiting the wave height range in the third extraction is the final. It is stored in the extracted area as a group of data sets extracted.

When the reduction amount of statistical variation is fixed as in this modification, for example, when the reduction amount of statistical variation or statistical variation is used as the end condition, the extraction process required until the end condition is satisfied. The number of times can be fixed. Further, when the reduction amount of the number of data sets is fixed as in the embodiment, for example, when the number of data sets or the reduction amount of the number of data sets is used as the end condition, the extraction process required until the end condition is satisfied. The number of times can be fixed.

In each of the above examples, the threshold of the number of data sets was used when ΔN was constant, and the threshold of statistical variation was used when Δσ was constant, but the present invention is not limited to this, and when ΔN is constant. The threshold of statistical variability or the threshold of reduction in statistical variability may be used, and the number of datasets or the threshold of reduction in the number of datasets may be used when Δσ is constant. In either case, the number of datasets, statistical variability, or reductions thereof defined as threshold values can be reduced to a certain level.

[2-3] Fluctuation of reduction amount In each of the above examples, the reduction amount of the number of data sets and the reduction amount of statistical variation were constant, but these reduction amounts may be changed. In this modification, when the data set extraction unit 103 performs extraction by limiting the range so that the number of datasets or the amount of reduction in statistical variation becomes a predetermined amount before and after extraction, the extraction up to the previous time is performed. The amount according to the number of times is set as a predetermined amount.

FIG. 13 shows an example of the statistical variation of this modified example and the transition of the number of data sets. In FIG. 13, as in the example of FIG. 8, the point E0 showing the relationship between the statistical variation of the initial data set group and the number of data sets, and the above-mentioned four data set groups extracted by the three extraction processes, respectively. The points showing the relationship are shown. In the example of FIG. 13, it is assumed that the end determination unit 106 determines that the end condition is satisfied when the statistical variation becomes less than the threshold Th3.

In this case, the data set extraction unit 103 sets a predetermined amount to ΔN1 in the first extraction, a predetermined amount to ΔN2 smaller than ΔN1 in the second extraction, and a predetermined amount to be smaller than ΔN2 in the third extraction. A predetermined amount is gradually reduced, such as ΔN3. As a result, the predetermined amount is increased to reduce the number of extractions until the statistical variation becomes small to some extent, while the predetermined amount is decreased as the statistical variation approaches the threshold value, and as many data sets as possible are obtained. The number can be used to satisfy the end condition.

In addition, even when the end condition is satisfied when the number of data sets becomes less than the threshold value and the data set extraction unit 103 extracts by limiting the range so that the amount of reduction in statistical variation becomes a predetermined amount. The predetermined amount may be reduced according to the number of extractions. In that case as well, the termination condition can be satisfied with as many data sets as possible by reducing the predetermined amount as the number of data sets approaches the threshold value while reducing the number of extractions.

[2-4] Change of narrowing item group In the embodiment, the extraction process is performed using the narrowing item group selected first to the end, but the narrowing item group may be changed in the middle. In this modification, the data set extraction unit 103 changes the narrowing item group before and after the extraction of the data set group. The data set extraction unit 103, for example, removes an item having a limited range when extracting a data set group, and instead adds another item as a narrowing item group.

In the example of FIG. 8, after performing the first extraction, the data set extraction unit 103 removes the "wind speed" that limits the range when extracting the data set group, and instead sets the "rotation speed" or the like. Make changes to add any of the items as a group of refined items. If the range of items that have been limited once is further limited, the number of data sets may be significantly reduced and the narrowing conditions may hardly be satisfied.

In that case, it is better to replace another item with a narrowing item group and perform extraction to reduce unnecessary temporary extraction and increase temporary extraction that is more likely to satisfy the narrowing condition than that extraction. be able to. In the above example, the data set extraction unit 103 exchanged the items of the narrowing down item group, but the present invention is not limited to this, and may be added or deleted. In either case, by appropriately changing the narrowed-down item group, it is possible to reduce unnecessary temporary extraction as compared with the case where the narrowed-down item group does not change.

[2-5] Limiting the range of measured values of narrowed-down items The method of limiting the range of measured values of narrowed-down items is not limited to the method described in the examples. The data set extraction unit 103 may limit the range by combining the measured values of two or more items, for example.

For example, when combining wind speed and wind direction, the data set extraction unit 103 limits the wind speed to a range below the first threshold when the wind direction is headwind, and limits the wind speed to a range below the second threshold when the wind direction is a tailwind. If the wind direction is crosswind, the wind speed is limited to a range below the third threshold. Further, when the pitch and roll are combined, the data set extraction unit 103 limits the value obtained by adding the pitch and the roll to a range in which the value obtained by adding the pitch and the roll is less than the threshold value.

Further, the relational expression between the measured values calculated for the measured values of the narrowed-down item group may be used to limit the range. In that case, the relational expression calculation unit 105 calculates the relational expression between the measured values from the set of the measured values of a plurality of predetermined items in the narrowed-down item group. Then, the data set extraction unit 103 extracts the data set group at the time when the statistical variation represented by the difference between the calculated measured values of the narrowed-down item group and the relational expression includes the measured values satisfying a predetermined condition. To do.

For example, when combining the number of revolutions and the change in ship speed, the data set extraction unit 103 uses the difference from the curve showing the relational expression calculated for the number of revolutions and the change in ship speed (difference in the X-axis direction, difference in the Y-axis direction, or The variance (which may be the mean value or standard deviation, etc.) of the shortest distance difference, etc.) is calculated as an index of statistical variation. The data set extraction unit 103 sequentially excludes the set of measured values having a large difference from the curve, and measures the set of measured values remaining when the calculated variance becomes less than the threshold value within a limited range. Used as a value.

When the items in the narrowed-down item group have a relationship that affects each other's increase / decrease in the measured value, the larger the statistical variation in the difference from the relational expression, the more abnormal the operation of the measuring device or the misunderstanding of the measurer, etc. There is a high possibility that an outlier is included. However, it may be difficult to determine the abnormal value by looking only at the measured value of one item. However, by limiting the range of measured values based on the relational expression between the measured values as described above, the data set when such an abnormal value is measured is excluded as compared with the case where the relational expression is not used. Can be made easier.

[2-6] Narrowing conditions The narrowing conditions may be different from those in the examples. In the example, the condition satisfaction determination unit 104 determined that the data set group after extraction of one item having the largest calculated reduction rate satisfies the narrowing condition, but the reduction rate is not limited to one, for example. It may be determined that the data set group after extracting a predetermined number of items from the larger one satisfies the narrowing condition.

In that case, further extraction processing is performed on the extracted predetermined number of data sets. For example, when it is determined that two data set groups satisfy the narrowing condition, if the extraction is performed three times, a maximum of eight data set groups will be extracted. In that case, a plurality of data set groups may be used for performance analysis, or a plurality of data set groups may be narrowed down to one at the end.

When the final refinement is performed, for example, the condition satisfaction determination unit 104 satisfies the final refinement condition in the data set group having the smallest final statistical variation or the data set group having the largest number of final data sets. You can judge that. Further, in the embodiment, the condition satisfaction determination unit 104 determines the satisfaction of the narrowing-down condition by comparing the reduction rate calculated for each item, but it is not such a relative condition but an absolute condition. May be used as the narrowing condition.

For example, the condition satisfaction determination unit 104 determines that the narrowing condition is satisfied when the calculated reduction amount ratio is larger than a predetermined threshold value (= a predetermined standard in this case). In this case as well, a plurality of data set groups may be extracted. In that case, the plurality of data set groups may be narrowed down to one at the end in the same manner as described above. Further, the condition satisfaction determination unit 104 may use a narrowing condition that combines an absolute condition and a relative condition.

In that case, the condition satisfaction determination unit 104 determines, for example, that when there are a plurality of data sets that satisfy the absolute conditions, one data set group that satisfies the relative conditions satisfies the narrowing condition. Regardless of which narrowing condition is used, a certain data set can be stably extracted without depending on the worker, as in the embodiment.

[2-7] Number of extractions The extraction process and the determination process of the narrowing conditions were repeated in the examples, but may be performed only once. In that case, the first extraction described in each of the above examples is performed, and when the data set group satisfying the narrowing condition is determined, the extraction process and the narrowing condition determination process are completed. Even in that case, the point of extracting so that the number of datasets does not become too small is the same as in the embodiment, so it tends to be easier to use in general situations than when extraction is performed regardless of the number of datasets. Data sets showing the above can be extracted.

[2-8] Output of data set group In the embodiment, the data set extraction unit 103 outputs a data set group determined to satisfy the narrowing condition, but the data set not determined to satisfy the narrowing condition is not satisfied. The group may be output. In that case, the output multiple data set groups may be narrowed down at the discretion of the worker. Even in that case, if the worker makes an appropriate judgment, it is possible to extract a data set group including the measured value group of the analysis item group showing the relationship with higher reproducibility than the other relationships as in the embodiment. it can. Then, according to this modification, it is possible to support narrowing down the measured value group of the analysis item group to the measured value group included in such a data set group.

When the condition satisfaction determination unit 104 determines the narrowing condition, as described in the embodiment, a certain data set group is stably extracted without depending on the ability and experience of the worker who performs the performance analysis. We were able to. On the other hand, for experienced workers, even if there was a set of data sets that were considered to be more appropriate and should be extracted, that judgment could not be reflected. In this modified example, the judgment of such an experienced worker or the like can be reflected in the extraction of the data set group.

[2-9] Statistical variation In the examples, statistical variation expressed by the difference between the measured values of each item in the analysis item group and the relational expression was used, but other statistical variations were used. May be good. For example, variations in the X-axis direction or the Y-axis direction may be used as statistical variations regardless of the relational expression between the measured values.

[2-10] Analysis item group In the examples, only two items were included in the analysis item group, but three or more items may be included. In that case, the narrowing condition may be determined based on the statistical variation expressed by the difference between the measured values of the three or more items and the relational expression.

[2-11] Analysis item group and narrowing item group In the examples, the items selected by the user's operation from the measurement items in which the measured values are accumulated were used as the analysis item group and the narrowing item group. However, for example, the analysis item group and the narrowing item group may be predetermined, and the measured value group of the defined analysis item group and the measured value group of the narrowing item group may be accumulated separately.

When the analysis item group and the narrowing item group are predetermined, the data set extraction unit 103 refers to, for example, an external system for storing each measurement value group instead of accumulating each measurement value group. The measured value group of the analysis item group and the measured value group of the narrowed down item group may be read. As described above, the method of determining the analysis item group and the narrowing item group and the input method of each measurement value group are any methods as long as the data set extraction unit 103 can perform the same processing as each of the above examples. May be used.

[2-12] Operations Performed by Each Function The functional configuration of the present invention is not limited to that shown in FIG. For example, in the embodiment, the data set extraction unit 103 outputs the finally extracted data set group, but another function may perform this output, and a new function not shown in FIG. 2 may output this output. May be done. Further, one function may perform extraction of the data set group by the data set extraction unit 103 and determination of the narrowing condition by the condition satisfaction determination unit 104. In short, as long as the functions shown in FIG. 2 are realized in the performance analysis device 10 as a whole, the range of operations performed by each function may be freely defined.

[2-13] Device for Realizing Each Function The device for realizing each function shown in FIG. 2 is not limited to the performance analysis device 10. For example, the functions realized by the performance analysis device 10 may be shared by two or more devices, or the resources provided by the cloud service may be realized. In either case, each function shown in FIG. 2 may be realized by some computer resource.

[2-14] Category of Invention The present invention can be regarded as an information processing method for realizing the processing performed by the information processing device in addition to the information processing device called the performance analysis device 10 described above. It can also be regarded as a program for operating the controlling computer. The program regarded as the present invention may be provided in the form of a recording medium such as an optical disk in which the program is stored, or may be downloaded to a computer via a network such as the Internet, and the downloaded program may be installed and used. It may be provided in the form of

10 ... Performance analysis device, 20 ... Inspection terminal, 101 ... Measurement value storage unit, 102 ... Item selection reception unit, 103 ... Data set extraction unit, 104 ... Condition satisfaction judgment unit, 105 ... Relational expression calculation unit, 106 ... End judgment Department.

Claims (12)

1. On the computer
   When the measured value group of the first item group and the second item group of the plurality of items at the same time is set as one data set, one or more items of the second item group from the data sets of the plurality of times. A program for executing the process of extracting a data set at a time when it contains a limited range of measurements.
2. Have the computer execute a process of determining whether or not the number of the data sets after extraction and the statistical variation of the measured values of the first item group included in the data set group after extraction satisfy a predetermined condition. The program according to claim 1.
3. The program according to claim 2, wherein the condition is satisfied when the ratio of the reduction amount of the statistical variation to the reduction amount of the number of the data sets before and after the extraction is larger than a predetermined criterion.
4. The program according to claim 2 or 3, wherein the computer is made to execute the extraction process and the determination process with respect to the extracted data set group determined to satisfy the above conditions.
5. The program according to claim 4, wherein the computer is made to change the second item group before and after extraction.
6. The extraction is performed on the computer until at least one of the statistical variability, the number of datasets, the reduction in the statistical variability before and after extraction, or the number of datasets before and after extraction meets the termination condition. The program according to claim 4 or 5, for executing the process to be performed and the process to be determined.
7. The computer is made to execute a process of calculating the relational expression between the measured values of each item from the set of the measured values of the first item group, and the statistical variation represented by the difference from the calculated relational expression is obtained. The program according to any one of claims 2 to 6, for executing a process of determining whether or not the above conditions are satisfied based on the above.
8. The computer is made to execute a process of calculating the relational expression between the measured values of each item from the set of the measured values of the second item group, and the statistical variation represented by the difference from the calculated relational expression is predetermined. The program according to any one of claims 2 to 7, for executing a process of extracting a data set group at a time including a measured value satisfying the above conditions.
9. Any one of claims 1 to 8 for causing the computer to perform a process of performing the extraction by limiting the range so that the amount of decrease in the number of the data sets becomes a predetermined amount before and after the extraction. The program described in.
10. According to any one of claims 2 to 8, for causing the computer to execute the process of performing the extraction by limiting the range so that the amount of reduction of the statistical variation becomes a predetermined amount before and after the extraction. Described program.
11. The program according to claim 9 or 10, wherein the computer is made to execute the extraction process with the amount corresponding to the number of extractions up to the previous time as the predetermined amount.
12. When the information processing device uses the measurement value group of the first item group and the second item group of the plurality of items at the same time as one data set, the second item is selected from the data set group of the plurality of times. An information processing method comprising the step of extracting a data set group at a time that includes a limited range of measurements of one or more items in a group.

Images