JP6987311B2 - Economic indicator forecasting device, production planning device, forecasting model generator and forecasting model generation method - Google Patents

Description

The present invention relates to an economic indicator prediction device, a production planning device, a prediction model generation device, and a prediction model generation method.

Conventionally, there is a need to know the predicted value of this economic indicator before the economic indicator is published. On the other hand, a technique for calculating a predicted value of an economic index using information indicating the positions of a plurality of mobile information terminals (hereinafter referred to as "position information") has been developed (see, for example, Patent Document 1). ). The technique described in Patent Document 1 uses a mathematical model that accepts input of position information and outputs predicted values of economic indicators. Mathematical models for prediction are generated by machine learning.

Japanese Unexamined Patent Publication No. 2019-46376

Hereinafter, a mathematical model by statistics or machine learning is simply referred to as a "model". When calculating the predicted value of an economic indicator using a model, it is possible to improve the reliability of the predicted value by using information that has a high correlation with the economic indicator to be predicted as input to the model. Conceivable. On the other hand, it is considered that the reliability of the predicted value is lowered by using the information having a low correlation with the economic indicator to be predicted as the input to the model.

Here, the location information corresponds to the moving state of the owner of each mobile information terminal, that is, the moving state of a large number of people. In this respect, location information also seems to correlate with economic indicators. However, the purpose of movement by the owner of each mobile information terminal is usually various. The location information is a mixture of information corresponding to movements for such various purposes. For this reason, location information does not always have a high correlation with a particular economic indicator. In the technique described in Patent Document 1, there is a problem that the reliability of the predicted value is low when the position information has a low correlation with the economic index to be predicted.

The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to improve the reliability of predicted values when calculating predicted values of economic indicators using a model.

The economic indicator prediction device of the present invention uses an operation information acquisition unit that acquires operation information indicating the operation state of the target elevator and a prediction model that accepts input of operation information and outputs a predicted value of the target economic indicator. It is equipped with an economic indicator prediction unit that predicts target economic indicators.

According to the present invention, since it is configured as described above, it is possible to improve the reliability of the predicted value when calculating the predicted value of the economic indicator using the model.

It is a block diagram which shows the main part of the economic indicator prediction apparatus which concerns on Embodiment 1. It is a block diagram which shows the main part of the model generation apparatus for prediction which concerns on Embodiment 1. FIG. It is a block diagram which shows the hardware composition of the economic indicator prediction apparatus which concerns on Embodiment 1. It is a block diagram which shows the other hardware composition of the economic indicator predictor which concerns on Embodiment 1. FIG. It is a block diagram which shows the hardware composition of the prediction model generation apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the other hardware configuration of the prediction model generation apparatus which concerns on Embodiment 1. FIG. It is a flowchart which shows the operation of the economic indicator prediction apparatus which concerns on Embodiment 1. It is a flowchart which shows the operation of the prediction model generation apparatus which concerns on Embodiment 1. FIG. It is explanatory drawing which shows the example of the 1st time series data before the execution of the trend extraction process. It is explanatory drawing which shows the example of the 1st time series data after execution of a trend extraction process. It is explanatory drawing which shows the example of the cross-correlation between the 1st time series data and the 2nd time series data. It is a block diagram which shows the main part of the other economic indicator prediction apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the main part of another prediction model generation apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the main part of the other economic indicator prediction apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the main part of the other economic indicator prediction apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the main part of the economic indicator prediction apparatus which concerns on Embodiment 2. It is a block diagram which shows the main part of the model generation apparatus for prediction which concerns on Embodiment 2. FIG. It is a flowchart which shows the operation of the economic indicator prediction apparatus which concerns on Embodiment 2. It is a flowchart which shows the operation of the prediction model generation apparatus which concerns on Embodiment 2. It is a block diagram which shows the main part of the other economic indicator prediction apparatus which concerns on Embodiment 2. FIG. It is a block diagram which shows the main part of another prediction model generation apparatus which concerns on Embodiment 2. FIG. It is a block diagram which shows the main part of the other economic indicator prediction apparatus which concerns on Embodiment 2. FIG. It is a block diagram which shows the main part of the other economic indicator prediction apparatus which concerns on Embodiment 2. FIG. It is a block diagram which shows the main part of the production planning apparatus which concerns on Embodiment 3. It is a block diagram which shows the hardware composition of the production planning apparatus which concerns on Embodiment 3. It is a block diagram which shows the other hardware configuration of the production planning apparatus which concerns on Embodiment 3. FIG. It is a flowchart which shows the operation of the production planning apparatus which concerns on Embodiment 3. It is a block diagram which shows the main part of the other production planning apparatus which concerns on Embodiment 3.

Hereinafter, in order to explain the present invention in more detail, embodiments for carrying out the present invention will be described with reference to the accompanying drawings.

Embodiment 1.
FIG. 1 is a block diagram showing a main part of the economic indicator prediction device according to the first embodiment. The economic indicator prediction device according to the first embodiment will be described with reference to FIG.

Now, one elevator E is installed in one facility F. Alternatively, a plurality of elevators E are installed in one facility F. Alternatively, a plurality of elevators E are installed in a plurality of facilities F. Each elevator E is composed of, for example, an elevator for luggage or an elevator for human cargo. The individual facility F is, for example, an office building, a factory or a commercial facility. Hereinafter, one or more selected elevators E among one elevator E, a plurality of elevators E, or a plurality of elevators E are collectively referred to as "target elevators".

Hereinafter, the economic indicator to be predicted by the economic indicator prediction device 100 is referred to as a “target economic indicator”. The target economic indicator includes one indicator I or a plurality of indicators I.

Here, the individual index I may be an index or index prepared and published by a public institution (for example, a government office or a central bank). Alternatively, the individual indicator I may be a private company (for example, a manufacturer of the economic indicator predictor 100, a user of the economic indicator predictor 100, a service provider using the economic indicator predictor 100, or an economic indicator predictor 100). It may be an index or index created by the user of the service used).

Further, the individual index I may use an index related to an interest rate (for example, a policy interest rate). Alternatively, the individual index I may use an index related to the economy (for example, GDP, business sentiment survey or consumption trend). Alternatively, the individual index I may use an index related to employment (for example, employment statistics). Alternatively, the individual index I may use an index related to prices (for example, the rate of price increase). Alternatively, the individual index I may use an index related to trade (for example, a trade deficit).

Further, each index I may use a so-called "leading index". Alternatively, the individual index I may use a so-called “match index”. Alternatively, the individual index I may use a so-called “lagging index”.

Further, the individual index I may be an index or an index related to a specific area.

Further, the individual index I may be an index or an index related to a specific industry.

As described above, the individual index I may be any index or index.

As shown in FIG. 1, the economic indicator prediction device 100 has an operation information acquisition unit 11, a data preprocessing unit 12, an index information acquisition unit 13, and an economic indicator prediction unit 14.

The operation information acquisition unit 11 acquires information indicating the operation state of the target elevator (hereinafter referred to as “operation information”). The operation information is acquired from, for example, a server device (not shown). The operation information acquisition unit 11 outputs the acquired operation information to the data preprocessing unit 12.

The operation information is composed of, for example, time-series data D1. The time-series data D1 indicates the operating state of the target elevator for each predetermined period (for example, one month) T2 in the latest predetermined period (for example, one year) T1. Specifically, for example, the time-series data D1 indicates the number of times the individual elevators E have been operated, the distance traveled, the travel time, the load weight, or the number of passengers for each period T2.

That is, the time series data D1 is composed of a plurality of values V1. Each of the plurality of values V1 indicates the operating state of the target elevator in the corresponding one period T2 of the plurality of period T2. Specifically, for example, each of the plurality of values V1 indicates the number of times the individual elevators E have been operated, the distance traveled, the travel time, the load weight, or the number of passengers in the corresponding period T2.

The data pre-processing unit 12 acquires the operation information output by the operation information acquisition unit 11. The data pre-processing unit 12 executes pre-processing for the acquired operation information. The data pre-processing unit 12 outputs the operation information on which such pre-processing has been executed to the economic indicator prediction unit 14.

Normally, the time series data D1 in the operation information has seasonal fluctuations. The preprocessing executed by the data preprocessing unit 12 includes a process of removing seasonal fluctuations in the time series data D1. By executing such a process, a so-called "trend" is extracted. Hereinafter, such processing is referred to as "trend extraction processing". Various known techniques can be used for trend extraction. Detailed description of these techniques will be omitted.

Here, "season" in seasonal fluctuation means "season" in a broad sense in the technical field related to statistics. That is, seasonal fluctuations can include not only fluctuations due to so-called "four seasons" but also fluctuations due to various other factors. For example, seasonal fluctuations are due to the fact that the total number of days is different for each month, the number of Sundays for each month is different, the fluctuations are due to the presence or absence of consecutive holidays (for example, year-end holidays or New Year holidays) in each month, and each month. It may include fluctuations due to the presence or absence of bonus payments in.

The index information acquisition unit 13 acquires information indicating the target economic index (hereinafter referred to as “index information”). The index information is acquired from, for example, a server device (not shown). The index information acquisition unit 13 outputs the acquired index information to the economic indicator prediction unit 14. The index information includes the latest value V2 of each index I.

The economic indicator prediction unit 14 acquires the operation information output by the data preprocessing unit 12, and also acquires the index information output by the index information acquisition unit 13. The economic indicator prediction unit 14 calculates the predicted value PV1 of the target economic indicator by using the acquired operation information and the acquired index information. The economic indicator prediction unit 14 outputs the calculated predicted value PV1.

That is, the economic indicator prediction unit 14 calculates the predicted value PV1 of one index I or the predicted value PV1 of each of the plurality of indexes I. The predicted value PV1 is a predicted value for the latest value V2 of each index I. In other words, the predicted value PV1 is a predicted value of each index I to be prepared, prepared and published in the future. The following model (hereinafter referred to as "prediction model") is used for calculating the predicted value PV1.

When the target economic indicator includes one indicator I, the forecasting model includes one model M corresponding to one indicator I. One model M accepts inputs of operation information and index information, and outputs a predicted value PV1 of one index I. On the other hand, when the target economic indicator includes a plurality of indicators I, the forecasting model includes a plurality of models M corresponding to the plurality of indicators I. Each of the plurality of models M accepts input of operation information and index information, and outputs a predicted value PV1 of one index I corresponding to the plurality of indexes I. Specific examples of the prediction model will be described later with reference to FIGS. 7 to 9.

Hereinafter, the processes executed by the operation information acquisition unit 11 may be collectively referred to as “operation information acquisition processing”. Further, the processes executed by the data pre-processing unit 12 may be collectively referred to as "data pre-processing". Further, the processes executed by the index information acquisition unit 13 may be collectively referred to as "index information acquisition process". In addition, the processing executed by the economic indicator prediction unit 14 may be collectively referred to as "economic indicator prediction processing".

FIG. 2 is a block diagram showing a main part of the prediction model generation device according to the first embodiment. The prediction model generation device according to the first embodiment will be described with reference to FIG.

As shown in FIG. 2, the prediction model generation device 200 has a history information acquisition unit 21, a data preprocessing unit 22, and a prediction model generation unit 23.

The history information acquisition unit 21 acquires information indicating the history of the operating state of the target elevator (hereinafter referred to as “first history information”) and information indicating the history of the target economic indicator (hereinafter referred to as “second history information”). .) Is to be acquired. The first history information and the second history information are acquired from, for example, a server device (not shown). The history information acquisition unit 21 outputs the acquired first history information to the data preprocessing unit 22, and outputs the acquired second history information to the prediction model generation unit 23.

The first history information is composed of, for example, time-series data (hereinafter referred to as "first time-series data") d1. The first time-series data d1 indicates the operating state of the target elevator for each predetermined period (for example, one month) t2 in the latest predetermined period (for example, 5 years) t1. Specifically, for example, the first time-series data d1 indicates the number of operating times, the mileage, the traveling time, the load weight, or the number of passengers of each elevator E for each period t2. That is, the first time-series data d1 in the first history information is the same data in different periods with respect to the time-series data D1 in the operation information.

The first time series data d1 is composed of a plurality of values v1. Each of the plurality of values v1 indicates the operating state of the target elevator in the corresponding one period t2 of the plurality of periods t2. Specifically, for example, each of the plurality of values v1 indicates the number of times the individual elevators E have been operated, the distance traveled, the travel time, the load weight, or the number of passengers in the corresponding period t2.

The second history information is composed of, for example, time-series data (hereinafter referred to as “second time-series data”) d2. The second time-series data d2 shows the history of the target economic indicators in the most recent predetermined period (for example, 5 years) t3. That is, the second time series data d2 shows the history of each index I in the latest period t3.

The data pre-processing unit 22 acquires the first history information output by the history information acquisition unit 21. The data pre-processing unit 22 executes pre-processing on the acquired first history information. The data pre-processing unit 22 outputs the first history information in which the pre-processing is executed to the prediction model generation unit 23.

Here, the first time-series data d1 may include an abnormal value. Specifically, for example, an abnormal value caused by a natural disaster such as an earthquake may be included. Therefore, the pre-processing executed by the data pre-processing unit 22 includes a process of removing an abnormal value included in the first time-series data d1 (hereinafter referred to as "abnormal value removing process"). Various known techniques can be used to remove the outliers. Detailed description of these techniques will be omitted.

For example, the data preprocessing unit 22 acquires information indicating the time of occurrence of a natural disaster such as an earthquake in the period t1 or the period t3. The data preprocessing unit 22 uses the acquired information to remove abnormal values caused by natural disasters such as earthquakes. Alternatively, for example, the data preprocessing unit 22 removes abnormal values by executing statistical processing on a plurality of values v1 included in the first time series data d1.

Further, usually, the first time-series data d1 includes seasonal fluctuations similar to the seasonal fluctuations in the time-series data D1. Therefore, the preprocessing executed by the data preprocessing unit 22 includes the same trend extraction processing as the trend extraction processing executed by the data preprocessing unit 12. That is, the pre-processing executed by the data pre-processing unit 22 includes a process of removing seasonal fluctuations in the first time-series data d1.

The prediction model generation unit 23 acquires the first history information output by the data preprocessing unit 22, and also acquires the second history information acquired by the history information acquisition unit 21. The prediction model generation unit 23 generates a prediction model by using the acquired first history information and the acquired second history information. The prediction model generation unit 23 outputs the generated prediction model. The output prediction model is used by the economic indicator prediction unit 14 of the economic indicator prediction device 100.

That is, the prediction model generation unit 23 generates one model M corresponding to one index I or a plurality of models M corresponding to the plurality of indexes I. Specific examples of the prediction model will be described later with reference to FIGS. 7 to 9.

Hereinafter, the processes executed by the history information acquisition unit 21 may be collectively referred to as “history information acquisition process”. Further, the processes executed by the data pre-processing unit 22 may be collectively referred to as "data pre-processing". Further, the processes executed by the prediction model generation unit 23 may be collectively referred to as "prediction model generation processing".

Next, with reference to FIG. 3, the hardware configuration of the main part of the economic indicator prediction device 100 will be described.

As shown in FIG. 3A, the economic indicator prediction device 100 has a processor 31 and a memory 32. The memory 32 stores a program for realizing the functions of the operation information acquisition unit 11, the data preprocessing unit 12, the index information acquisition unit 13, and the economic indicator prediction unit 14. When the processor 31 reads out and executes such a program, the functions of the operation information acquisition unit 11, the data preprocessing unit 12, the index information acquisition unit 13, and the economic indicator prediction unit 14 are realized.

Alternatively, as shown in FIG. 3B, the economic indicator prediction device 100 has a processing circuit 33. In this case, the functions of the operation information acquisition unit 11, the data preprocessing unit 12, the index information acquisition unit 13, and the economic indicator prediction unit 14 are realized by the dedicated processing circuit 33.

Alternatively, the economic indicator prediction device 100 includes a processor 31, a memory 32, and a processing circuit 33 (not shown). In this case, some of the functions of the operation information acquisition unit 11, the data preprocessing unit 12, the index information acquisition unit 13, and the economic indicator prediction unit 14 are realized by the processor 31 and the memory 32, and the remaining functions are realized. Is realized by a dedicated processing circuit 33.

The processor 31 is composed of one or a plurality of processors. The individual processor uses, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a microprocessor, a microprocessor, or a DSP (Digital Signal Processor).

The memory 32 is composed of one or a plurality of non-volatile memories. Alternatively, the memory 32 is composed of one or a plurality of non-volatile memories and one or a plurality of volatile memories. That is, the memory 32 is composed of one or a plurality of memories. The individual memory uses, for example, a semiconductor memory, a magnetic disk, an optical disk, a magneto-optical disk, or a magnetic tape. More specifically, each volatile memory uses, for example, a RAM (Random Access Memory). Further, the individual non-volatile memory is, for example, a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable Advanced Memory), a hard disk, a solid drive A compact disc, a DVD (Digital Vertical Disc), a Blu-ray disc, or a mini disc is used.

The processing circuit 33 is composed of one or a plurality of digital circuits. Alternatively, the processing circuit 33 is composed of one or a plurality of digital circuits and one or a plurality of analog circuits. That is, the processing circuit 33 is composed of one or a plurality of processing circuits. The individual processing circuits are, for example, ASIC (Application Specific Integrated Circuit), PLD (Programmable Logic Device), FPGA (Field Programmable Gate Array), System LSI (System) System Is.

Next, with reference to FIG. 4, the hardware configuration of the main part of the prediction model generator 200 will be described.

As shown in FIG. 4A, the predictive model generator 200 has a processor 41 and a memory 42. The memory 42 stores a program for realizing the functions of the history information acquisition unit 21, the data preprocessing unit 22, and the prediction model generation unit 23. When the processor 41 reads and executes such a program, the functions of the history information acquisition unit 21, the data preprocessing unit 22, and the prediction model generation unit 23 are realized.

Alternatively, as shown in FIG. 4B, the prediction model generation device 200 has a processing circuit 43. In this case, the functions of the history information acquisition unit 21, the data preprocessing unit 22, and the prediction model generation unit 23 are realized by the dedicated processing circuit 43.

Alternatively, the predictive model generator 200 has a processor 41, a memory 42, and a processing circuit 43 (not shown). In this case, some of the functions of the history information acquisition unit 21, the data preprocessing unit 22, and the prediction model generation unit 23 are realized by the processor 41 and the memory 42, and the remaining functions are realized by the processing circuit 43. It will be realized.

The processor 41 is composed of one or a plurality of processors. The individual processors use, for example, CPUs, GPUs, microprocessors, microcontrollers or DSPs.

The memory 42 is composed of one or a plurality of non-volatile memories. Alternatively, the memory 42 is composed of one or a plurality of non-volatile memories and one or a plurality of volatile memories. That is, the memory 42 is composed of one or a plurality of memories. The individual memory uses, for example, a semiconductor memory, a magnetic disk, an optical disk, a magneto-optical disk, or a magnetic tape. More specifically, each volatile memory uses, for example, RAM. Further, as the individual non-volatile memory, for example, ROM, flash memory, EPROM, EEPROM, solid state drive, hard disk drive, flexible disk, compact disk, DVD, Blu-ray disk or mini disk are used.

The processing circuit 43 is composed of one or a plurality of digital circuits. Alternatively, the processing circuit 43 is composed of one or a plurality of digital circuits and one or a plurality of analog circuits. That is, the processing circuit 43 is composed of one or a plurality of processing circuits. The individual processing circuits use, for example, an ASIC, PLD, FPGA, SoC or system LSI.

Next, the operation of the economic indicator prediction device 100 will be described with reference to the flowchart of FIG.

First, the operation information acquisition unit 11 executes the operation information acquisition process (step ST1). Next, the data pre-processing unit 12 executes the data pre-processing (step ST2). Next, the index information acquisition unit 13 executes the index information acquisition process (step ST3). Next, the economic indicator prediction unit 14 executes the economic indicator prediction process (step ST4).

The execution order of the operation information acquisition process, the data preprocessing, and the index information acquisition process is not limited to the example shown in FIG. The operation information acquisition process may be executed before the economic indicator prediction process is executed. The data pre-processing may be performed as long as it is executed after the operation information acquisition process is executed and before the economic indicator prediction process is executed. The index information acquisition process may be executed before the economic indicator prediction process is executed.

Next, the operation of the prediction model generation device 200 will be described with reference to the flowchart of FIG.

First, the history information acquisition unit 21 executes the history information acquisition process (step ST11). Next, the data pre-processing unit 22 executes the data pre-processing (step ST12). Next, the prediction model generation unit 23 executes the prediction model generation process (step ST13).

Next, a specific example of the prediction model will be described with reference to FIGS. 7 to 9. In addition, a specific example of the model generation process for prediction and a specific example of the economic indicator prediction process will be described.

Hereinafter, in the description with reference to FIGS. 7 to 9, the target elevator includes one elevator E. One elevator E is installed in an office building. The time-series data D1 in the operation information indicates the number of operations of one elevator E every month. The first time-series data d1 in the first history information also indicates the number of times of operation of one elevator E every month. The target economic indicator includes one indicator I. One index I uses an arbitrary concordance index. The prediction model includes one model M corresponding to one index I.

Normally, the elevator E installed in an office building is used by an office worker. Therefore, it is considered that the number of times of operation of the elevator E installed in the office building has a high correlation with the economic index. Further, it is considered that the time variation of the number of times of operation of the elevator E installed in the office building precedes the time variation of any coincidence index. This also applies to the mileage, travel time, load weight and number of passengers of the elevator E installed in the office building.

Therefore, as follows, a model M based on the time difference n between the history of the number of operations indicated by the first history information and the history of the index I indicated by the second history information is generated. That is, a model M based on the temporal correlation between the first time-series data d1 and the second time-series data d2 is generated. Specifically, for example, a model M based on the cross-correlation between the first time-series data d1 and the second time-series data d2 is generated.

First, the prediction model generation device 200 executes history information acquisition processing and data preprocessing. As described above, the data pre-processing in the prediction model generation device 200 includes an outlier removal process and a trend extraction process. Here, FIG. 7 shows an example of the first time series data d1 before the execution of the trend extraction process. As shown in FIG. 7, the first time-series data d1 before the execution of the trend extraction process has seasonal fluctuations. On the other hand, FIG. 8 shows an example of the first time series data d1 after the execution of the trend extraction process. As shown in FIG. 8, the trend is extracted by removing the seasonal variation in the first time series data d1.

Next, the prediction model generation device 200 executes the prediction model generation process. At this time, the prediction model generation unit 23 calculates the cross-correlation between the first time-series data d1 and the second time-series data d2 after the trend extraction process is executed. As a result, the prediction model generation unit 23 calculates the amount of time lag n at which the correlation value C between the first time-series data d1 and the second time-series data d2 becomes maximum. FIG. 9 shows an example of the correlation value C with respect to the deviation amount n. In the example shown in FIG. 9, when the deviation amount n is 7 months (that is, when n = +7), the correlation value C becomes maximum. This indicates that the time variation of the number of times of operation of the elevator E precedes the time variation of the index I by 7 months.

The prediction model generation unit 23 generates a model M according to the following equation (1) based on the calculated deviation amount n. That is, the prediction model generation unit 23 generates the model M by the function f (x).

f (x) = a × x (1)

Here, the coefficient a in the function f (x) is substituted with the value V2 included in the index information in the economic indicator prediction process. That is, the latest value V2 of the index I is substituted.

Further, the variable x in the function f (x) is assigned a value indicating the volatility R as shown below in the economic indicator prediction process. That is, as described above, the operation information includes a plurality of values V1. The plurality of values V1 include the value V1 corresponding to the time n months before the time corresponding to the value V2. Therefore, a value indicating the volatility R of the value V1 in the period n months ago is assigned to the variable x. As a result, the predicted value PV1 is calculated.

After the model M is generated, the economic indicator prediction device 100 executes operation information acquisition processing, data preprocessing, and index information acquisition processing. Next, the economic indicator prediction process is executed by the economic indicator prediction device 100. At this time, the model M accepts the input of the operation information and the index information. As a result, the latest value V2 of the index I is assigned to the coefficient a of the function f (x), and the value indicating the fluctuation rate R is assigned to the variable x of the function f (x). The model M outputs the predicted value PV1 based on the function f (x).

When the target economic indicator includes a plurality of indicators I and the prediction model includes a plurality of models M, the same processing as these processing is performed for each of the plurality of models M. Will be executed. As a result, the prediction model generation device 200 generates a plurality of models M corresponding to the plurality of indexes I. Further, the economic indicator prediction device 100 calculates the predicted value PV1 of each of the plurality of indicators I.

Next, a modification of the economic indicator prediction device 100 will be described with reference to FIG. 10. Further, with reference to FIG. 11, a modification of the prediction model generation device 200 will be described.

The operation information acquired by the operation information acquisition unit 11 may be composed of time-series data D1 having no seasonal variation or time-series data D1 from which seasonal variation has been removed. In this case, as shown in FIG. 10, the economic indicator prediction device 100 may not have the data preprocessing unit 12. That is, the main part of the economic indicator prediction device 100 may be configured by the operation information acquisition unit 11, the index information acquisition unit 13, and the economic indicator prediction unit 14.

Further, the first history information acquired by the history information acquisition unit 21 is composed of the first time-series data d1 having no abnormal value and seasonal variation, or the first time-series data d1 from which the abnormal value and seasonal variation are removed. It may be the one that has been done. In this case, as shown in FIG. 11, the prediction model generation device 200 may not have the data preprocessing unit 22. That is, the main part of the prediction model generation device 200 may be configured by the history information acquisition unit 21 and the prediction model generation unit 23.

Next, another modification of the economic indicator prediction device 100 will be described with reference to FIGS. 12 and 13.

The coefficient a in the function f (x) may be a constant. Alternatively, the coefficient a in the function f (x) may be a value determined according to a plurality of values V1 included in the operation information. In this case, it is not necessary to input the index information in each model M. Further, in the economic indicator prediction device 100, the index information acquisition unit 13 is unnecessary. That is, the prediction model may be one that accepts the input of operation information and outputs the predicted value PV1 of the target economic index.

In this case, as shown in FIG. 12 or 13, the economic indicator prediction device 100 may not have the index information acquisition unit 13. That is, as shown in FIG. 12, the main part of the economic indicator prediction device 100 may be configured by the operation information acquisition unit 11, the data preprocessing unit 12, and the economic indicator prediction unit 14. Alternatively, as shown in FIG. 13, the main part of the economic indicator prediction device 100 may be configured by the operation information acquisition unit 11 and the economic indicator prediction unit 14.

Next, other modification examples will be described.

The pre-processing executed by the data pre-processing unit 12 may include an abnormal value removing process similar to the abnormal value removing process executed by the data pre-processing unit 22. That is, the preprocessing executed by the data preprocessing unit 12 may include a process for removing an abnormal value included in the time series data D1.

The operation information may include two or more time series data D1s. For example, the operation information includes time-series data D1 indicating the number of times the target elevator has been operated, time-series data D1 indicating the mileage of the target elevator, time-series data D1 indicating the traveling time of the target elevator, and time indicating the load weight of the target elevator. It may include any two or more time-series data D1 of the series data D1 or the time-series data D1 indicating the number of passengers of the target elevator. The economic indicator prediction unit 14 may use two or more time-series data D1 in combination when executing the economic indicator prediction process. Alternatively, at this time, the economic indicator prediction unit 14 may selectively use the two or more time-series data D1s.

The first history information may include two or more first time series data d1. For example, the first history information includes first time-series data d1 showing the history of the number of times the target elevator has been operated, first time-series data d1 showing the history of the travel distance of the target elevator, and first history of the travel time of the target elevator. 1 time series data d1, 1st time series data d1 showing the history of the load weight of the target elevator, or 1st time series data d1 showing the history of the number of passengers of the target elevator, any two or more of the first time It may include the series data d1. The prediction model generation unit 23 may use the two or more first time-series data d1 in combination when executing the prediction model generation process. Alternatively, at this time, the prediction model generation unit 23 may selectively use the two or more first time series data d1.

The prediction model is not limited to the specific examples described with reference to FIGS. 7 to 9. The forecasting model may be any model that accepts the input of operation information or operation information and index information and outputs the predicted value PV1 of the target economic index.

The prediction model generation process is not limited to the specific examples described with reference to FIGS. 7 to 9. Various known techniques related to statistics or machine learning can be used for the prediction model generation process. Detailed description of these techniques will be omitted.

As described above, the economic indicator prediction device 100 according to the first embodiment has the operation information acquisition unit 11 that acquires the operation information indicating the operation state of the target elevator, and the operation information acquisition unit 11 that receives the input of the operation information and predicts the target economic indicator. It includes an economic indicator prediction unit 14 that predicts a target economic indicator using a prediction model that outputs PV1. By using the operation information, it is possible to improve the reliability of the predicted value PV1 as compared with the case of using the position information.

Further, the prediction model generation device 200 according to the first embodiment acquires the first history information indicating the history of the operating state of the target elevator and the second history information indicating the history of the target economic indicator. The prediction model includes an acquisition unit 21 and a prediction model generation unit 23 that generates a prediction model using the first history information and the second history information, and the prediction model accepts input of operation information indicating an operation state. It outputs the predicted value PV1 of the target economic indicator. By using the first history information and the second history information, it is possible to generate a prediction model in the economic indicator prediction device 100.

Further, in the prediction model generation method according to the first embodiment, the history information acquisition unit 21 acquires the first history information indicating the history of the operating state of the target elevator and the second history showing the history of the target economic indicator. The prediction model includes a step ST11 for acquiring information and a step ST12 for the prediction model generation unit 23 to generate a prediction model using the first history information and the second history information, and the prediction model indicates an operating state. It accepts the input of operation information and outputs the predicted value PV1 of the target economic indicator. By using the first history information and the second history information, it is possible to generate a prediction model in the economic indicator prediction device 100.

Embodiment 2.
FIG. 14 is a block diagram showing a main part of the economic indicator prediction device according to the second embodiment. The economic indicator prediction device according to the second embodiment will be described with reference to FIG. In FIG. 14, the same blocks as those shown in FIG. 1 are designated by the same reference numerals and the description thereof will be omitted.

In the second embodiment, the target elevator includes one or more selected elevators E among the plurality of elevators E. The plurality of elevators E are installed in the plurality of facilities F.

Hereinafter, the information indicating the area including each of the plurality of facilities F is referred to as "regional information". Further, the information indicating each type of the plurality of facilities F is referred to as "type information". Further, information indicating the use of each of the plurality of elevators E is referred to as "use information". In addition, information including at least one of regional information, type information, and usage information is referred to as "additional information".

The area information is, for example, information indicating a prefecture or a city, ward, town, or village including each of a plurality of facilities F. The type information is, for example, information indicating whether each of the plurality of facilities F is an office building, a commercial facility, or a factory. The usage information is, for example, information indicating whether each of the plurality of elevators E is for luggage or for humans.

As shown in FIG. 14, the economic indicator prediction device 100a has an operation information acquisition unit 11, a data preprocessing unit 12, an index information acquisition unit 13, and an economic indicator prediction unit 14. In addition to this, the economic indicator prediction device 100a has an additional information acquisition unit 15 and an elevator selection unit 16.

The additional information acquisition unit 15 acquires additional information. The additional information is acquired from, for example, a server device (not shown). The additional information acquisition unit 15 outputs the acquired additional information to the elevator selection unit 16.

The elevator selection unit 16 acquires additional information output by the additional information acquisition unit 15. The elevator selection unit 16 selects one or more elevators E corresponding to the target economic indicators among the plurality of elevators E by using the acquired additional information. As a result, the selected one or more elevators E become the target elevators. The elevator selection unit 16 outputs information indicating the result of the selection (hereinafter referred to as “selection result information”) to the operation information acquisition unit 11 and the economic indicator prediction unit 14.

When the target economic indicator includes one indicator I, one or more elevators E corresponding to one indicator I are selected. On the other hand, when the target economic indicator includes a plurality of indicators I, one or more elevators E corresponding to each of the plurality of indicators I are selected. That is, one or more elevators E corresponding to the individual indicators I are selected.

The operation information acquisition unit 11 acquires operation information related to one or more elevators E indicated by the selection result information. Based on the selection result information, the economic indicator prediction unit 14 uses the corresponding operation information among the acquired operation information to calculate the predicted value PV1 of each index I.

That is, when the target economic indicator includes one index I, the operation information acquisition unit 11 acquires operation information related to one or more elevators E corresponding to one index I. The economic indicator prediction unit 14 uses the acquired operation information when calculating the predicted value PV1 of one index I.

On the other hand, when the target economic indicator includes a plurality of indicators I, the operation information acquisition unit 11 acquires operation information related to one or more elevators E corresponding to each of the plurality of indicators I. When calculating the predicted value PV1 of each of the plurality of indicators I, the economic indicator prediction unit 14 uses the corresponding operation information among the acquired operation information.

Hereinafter, the processes executed by the additional information acquisition unit 15 may be collectively referred to as “additional information acquisition processing”. Further, the processes executed by the elevator selection unit 16 may be collectively referred to as "elevator selection process".

Here, a specific example of the elevator selection process using the type information will be described.

Now, one elevator E installed in an office building, one elevator E installed in a commercial facility, and one elevator E installed in a factory are included in a plurality of elevators E. do. Further, it is assumed that one index I using the leading index, one index I using the coincidence index, and one index I using the lagging index are included in the target economic index.

For example, the operating state of the elevator E installed in an office building is considered to behave as a leading exponential with respect to economic trends. Therefore, it is considered that the time fluctuation of the operating state has a high probability of leading to the time fluctuation of any coincidence index and also has a high probability of leading to the time fluctuation of any lagging index. In addition, it is considered that the time variation of the operation information can precede the time variation of some leading indices. This also applies to the elevator E installed in the factory.

On the other hand, the operating state of the elevator E installed in the commercial facility is considered to behave in a lagging exponential manner with respect to economic trends. Therefore, it is considered that the time variation of the operating state is unlikely to precede the time variation of any leading index and is unlikely to precede the time variation of any leading index. Further, it is considered that the time variation of the operation information can precede the time variation of some lagging indexes.

Therefore, the elevator selection unit 16 selects at least one of the elevator E installed in the office building and the elevator E installed in the factory for the index I using the leading index. Further, the elevator selection unit 16 selects at least one of the elevator E installed in the office building and the elevator E installed in the factory for the index I using the coincidence index. Further, the elevator selection unit 16 uses at least one of the elevator E installed in the office building, the elevator E installed in the commercial facility, or the elevator E installed in the factory for the index I using the lagging index. select.

Next, a specific example of the elevator selection process using regional information will be described.

Now, one elevator E is installed in one facility F in one specific area (hereinafter referred to as "first area"), and another area (hereinafter referred to as "second area"). It is assumed that one elevator E is installed in one facility F in the above. Further, it is assumed that one index I using the index related to the first region and one index I using the index related to the second region are included in the target economic index.

In this case, the elevator selection unit 16 selects the elevator E installed in the facility F in the first area for the index I using the index related to the first area. Further, the elevator selection unit 16 selects the elevator E installed in the facility F in the second area for the index I using the index related to the second area.

FIG. 15 is a block diagram showing a main part of the prediction model generation device according to the second embodiment. The prediction model generation device according to the second embodiment will be described with reference to FIG. In FIG. 15, the same blocks as those shown in FIG. 2 are designated by the same reference numerals and the description thereof will be omitted.

As shown in FIG. 15, the prediction model generation device 200a has a history information acquisition unit 21, a data preprocessing unit 22, and a prediction model generation unit 23. In addition to this, the prediction model generation device 200a has an additional information acquisition unit 24 and an elevator selection unit 25.

The additional information acquisition unit 24 executes an additional information acquisition process similar to the additional information acquisition process executed by the additional information acquisition unit 15. The elevator selection unit 25 executes an elevator selection process similar to the elevator selection process executed by the elevator selection unit 16. However, the selection result information by the elevator selection unit 25 is output to the history information acquisition unit 21 and the prediction model generation unit 23.

The history information acquisition unit 21 acquires the first history information related to one or more elevators E indicated by the selection result information. The prediction model generation unit 23 uses the corresponding first history information of the acquired first history information to generate the individual model M based on the selection result information.

That is, when the target economic indicator includes one index I, the history information acquisition unit 21 acquires the first history information related to one or more elevators E corresponding to one index I. The prediction model generation unit 23 uses the acquired operation information when generating one model M corresponding to one index I.

On the other hand, when the target economic indicator includes a plurality of indicators I, the history information acquisition unit 21 acquires the first history information related to one or more elevators E corresponding to each of the plurality of indicators I. When the prediction model generation unit 23 generates each of the plurality of models M corresponding to the plurality of indexes I, the prediction model generation unit 23 uses the corresponding first history information among the acquired first history information.

The hardware configuration of the main part of the economic indicator prediction device 100a is the same as that described with reference to FIG. 3 in the first embodiment. Therefore, illustration and description will be omitted. That is, each function of the operation information acquisition unit 11, the data preprocessing unit 12, the index information acquisition unit 13, the economic indicator prediction unit 14, the additional information acquisition unit 15, and the elevator selection unit 16 is realized by the processor 31 and the memory 32. It may be one, or it may be realized by a dedicated processing circuit 33.

The hardware configuration of the main part of the prediction model generation device 200a is the same as that described with reference to FIG. 4 in the first embodiment. Therefore, illustration and description will be omitted. That is, each function of the history information acquisition unit 21, the data preprocessing unit 22, the prediction model generation unit 23, the additional information acquisition unit 24, and the elevator selection unit 25 is realized by the processor 41 and the memory 42. It may be realized by a dedicated processing circuit 43.

Next, the operation of the economic indicator prediction device 100a will be described with reference to the flowchart of FIG. In FIG. 16, the same steps as those shown in FIG. 5 are designated by the same reference numerals and the description thereof will be omitted.

First, the additional information acquisition unit 15 executes the additional information acquisition process (step ST5). Next, the elevator selection unit 16 executes the elevator selection process (step ST6). Next, the processes of steps ST1 to ST4 are executed.

Next, the operation of the prediction model generation device 200a will be described with reference to the flowchart of FIG. In FIG. 17, the same steps as those shown in FIG. 6 are designated by the same reference numerals and the description thereof will be omitted.

First, the additional information acquisition unit 24 executes the additional information acquisition process (step ST14). Next, the elevator selection unit 25 executes the elevator selection process (step ST15). Next, the processes of steps ST11 to ST13 are executed.

As the economic indicator prediction device 100a, various modifications similar to those described in the first embodiment can be adopted. Further, as the prediction model generation device 200a, various modifications similar to those described in the first embodiment can be adopted.

For example, as shown in FIG. 18, the economic indicator prediction device 100a may not have the data preprocessing unit 12. That is, the main part of the economic indicator prediction device 100a may be configured by the operation information acquisition unit 11, the index information acquisition unit 13, the economic indicator prediction unit 14, the additional information acquisition unit 15, and the elevator selection unit 16. ..

Further, as shown in FIG. 19, the prediction model generation device 200a may not have the data preprocessing unit 22. That is, the main part of the prediction model generation device 200a may be configured by the history information acquisition unit 21, the prediction model generation unit 23, the additional information acquisition unit 24, and the elevator selection unit 25.

Further, as shown in FIG. 20 or 21, the economic indicator prediction device 100a may not have the index information acquisition unit 13. That is, as shown in FIG. 20, the main part of the economic indicator prediction device 100a is composed of the operation information acquisition unit 11, the data preprocessing unit 12, the economic indicator prediction unit 14, the additional information acquisition unit 15, and the elevator selection unit 16. It may be a thing. Alternatively, as shown in FIG. 21, the main part of the economic indicator prediction device 100a may be configured by the operation information acquisition unit 11, the economic indicator prediction unit 14, the additional information acquisition unit 15, and the elevator selection unit 16. ..

Next, other modification examples will be described.

The additional information is not limited to the above specific example. Further, the elevator selection process is not limited to the above specific example. The elevator selection process may be performed as long as one or more elevators E corresponding to the individual indexes I are selected. That is, the elevator selection process may select one or more elevators E that are expected to have a high correlation with the time variation of the individual index I in the operating state. The additional information may include any information as long as such selection is feasible.

For example, the additional information includes an industry to which a company that uses each of the one or more facilities F belongs when one or more of the facilities F are used by a plurality of companies. It may include information indicating (hereinafter referred to as "industry information"). In this case, with respect to the index I using the index related to the specific industry, the elevator E installed in the facility F used by the company belonging to the specific industry may be selected as the target elevator.

As described above, the economic index prediction device 100a according to the second embodiment corresponds to the additional information acquisition unit 15 for acquiring additional information and the target economic index among the plurality of elevators E by using the additional information1. Elevator selection unit 16 for selecting one or more elevators E as the target elevator is provided, and additional information is provided for each of the plurality of facilities F when the plurality of elevators E are installed in the plurality of facilities F. It contains at least one of information indicating an area including, information indicating each type of the plurality of facilities F, or information indicating the use of each of the plurality of elevators E. Thereby, in calculating the predicted value PV1 of each index I, the operation information related to the elevator E suitable for the calculation of the predicted value PV1 can be used for the calculation of the predicted value PV1.

Further, the prediction model generation device 200a according to the second embodiment has an additional information acquisition unit 24 for acquiring additional information, and one or more of the plurality of elevators E corresponding to the target economic index using the additional information. Elevator selection unit 25 for selecting the elevator E of the above as the target elevator, and additional information includes the area including each of the plurality of facilities F when the plurality of elevators E are installed in the plurality of facilities F. Includes at least one of information indicating the above, information indicating each type of the plurality of facilities F, or information indicating the use of each of the plurality of elevators E. Thereby, in generating the model M corresponding to each index I, the first history information related to the elevator E suitable for generating the model M can be used for the generation of the model M.

Embodiment 3.
FIG. 22 is a block diagram showing a main part of the production planning apparatus according to the third embodiment. The production planning apparatus according to the third embodiment will be described with reference to FIG. 22.

As shown in FIG. 22, the production planning device 300 includes an economic indicator prediction device 100, an order information acquisition unit 51, and a production planning unit 52.

The economic indicator prediction device 100 is the same as that described in the first embodiment. Therefore, a detailed description of the economic indicator prediction device 100 will be omitted. The economic indicator prediction device 100 outputs the predicted value PV1 calculated by the economic indicator prediction unit 14 to the production planning unit 52.

Hereinafter, one product P, a plurality of products P, or one or more selected products P among the plurality of products P are collectively referred to as "target products". The individual product P is any system, any device or any component.

The order information acquisition unit 51 acquires information indicating the number of orders for the target product (hereinafter referred to as "order information"). The order information is, for example, input by the producer of the target product (that is, the user of the production planning device 300 or the user of the service using the production planning device 300). The order information acquisition unit 51 outputs the acquired order information to the production planning unit 52.

The production planning unit 52 acquires the predicted value PV1 output by the economic indicator prediction device 100, and also acquires the order information output by the order information acquisition unit 51. The production planning unit 52 calculates the planned value PV2 of the production number of the target product by using the acquired predicted value PV1 and the acquired order information. The production planning unit 52 outputs the calculated planned value PV2.

That is, usually, when a producer produces a target product according to the number of orders, the number of orders may increase due to sudden addition of orders within the period after the order is placed and before the delivery date. Alternatively, the number of orders may decrease due to the partial cancellation of the order within the period after the order is placed and before the delivery date. From the viewpoint of controlling the number of inventories of the target product, it is required to obtain the planned value PV2 of the production number in consideration of the increase / decrease in the number of orders before starting the production of the target product.

Here, it is considered that the increase / decrease in the number of orders has a correlation with the economic trend in the period after the order is received and before the delivery date. For example, during the transition from recession to diastole, additional orders are likely to occur. On the other hand, in the period of transition from the economic expansion period to the recession period, it is considered that the cancellation of orders is likely to occur.

Therefore, the production planning unit 52 uses the predicted value PV1 calculated by the economic indicator prediction device 100 to calculate the planned value PV2. By using the predicted value PV1, it is possible to calculate the planned value PV2 of the production number in consideration of the increase / decrease in the number of orders.

For the calculation of the planned value PV2, for example, a dedicated model (hereinafter referred to as “planning model”) is used. That is, the planning model accepts the input of the predicted value PV1 and the order information and outputs the planned value PV2. Various known techniques related to statistics or machine learning can be used to generate a planning model. Detailed description of these techniques will be omitted.

Alternatively, for example, a predetermined mathematical formula is used for calculating the planned value PV2. That is, the formula includes one or more variables. The individual variables include the predicted value PV1, the value calculated based on the predicted value PV1, the value indicating the number of orders, the value calculated based on the value indicating the number of orders, or the predicted value PV1 and the value indicating the number of orders. The calculated value is entered. As a result, the planned value PV2 is calculated.

In addition, various known techniques can be used to calculate the planned value PV2. Detailed description of these techniques will be omitted.

Hereinafter, the processes executed by the economic indicator prediction device 100 may be collectively referred to as “economic indicator prediction processing, etc.”. The individual processes included in the economic indicator prediction process and the like are the same as those described in the first embodiment. Therefore, detailed description of these processes will be omitted. In addition, the processing executed by the order information acquisition unit may be collectively referred to as "order information acquisition processing". In addition, the processing executed by the production planning unit 52 may be collectively referred to as "production planning processing".

Next, with reference to FIG. 23, the hardware configuration of the main part of the production planning apparatus 300 will be described.

As shown in FIG. 23A, the production planning apparatus 300 has a processor 61 and a memory 62. The memory 62 stores a program for realizing the functions of the economic indicator prediction device 100, the order information acquisition unit 51, and the production planning unit 52. When the processor 61 reads out and executes such a program, the functions of the economic indicator prediction device 100, the order information acquisition unit 51, and the production planning unit 52 are realized.

Alternatively, as shown in FIG. 23B, the production planning apparatus 300 has a processing circuit 63. In this case, the functions of the economic indicator prediction device 100, the order information acquisition unit 51, and the production planning unit 52 are realized by the dedicated processing circuit 63.

Alternatively, the production planning apparatus 300 has a processor 61, a memory 62, and a processing circuit 63 (not shown). In this case, some of the functions of the economic indicator prediction device 100, the order information acquisition unit 51, and the production planning unit 52 are realized by the processor 61 and the memory 62, and the remaining functions are realized by the processing circuit 63. To.

The processor 61 is composed of one or a plurality of processors. The individual processors use, for example, CPUs, GPUs, microprocessors, microcontrollers or DSPs.

The memory 62 is composed of one or a plurality of non-volatile memories. Alternatively, the memory 62 is composed of one or more non-volatile memories and one or more volatile memories. That is, the memory 62 is composed of one or a plurality of memories. The individual memory uses, for example, a semiconductor memory, a magnetic disk, an optical disk, a magneto-optical disk, or a magnetic tape. More specifically, the individual volatile memories use, for example, RAM. Further, as the individual non-volatile memory, for example, ROM, flash memory, EPROM, EEPROM, solid state drive, hard disk drive, flexible disk, compact disk, DVD, Blu-ray disk or mini disk are used.

The processing circuit 63 is composed of one or a plurality of digital circuits. Alternatively, the processing circuit 63 is composed of one or a plurality of digital circuits and one or a plurality of analog circuits. That is, the processing circuit 63 is composed of one or a plurality of processing circuits. The individual processing circuits use, for example, an ASIC, PLD, FPGA, SoC or system LSI.

Next, the operation of the production planning apparatus 300 will be described with reference to the flowchart of FIG. 24.

First, the economic indicator prediction device 100 executes economic indicator prediction processing and the like (step ST21). Next, the order information acquisition unit 51 executes the order information acquisition process (step ST22). Next, the production planning unit 52 executes the production planning process (step ST23).

The execution order of the economic indicator prediction process and the order information acquisition process is not limited to the example shown in FIG. 24. The economic indicator prediction process or the like may be executed before the production planning process is executed. The order information acquisition process may be executed before the production planning process is executed.

Next, a modified example of the production planning apparatus 300 will be described with reference to FIG. 25.

As shown in FIG. 25, the production planning device 300 may have the economic indicator prediction device 100a instead of the economic indicator prediction device 100. The economic indicator prediction device 100a is the same as that described in the second embodiment. Therefore, a detailed description of the economic indicator prediction device 100a will be omitted.

As the economic indicator prediction device 100 in the production planning device 300 shown in FIG. 22, various modifications similar to those described in the first embodiment can be adopted. Further, as the economic indicator prediction device 100a in the production planning device 300 shown in FIG. 25, various modifications similar to those described in the second embodiment can be adopted.

As described above, the production planning device 300 according to the third embodiment includes the economic index prediction device 100 or the economic index prediction device 100a, and the order information acquisition unit 51 that acquires order information indicating the number of orders for the target product. It includes a production planning unit 52 that calculates a planned value PV2 of the number of production of the target product using the value PV1 and the order information. By using the predicted value PV1, it is possible to calculate the planned value PV2 of the production number in consideration of the increase in the number of orders due to the addition of orders and the decrease in the number of orders due to the cancellation of orders.

It should be noted that, within the scope of the present invention, any combination of embodiments can be freely combined, any component of each embodiment can be modified, or any component can be omitted in each embodiment. ..

Further, in each embodiment, the operation information acquisition unit acquires operation information indicating the operation state of other equipment or the like in place of or in addition to acquiring the operation information indicating the operation state of the target elevator. There may be. For example, the operation information acquisition unit may acquire operation information indicating the operation state of the production equipment (hereinafter referred to as “target production equipment”) installed in a factory or the like. The forecasting model may be one that accepts the input of the acquired operation information and outputs the forecasted value of the target economic indicator. The operation information in this case is, for example, at least one of information indicating the operating rate of the target production equipment, information indicating the production amount by the target production equipment, or information indicating the amount of parts used for production by the target production equipment. It may include one.

Further, in each embodiment, the history information acquisition unit may acquire the first history information corresponding to the operation information acquired by the operation information acquisition unit. That is, the history information acquisition unit acquires, in addition to, instead of acquiring the first history information indicating the history of the operating state of the target elevator, the first history information indicating the history of the operating state of other devices and the like. It may be a thing. For example, the history information acquisition unit may acquire the first history information indicating the history of the operating state of the target production equipment. The prediction model generation unit may use the acquired first history information to generate a prediction model.

The economic indicator prediction device, the production planning device, the prediction model generation device, and the prediction model generation method of the present invention can be used, for example, for planning the production number of the target product.

11 Operation information acquisition unit, 12 Data preprocessing unit, 13 Indicator information acquisition unit, 14 Economic index prediction unit, 15 Additional information acquisition unit, 16 Elevator selection unit, 21 History information acquisition unit, 22 Data preprocessing unit, 23 For prediction Model generation unit, 24 additional information acquisition unit, 25 elevator selection unit, 31 processor, 32 memory, 33 processing circuit, 41 processor, 42 memory, 43 processing circuit, 51 order information acquisition unit, 52 production planning unit, 61 processor, 62 Memory, 63 processing circuit, 100, 100a economic index prediction device, 200, 200a prediction model generation device, 300 production planning device.

Claims (18)

An operation information acquisition unit that acquires operation information indicating the operation status of the target elevator,
An economic indicator prediction unit that predicts the target economic indicator using a forecasting model that accepts the input of the operation information and outputs the predicted value of the target economic indicator.
An economic indicator forecaster equipped with. The operation information includes information indicating the number of times the target elevator has been operated, information indicating the mileage of the target elevator, information indicating the traveling time of the target elevator, information indicating the load weight of the target elevator, or boarding of the target elevator. The economic indicator predictor according to claim 1, wherein the device includes at least one of information indicating the number of persons. The additional information acquisition unit that acquires additional information,
It is provided with an elevator selection unit that selects one or more elevators corresponding to the target economic index among the plurality of elevators as the target elevator by using the additional information.
The additional information is information indicating an area including each of the plurality of facilities, information indicating the type of each of the plurality of facilities, or the above-mentioned information indicating the type of each of the plurality of facilities when the plurality of elevators are installed in the plurality of facilities. The economic indicator predictor according to claim 1, further comprising at least one of information indicating the use of each of the plurality of elevators. A data pre-processing unit that executes pre-processing for the operation information is provided.
The operation information is composed of time series data, and is composed of time series data.
The economic indicator prediction device according to claim 1, wherein the preprocessing includes a process for removing seasonal fluctuations in the time series data. A data pre-processing unit that executes pre-processing for the operation information is provided.
The operation information is composed of time series data, and is composed of time series data.
The economic indicator prediction device according to claim 1, wherein the preprocessing includes a process of removing an abnormal value included in the time series data. The economic indicator prediction device according to claim 1, wherein the forecasting model is based on a time lag between the history of the operating state and the history of the target economic indicator. The forecasting model is characterized in that it is based on a time-series correlation between the first time-series data showing the history of the operating state and the second time-series data showing the history of the target economic indicator. The economic indicator prediction device according to claim 6. It is equipped with an index information acquisition unit that acquires index information indicating the target economic indicator.
The economic indicator prediction device according to claim 1, wherein the prediction model accepts inputs of the operation information and the index information and outputs the predicted value. The economic indicator prediction device according to claim 1 and
The order information acquisition department that acquires order information indicating the number of orders for the target product,
A production planning unit that calculates a planned value of the number of production of the target product using the predicted value and the order information.
Production planning equipment. A history information acquisition unit that acquires the first history information indicating the history of the operating state of the target elevator and the second history information indicating the history of the target economic indicator.
A prediction model generation unit that generates a prediction model using the first history information and the second history information is provided.
The forecasting model is a forecasting model generator, characterized in that it receives an input of operating information indicating the operating state and outputs a predicted value of the target economic indicator. The first history information includes information indicating the history of the number of times the target elevator has been operated, information indicating the history of the travel distance of the target elevator, information indicating the history of the travel time of the target elevator, and the load weight of the target elevator. The predictive model generation device according to claim 10, further comprising at least one of information indicating a history or information indicating a history of the number of passengers of the target elevator. The additional information acquisition unit that acquires additional information,
It is provided with an elevator selection unit that selects one or more elevators corresponding to the target economic index among the plurality of elevators as the target elevator by using the additional information.
When the plurality of elevators are installed in a plurality of facilities, the additional information includes information indicating an area including each of the plurality of facilities, information indicating the type of each of the plurality of facilities, or the above-mentioned information. The predictive model generator according to claim 10, further comprising at least one of information indicating the use of each of the plurality of elevators. A data pre-processing unit that executes pre-processing for the first history information is provided.
The first history information is composed of the first time series data.
The prediction model generation device according to claim 10, wherein the pre-processing includes a processing for removing seasonal fluctuations in the first time-series data. A data pre-processing unit that executes pre-processing for the first history information is provided.
The first history information is composed of the first time series data.
The predictive model generation device according to claim 10, wherein the preprocessing includes a process of removing an abnormal value included in the first time series data. The prediction model generation device according to claim 10, wherein the prediction model generation unit generates the prediction model based on the amount of time lag between the history of the operating state and the history of the target economic indicator. .. The first history information is composed of the first time series data.
The second history information is composed of second time series data.
The prediction model according to claim 15, wherein the prediction model generation unit generates the prediction model based on the time-series correlation between the first time-series data and the second time-series data. Model generator. The prediction model generation device according to claim 10, wherein the prediction model accepts input of operation information and index information indicating the target economic indicator and outputs the prediction value. A step in which the history information acquisition unit acquires the first history information indicating the history of the operating state of the target elevator and the second history information indicating the history of the target economic indicator.
The prediction model generation unit includes a step of generating a prediction model using the first history information and the second history information.
The prediction model is a method for generating a prediction model, characterized in that it receives an input of operation information indicating the operation state and outputs a predicted value of the target economic index.

Images