JP7198700B2 - Decision support device, decision support program and decision support method - Google Patents

Description

The present invention relates to a decision support device, a decision support program, and a decision support method.

In industries that handle inventories, understanding future demand will help control excess inventories and reduce opportunity losses. Currently, it is possible to predict future demand using statistical methods, machine learning, etc., based on data such as past shipment results, and research is being conducted to improve the accuracy of this prediction. In particular, high prediction accuracy is required in order to limit losses when placing orders or purchasing based on the results of demand predicted using statistical methods, machine learning, or the like.

Note that Patent Literature 1 describes a demand forecast model evaluation method.

JP-A-10-228463

Absolute value error and coefficient of determination are general indicators for evaluating the accuracy of prediction results obtained using statistical methods and machine learning. Indicators have no business meaning, so it is difficult to decide which indicator to choose for operations such as ordering and purchasing. However, it is unknown.

Companies that handle inventories are required to make appropriate decisions from a practical point of view on a product-by-product basis as to whether or not to use the results of predictions based on statistical methods and machine learning for ordering and purchasing. Until now, there has been no information processing means capable of providing useful information for judgment.

The present invention has been made in view of the above, and from a practical point of view regarding whether or not to adopt prediction results by statistical methods, machine learning, etc. in ordering, purchasing, etc., and for each product. It is an object of the present invention to provide a decision support device, a decision support program, and a decision support method capable of providing useful information for appropriate decisions.

In order to solve the above-described problems and achieve the object, a decision support device according to the present invention is a decision support device comprising a control unit, wherein the control unit predicts product demand obtained by a demand prediction model. and the actual demand value of the product, an index for evaluating the degree of suppression of opportunity loss, which is weighted according to the magnitude relationship between the predicted demand value and the actual demand value and the excess Calculation means for calculating the value of either one or both of the excess inventory indicators, which are indicators for evaluating the degree of inventory control and are weighted according to the magnitude relationship between the demand forecast value and the actual demand value characterized by comprising

In addition, in the judgment support device according to the present invention, the control unit causes the display unit to display the value calculated by the calculation means and the graph showing the transition of the predicted demand value and the actual demand value used by the calculation means. A control means may further be provided.

Further, a judgment support program according to the present invention uses a product demand forecast value and a product demand actual value obtained by a demand forecast model to be executed by the control unit of an information processing apparatus including a control unit. , an index for evaluating the degree of suppression of opportunity loss, which is weighted according to the magnitude relationship between the demand forecast value and the actual demand value, and an index for evaluating the degree of suppression of excess inventory and a calculating step of calculating the value of either or both of the excess inventory indexes weighted according to the magnitude relationship between the predicted demand value and the actual demand value.

In addition, a decision support method according to the present invention is executed by the control unit of an information processing apparatus having a control unit, and uses a product demand forecast value and a product demand actual value obtained by a demand forecast model to determine whether an opportunity is available. It is an index for evaluating the degree of loss control, and is an opportunity loss index weighted according to the magnitude relationship between the demand forecast value and the actual demand value, and an index for evaluating the degree of control of excess inventory. and a calculating step of calculating the value of either or both of the excess inventory indexes weighted according to the magnitude relationship between the predicted demand value and the actual demand value.

The present invention provides information that is useful for making appropriate judgments on a product-by-product basis from a practical point of view as to whether or not to adopt prediction results based on statistical techniques, machine learning, etc., in ordering, purchasing, etc. It has the effect of being able to

FIG. 1 is a diagram showing an example of the configuration of a judgment support device. FIG. 2 is a diagram showing an example of the prediction data 106a. FIG. 3 is a diagram showing an example of the relationship between prediction data and performance data used when explaining a calculation example of an opportunity loss index. FIG. 4 is a diagram showing an example of the relationship between forecast data and performance data used when explaining a calculation example of an excess inventory index. FIG. 5 is a diagram showing an example of a display mode when the index value and the graph are displayed together. FIG. 6 is a diagram showing an example of a display mode when the index value and the graph are displayed together. FIG. 7 is a diagram showing an example of the relationship between prediction data and performance data used when explaining an example of absolute value error calculation.

EMBODIMENT OF THE INVENTION Below, embodiment of the judgment assistance apparatus, the judgment assistance program, and the judgment assistance method which concern on this invention is described in detail based on drawing. It should be noted that the present invention is not limited by this embodiment.

[1. composition]
An example of the configuration of the decision support device according to this embodiment will be described with reference to FIG. 1 and the like. FIG. 1 is a block diagram showing an example of the configuration of a decision support device.

The judgment support device 100 is a commercially available desktop personal computer. Note that the judgment support device 100 is not limited to a stationary information processing device such as a desktop personal computer, and may be a portable type such as a commercially available notebook personal computer, PDA (Personal Digital Assistants), a smart phone, or a tablet personal computer. It may be an information processing device.

The decision support device 100 includes a control unit 102 , a communication interface unit 104 , a storage unit 106 and an input/output interface unit 108 . Each unit included in the determination support device 100 is communicably connected via an arbitrary communication path.

The communication interface unit 104 communicably connects the determination support device 100 to the network 300 via a communication device such as a router and a wired or wireless communication line such as a dedicated line. The communication interface unit 104 has a function of communicating data with another device via a communication line. Here, the network 300 has a function of connecting the determination support device 100 and the server 200 so that they can communicate with each other, and is, for example, the Internet or a LAN (Local Area Network). Note that the data stored in the storage unit 106 may be stored in the server device 200, for example.

An input device 112 and an output device 114 are connected to the input/output interface section 108 . The output device 114 can be a monitor (including a home television), a speaker, or a printer. The input device 112 can be a keyboard, a mouse, a microphone, or a monitor that realizes a pointing device function in cooperation with a mouse. Note that, hereinafter, the output device 114 may be referred to as the monitor 114 and the input device 112 may be referred to as the keyboard 112 or the mouse 112 .

The storage unit 106 stores various databases, tables, files, and the like. The storage unit 106 stores a computer program for performing various processes by giving commands to a CPU (Central Processing Unit) in cooperation with an OS (Operating System). As the storage unit 106, for example, memory devices such as RAM (Random Access Memory) and ROM (Read Only Memory), fixed disk devices such as hard disks, flexible disks, and optical disks can be used.

In the storage unit 106, demand forecast values (demand forecast quantities) of products (for example, SKU (Stock Keeping Unit: minimum unit of inventory management)) obtained by the demand forecast model are stored by time period (for example, sorted, hourly, hourly Forecast data 106a (see FIG. 2) stored in bands, days, weeks, months, periods, years, etc., and results in which actual demand values (actual demand quantities) for products are stored by period Data 106b and the like are stored. Note that the data structure of the performance data 106b may be the same as that of the prediction data 106a, for example. The demand forecast model is conceptually a mathematical technique (for example, statistical technique, machine learning, etc.) for predicting future demand based on past demand.

The control unit 102 is a CPU or the like that controls the determination support device 100 in an integrated manner. The control unit 102 has an internal memory for storing a control program such as an OS, a program defining various processing procedures, required data, and the like, and performs various information processing based on these stored programs. Run.

The control unit 102 functionally and conceptually includes a calculation unit 102a and a display control unit 102b.

The calculation unit 102a performs information processing for calculating the value of either one or both of the opportunity loss index and the excess inventory index using the predicted demand value for the product obtained by the demand prediction model and the actual demand value for the same product. It is a means. Opportunity loss index is an index for evaluating the degree of suppression of opportunity loss, depending on the magnitude relationship between the demand forecast value and the actual demand value (specifically, when the forecast value is smaller than the actual value) A weight is imposed. In addition, the excess inventory index is an index for evaluating the degree of suppression of excess inventory, and it depends on the magnitude relationship between the demand forecast value and the actual demand value (specifically, when the forecast value is greater than the actual value ) is the one to which the weighting is imposed.

The display control unit 102b causes the monitor 114 (corresponding to the display unit of the present invention) to display the value calculated by the calculation unit 102a, the demand forecast value used by the calculation unit 102a, and the graph showing the transition of the actual demand value. processing means. Note that the display control unit 102b is not limited to displaying on the monitor 114, and may perform printing via a printer, for example.

[2. Specific example of processing]
Here, a specific example of the processing executed by the judgment support device 100 will be described with reference to each drawing. In addition, in this description, the following 1. and 2. is assumed.
1. A company carries out ordering operations in units of SKUs.
2. The forecast data 106a stores the past daily demand forecast values (quantity) forecasted in the past by the demand forecasting model that forecasts the demand amount in SKU units (see FIG. 2). is stored in the performance data 106b.

First, the calculation unit 102a calculates demand forecast values by SKU and by day stored in the forecast data 106a and From the actual demand values by SKU and by day stored in the actual data 106b, those that satisfy the extraction conditions are extracted, and the extracted demand forecast values by SKU and by day and actual demand by SKU and by day are extracted. The values are aggregated by SKU and period (step SA1). If no extraction condition is specified, all the data stored in the prediction data 106a and the performance data 106b are extracted.

Next, the calculation unit 102a uses the SKU-by-SKU/period-by-SKU demand forecast value and the demand actual demand value aggregated in step SA1 to calculate an opportunity loss index for each SKU defined by Equation 1 below and Equation 2 below. Both values of the SKU unit excess inventory index defined in are calculated for each SKU (step SA2).

Opportunity loss index _*1 =|Demand forecast value－Actual demand value| _*2 ×Total sum of weighting factor _*3
... (Formula 1)
*1: The smaller the indicator value, the better (that is, the higher the validity of the demand forecast value).
*2: || is an absolute value symbol. Also, the demand forecast value and actual demand value to be substituted are those of the same period.
*3: Regarding the weighting factor, if the predicted value is greater than the actual value, it is considered to be acceptable from the perspective of opportunity loss, and a constant of 1 is set, which means that no penalty is imposed. If it is smaller than the actual value, it is considered unacceptable from the viewpoint of opportunity loss, and a constant greater than 1 (for example, 2) is set, which means that a penalty is imposed (see Figure 3). .

Excess inventory index _*1 = | Demand forecast value - Actual demand value | _*2 × Total sum of weighting factor _*4
... (Formula 2)
*4: Regarding the weighting factor, if the predicted value is greater than the actual value, it is considered unacceptable from the perspective of excess inventory, and a constant greater than 1 (such as 2) means that a penalty is imposed. ) is set, and if the predicted value is less than the actual value, it is considered acceptable from the perspective of excess inventory, and a constant of 1 is set, meaning that no penalty is imposed (see Figure 4).

Next, the display control unit 102b displays an SKU-by-SKU display showing changes in forecast and actual values for each period based on the SKU-by-SKU/period-by-period demand forecast value and actual demand value used in the calculation in step SA2. A graph (for example, a line graph) is generated, and a screen MA as shown in FIG. (For each SKU, the value of the opportunity loss indicator, the value of the excess inventory indicator, and a graph of a display mode that can be specified items, etc.) are arranged in a list format, and the generated screen MA is displayed on the monitor 114 (step S3). Note that when each graph in the screen MA displayed on the monitor 114 is specified by, for example, cursor alignment or click operation, the display control unit 102b controls the screen on which the enlarged graph is displayed (for example, screen MB or screen MC, etc.) may be generated and displayed on the monitor 114 (see FIG. 5).

In this way, by displaying the index and the graph together, it is possible to compensate for human judgment that is difficult to index. For example, an operator who sees the display contents of FIG. In addition, it is possible to decide which indicators should be focused on for each product category, taking into consideration the characteristics of the product category (microwave oven, smartphone, etc.). Specifically, when the operator compares the graph and the index, for example, "It is true that the product whose forecast tends to be smaller than the actual has a high opportunity loss index, and the product whose forecast tends to be larger than the actual is excessive." The inventory index is high. This fits my sense of the business.”, and you can go on to say, “Okay, then about the microwave oven, where inventory takes up a lot of space. We will focus on the excess inventory index, and for smartphones that have little space to secure inventory and are expected to generate a large amount of sales, we will focus on the opportunity loss index and decide on SKUs for which demand forecast values are likely to be effective.” can also think.

In addition, by displaying the index and the graph together, the operator can determine the product for which the demand forecast value is used when ordering or purchasing from the selected product group and index. For example, the operator who sees the display contents of FIG. 6 displayed by the processing of the display control unit 102b can decide the threshold value of the index value that is permissible with respect to the selected product group or index, or set the threshold value. Based on this, it is possible to decide which products to use the demand forecast value for when ordering or purchasing. Specifically, the operator compares the graph and the index with respect to the product group of microwave ovens, and says, for example, "For products with an excess inventory index of approximately 200 or more, it is likely to become an excess inventory if the demand forecast value is used. In addition, it is likely to be an opportunity loss if the demand forecast value is used for products with an opportunity loss index of 150 or more.Therefore, 200 was set as the allowable threshold for the excess inventory index and 150 was set as the allowable threshold for the opportunity loss index. Above, for products below each threshold, it would be good to use the demand forecast value _*5 ." Since it takes a lot of space to secure inventory, we will use the excess inventory index.Therefore, the demand forecast value for microwave ovens (large, white) and microwave ovens (medium, white) whose excess inventory index is 200 or less is Let's decide to place orders and purchases using
*5: The marks “○” and “X” in FIG. Such a mark (for example, ○ meaning "adopted" or "X" meaning "not adopted") for identifying the judgment result by the operator as to whether or not to adopt the demand forecast value in operations such as importing , may be displayed on the actual screen by being designated by the operator.

[3. Summary of this embodiment]
As described above, according to the present embodiment, it is useful for making appropriate judgments on a product-by-product basis from a practical point of view as to whether or not to adopt prediction results based on statistical methods, machine learning, etc. in ordering, purchasing, etc. can provide information. In this embodiment, the difference between the demand forecast value and the actual demand value is used as an example of the opportunity loss index and the excess inventory evaluation index (see the above formulas 1 and 2), but the demand forecast value and the ratio of the actual demand value may be used. Further, as shown in FIG. 6, the display control unit 102b displays an enlarged graph when an area in which the characters "GRAPH" are written in the screen MD is designated by, for example, cursor alignment or a click operation. Screens ME to MH of all displayed SKUs may be generated and displayed side by side on the monitor 114 .

Here, the problems hitherto and the effects of the present embodiment will be described below.

Until now, as a general indicator for evaluating the accuracy of prediction results obtained using statistical methods, machine learning, etc., the absolute value error (see Fig. 7), which is evaluated by the difference between the overall predicted value and the actual value ) and commodities that fluctuate greatly, there are effective coefficients of determination, but the evaluation of accuracy varies depending on the index, and these indices have no business meaning. In addition, even if an indicator is selected, it is of course unclear whether the indicator is practically appropriate.

In general, when AI (artificial intelligence) is used for demand forecasting, if there are many kinds of SKUs to be handled, the accuracy of the forecast varies depending on the SKUs. On the other hand, when ordering and purchasing work based on forecasts, the person in charge selects only SKUs with stable forecast accuracy and a high level, even though they do not have knowledge of statistics. sea bream. Therefore, it is desirable that the accuracy evaluation index for selecting such SKUs can be easily understood even without knowledge of statistics, etc., in order to avoid the risk that the demand forecast will not be fully utilized.

In addition, if there are many products, it is a difficult task to see the predicted value and the actual value for each product numerically. Therefore, it is unknown for which product the predicted value is correct. Also, the definition of "prediction is correct" varies depending on the customer's issue or need. In particular, "hitting" is determined intuitively, and the definition is not clear. As a customer, we want to prevent excess inventory and lost opportunities by considering the unit price, warehouse size, and company strategy.

In order to solve these problems, according to the present embodiment, new accuracy evaluation indicators, such as an "opportunity loss index" and an "excess inventory index", which have practical significance in evaluating the accuracy of demand forecasting, are presented. By visualizing the index for each product, it is possible to support the person in charge in deciding the accuracy evaluation index and selecting SKUs whose demand forecast values are used as a reference when purchasing and ordering. Such ``presentation of indicators with practical meaning'' and ``easy-to-understand visualization'' enables the person in charge to confirm the reliability of the accuracy evaluation index. As a result, the person in charge can determine the accuracy evaluation index with a sense of convincing even if they do not have specialized knowledge of machine learning, and can select the products that refer to the predicted value. In addition, by presenting new accuracy evaluation indicators such as "opportunity loss index" and "excess inventory index" that have practical meaning, and by visualizing the indicators for each product, we will contribute to the prevention of excess inventory and opportunity loss. be able to.

[4. Other embodiments]
The present invention may be implemented in various different embodiments other than the embodiments described above within the scope of the technical idea described in the claims.

For example, among the processes described in the embodiments, all or part of the processes described as being automatically performed can be manually performed, or all of the processes described as being manually performed Alternatively, some can be done automatically by known methods.

In addition, unless otherwise specified, the processing procedures, control procedures, specific names, information including parameters such as registration data and search conditions for each process, screen examples, and database configurations shown in this specification and drawings can be changed arbitrarily.

Also, with regard to the decision support device 100, each illustrated component is functionally conceptual, and does not necessarily need to be physically configured as illustrated.

For example, all or any part of the processing functions of the decision support device 100, particularly the processing functions performed by the control unit, may be realized by a CPU and a program interpreted and executed by the CPU. Alternatively, it may be implemented as hardware by wired logic. The program is recorded on a non-temporary computer-readable recording medium containing programmed instructions for causing the information processing apparatus to execute the processing described in the present embodiment, and the determination support apparatus as necessary. 100 mechanically read. That is, a storage unit such as a ROM or HDD (Hard Disk Drive) stores a computer program for giving commands to the CPU in cooperation with the OS to perform various processes. This computer program is executed by being loaded into the RAM and constitutes a control section in cooperation with the CPU.

Also, this computer program may be stored in an application program server connected to the decision support apparatus 100 via any network, and all or part of it can be downloaded as necessary. be.

Also, the program for executing the processing described in this embodiment may be stored in a non-temporary computer-readable recording medium, or may be configured as a program product. Here, the term "recording medium" refers to memory cards, USB (Universal Serial Bus) memories, SD (Secure Digital) cards, flexible disks, magneto-optical disks, ROMs, EPROMs (Erasable Programmable Read Only Memory), EEPROMs (registered (trademark) (Electrically Erasable and Programmable Read Only Memory), CD-ROM (Compact Disk Read Only Memory), MO (Magneto-Optical disk), DVD (Digital Versatile Disk), and Disc (registered trademark) such as Blu- shall include any "portable physical medium".

A "program" is a data processing method written in any language or writing method, regardless of the format such as source code or binary code. In addition, the "program" is not necessarily limited to a single configuration, but is distributed as multiple modules or libraries, or cooperates with a separate program represented by the OS to achieve its function. Including things. It should be noted that well-known configurations and procedures can be used for the specific configuration and reading procedure for reading the recording medium in each device shown in the embodiments, the installation procedure after reading, and the like.

The various databases stored in the storage unit are storage means such as memory devices such as RAM and ROM, fixed disk devices such as hard disks, flexible disks, and optical disks. It stores programs, tables, databases, files for web pages, and so on.

Further, the judgment support device 100 may be configured as an information processing device such as a known personal computer or workstation, or may be configured as the information processing device to which arbitrary peripheral devices are connected. Further, the determination support device 100 may be implemented by installing software (including programs, data, etc.) that causes the device to implement the processing described in the present embodiment.

Furthermore, the specific forms of distribution and integration of devices are not limited to those shown in the figures, and all or part of them can be functionally or physically arranged in arbitrary units according to various additions or functional loads. It can be distributed and integrated. In other words, the embodiments described above may be arbitrarily combined and implemented, or the embodiments may be implemented selectively.

The present invention is particularly useful in various industries and types of business dealing with inventory.

100 judgment support device 102 control unit
102a calculator
102b display control unit 104 communication interface unit 106 storage unit
106a prediction data
106b performance data 108 input/output interface unit 112 input device 114 output device 200 server 300 network

Claims (4)

A judgment support device comprising a control unit,
The control unit
An indicator for evaluating the degree of suppression of opportunity loss using the forecasted demand value for a product obtained by a demand forecast model and the actual demand value for the product, and according to the magnitude relationship between the forecasted demand value and the actual demand value Either an opportunity loss index that is weighted according to the actual demand or an excess inventory index that is weighted according to the magnitude relationship between the demand forecast value and the actual demand value, which is an index for evaluating the degree of suppression of excess inventory calculating means for calculating one or both values;
A judgment support device characterized by: The control unit
further comprising display control means for causing a display unit to display the value calculated by the calculation means and a graph showing changes in the predicted demand value and the actual demand value used by the calculation means;
The decision support device according to claim 1, characterized by: For causing the control unit of an information processing device comprising a control unit to execute
An indicator for evaluating the degree of suppression of opportunity loss using the forecasted demand value for a product obtained by a demand forecast model and the actual demand value for the product, and according to the magnitude relationship between the forecasted demand value and the actual demand value Either an opportunity loss index that is weighted according to the actual demand or an excess inventory index that is weighted according to the magnitude relationship between the demand forecast value and the actual demand value, which is an index for evaluating the degree of suppression of excess inventory A decision support program comprising a calculation step of calculating one or both values. Executed by the control unit of an information processing device comprising a control unit,
An indicator for evaluating the degree of suppression of opportunity loss using the forecasted demand value for a product obtained by a demand forecast model and the actual demand value for the product, and according to the magnitude relationship between the forecasted demand value and the actual demand value Either an opportunity loss index that is weighted according to the actual demand or an excess inventory index that is weighted according to the magnitude relationship between the demand forecast value and the actual demand value, which is an index for evaluating the degree of suppression of excess inventory A decision support method, comprising a calculating step of calculating one or both values.

Images