Data generation device, data generation method, and data generation program

Abstract

To provide a data generation device, a data generation method, and a data generation program that can present a sufficient number of patterns related to item selection used when performing highly accurate layout design by simulation or the like.SOLUTION: A data generation device, a data generation method, and a data generation program relate to a controller connected to an input unit and an output unit. For a pattern where one or more types of items are selected from a plurality of items, a first probability that an item of interest is included in the pattern is obtained for each item via the input unit. A second probability that a first item and a second item, which are two distinct items of interest, are both included in the pattern is obtained for each of a part of or all combinations of the first item and the second item. Then, a pattern is generated on the basis of the first probability and the second probability, and the pattern is output via the output unit.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a data generation device, a data generation method, and a data generation program.

Background Art

[0002] Patent Document 1 discloses a system, a method, and a program for assisting in the layout design of a logistics warehouse by movably attaching and displaying storage equipment icons containing shipped products on a warehouse floor plan defined on a display screen.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] According to the technique described in Patent Document 1, in order to perform the layout design of a logistics warehouse, data on the conveyance pattern when products are actually shipped from storage equipment is required. However, there is a problem in that the number of data on the conveyance pattern may be insufficient in order to achieve a highly accurate layout design by means of simulation or the like.

[0005] The present disclosure has been made in view of the above problems. An object thereof is to provide a data generation device, a data generation method, and a data generation program capable of presenting a sufficient number of patterns related to the selection of items to be used when performing a highly accurate layout design by means of simulation or the like.

Means for Solving the Problems

[0006] The data generation device, data generation method, and data generation program according to the present disclosure relate to a controller connected to an input unit and an output unit. Regarding a pattern of selecting one or more items from a plurality of items, the first probability that an item of interest is included in the pattern is acquired for each item via the input unit. The second probability that a first item and a second item, which are two different items of interest, are both included in the pattern is acquired for each combination of some or all of the first item and the second item via the input unit. Then, a pattern is generated based on the first probability and the second probability, and the pattern is output via the output unit.

[0007] The controller may calculate a selection probability that an item is selected by the pattern and a conditional probability that the second item is selected when the first item is selected by the pattern based on the first probability and the second probability. And a pattern may be generated based on the selection probability and the conditional probability.

[0008] The controller may determine whether the first item related to the selection probability is included in the pattern based on the selection probability. When the first item is selected, it may be determined whether to select the second item related to the conditional probability based on the conditional probability related to the selected first item. And a pattern may be generated based on the presence or absence of the selection of the first item and the presence or absence of the selection of the second item.

[0009] The controller may calculate the product of the selection probability that the first item is selected and the conditional probability that the second item is selected when the first item is selected. It may also change the first item over a plurality of items and sum the values of the products. It may change the selection probability and the conditional probability in a direction such that the total value obtained by summing the values of the products approaches the first probability that the second item is included in the pattern. Further, the controller may calculate the product (the first product) of the selection probability that the first item is selected and the conditional probability that the second item is selected when the first item is selected. Additionally, it may calculate the product (the second product) of the selection probability that the second item is selected and the conditional probability that the first item is selected when the second item is selected. Then, it may change the selection probability and the conditional probability in a direction such that the sum of the first product and the second product approaches the second probability that both the first item and the second item are included in the pattern.

[0010] The items may be arranged at a plurality of different locations included in the warehouse or area. And the pattern may describe the conveyance of items from a location or the conveyance of items to a location.

[0011] Based on the generated pattern, it may determine the arrangement of items at a location.

Advantages of the Invention

[0012] According to the present disclosure, it is possible to provide a data generation device, a data generation method, and a data generation program that can present a sufficient number of patterns related to the selection of items for use in performing highly accurate layout design by simulation or the like.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Embodiments for Carrying Out the Invention

[0014] Hereinafter, some exemplary embodiments will be described with reference to the drawings. In the drawings, the same reference numerals are given to the common parts in each figure, and duplicate explanations are omitted.

[0015] [Configuration of Data Generation Device] FIG. 1 is a block diagram showing the configuration of a data generation device according to an embodiment of the present disclosure. As shown in FIG. 1, the data generation device 20 includes an input unit 21, an output unit 23, and a controller 25. Further, the data generation device 20 may include an operation unit 27 and a database 29. The controller 25 is connected so as to be communicable with the input unit 21, the output unit 23, the operation unit 27, and the database 29.

[0016] In addition, the input unit 21, the output unit 23, the operation unit 27, and the database 29 may be provided in the data generation device 20 itself, or may be installed outside the data generation device 20 and connected to the data generation device 20.

[0017] The input unit 21 acquires a first probability and a second probability regarding a pattern of selecting one or more items from a plurality of items. Here, the "first probability" is the probability that the item of interest is included in the pattern among the plurality of items. The "second probability" is the probability that a first item and a second item, which are two different items of interest among the plurality of items, are both included in the pattern.

[0018] Note that the input unit 21 may obtain a "conditional second probability" that the second item of interest is included in the same pattern when the first item, which is the item of interest, is included in the pattern, instead of the "second probability". In this case, the input unit 21 may calculate the "second probability" from the "first probability" and the "conditional second probability".

[0019] Alternatively, the input unit 21 may obtain information indicating a pattern. Then, based on the plurality of obtained patterns, the first probability, the second probability, and the conditional second probability may be calculated.

[0020] Here, an "item" is an object to be selected described by a pattern generated by the data generation device 20. For example, the "item" may be arranged at a plurality of different locations included in a warehouse or an area. In this case, the "pattern" may describe the conveyance of items from a plurality of locations or the conveyance of items to a plurality of locations.

[0021] For example, the "item" may be a plurality of types of goods stored and managed in an inventory-type logistics center. Also, the "item" may be a product sold by a sales store such as a supermarket. Alternatively, the "item" may be an item displayed on an Internet site such as an e-commerce site, or an article sold, leased, etc. through the selection of an item.

[0022] The output unit 23 outputs the pattern generated by the controller 25 described later. The generated pattern may be used for purposes such as simulation for performing a highly accurate layout design.

[0023] For example, the output unit 23 may be connected to a computer or the like (not shown) and output the generated pattern to the computer. Then, on the computer, an item placement determination method for determining the placement of items at a plurality of locations may be executed based on the generated pattern. For example, according to the item placement determination method, by placing items that are likely to be transported simultaneously by a pattern at the same location as much as possible, the movement cost between a plurality of locations during transportation is reduced. In addition, by placing items that are likely to be selected simultaneously by a pattern within the same page such as an Internet site, the load associated with page transitions is reduced.

[0024] The operation unit 27 is an input device through which the user of the data generation device 20 can perform operations. For example, the operation unit 27 is a keyboard, a mouse, a trackball, a touch panel, or the like. The operation unit 27 is not limited to the examples listed here. The operation content of the user input through the operation unit 27 is transmitted to the controller 25.

[0025] In addition, the operation unit 27 may acquire the above-described "first probability", "second probability", or "pattern" based on the user's operation. Further, the operation unit 27 may acquire a list of a plurality of items based on the user's operation. For example, selection may be performed on a plurality of items included in the list acquired by the operation unit 27.

[0026] The database 29 stores the "first probability", "second probability", and "pattern" acquired via the input unit 21 or the operation unit 27. Further, the database 29 may store the pattern generated by the controller 25.

[0027] The controller 25 is a general-purpose computer including a CPU (Central Processing Unit), a memory, and an input / output unit. A computer program (data generation program) for functioning as the data generation device 20 is installed in the controller 25. By executing the computer program, the controller 25 functions as a plurality of information processing circuits (251, 253, 255, 257) included in the data generation device 20.

[0028] In the present disclosure, an example of realizing a plurality of information processing circuits (251, 253, 255, 257) by software is shown. However, it is also possible to prepare dedicated hardware for executing each of the information processes shown below to configure the information processing circuits (251, 253, 255, 257). Further, the plurality of information processing circuits (251, 253, 255, 257) may be configured by individual hardware.

[0029] As shown in FIG. 1, the controller 25 includes, as a plurality of information processing circuits (251, 253, 255, 257), a first probability acquisition unit 251, a second probability acquisition unit 253, a calculation unit 255, and a generation unit 257.

[0030] The first probability acquisition unit 251 acquires the first probability for each item. Further, the first probability acquisition unit 251 may acquire information indicating a pattern and calculate the first probability based on the plurality of acquired patterns.

[0031] The second probability acquisition unit 253 acquires the second probability for each combination of some or all of two different items. That is, the second probability acquisition unit 253 may acquire the second probability for each combination of all combinations of two different items, or may acquire the second probability for each combination of some combinations of two different items. For example, the second probability acquisition unit 253 may not acquire the second probability for the combination of the first item and the second item that are less likely to be selected together. Further, the second probability acquisition unit 253 may acquire information indicating a pattern and calculate the second probability based on the plurality of acquired patterns.

[0032] In addition, the second probability acquisition unit 253 may acquire the above-mentioned conditional second probability. In this case, the second probability acquisition unit 253 may calculate the second probability from the first probability and the conditional second probability.

[0033] The calculation unit 255 calculates the selection probability that an item is selected by a randomly extracted pattern based on the first probability and the second probability. Further, the calculation unit 255 calculates the conditional probability that the second item is selected when the first item is selected by a randomly extracted pattern based on the first probability and the second probability.

[0034] For example, the calculation unit 255 calculates the product of the selection probability that the first item is selected and the conditional probability that the second item is selected when the first item is selected. The calculation unit 255 changes the first item over a plurality of items and sums the values of the products. Then, the calculation unit 255 changes the selection probability and the conditional probability in a direction that the total value obtained by summing the values of the products approaches the first probability that the second item is included in the pattern.

[0035] Further, the calculation unit 255 calculates the product (the first product) of the selection probability that the first item is selected and the conditional probability that the second item is selected when the first item is selected. Further, the calculation unit 255 calculates the product (the second product) of the selection probability that the second item is selected and the conditional probability that the first item is selected when the second item is selected. Then, the calculation unit 255 changes the selection probability and the conditional probability in a direction such that the sum of the first product and the second product approaches the second probability that both the first item and the second item are included in the pattern.

[0036] The processing of the calculation unit 255 will be described using mathematical formulas. Here, it is assumed that there are N types of items (N is a natural number of 1 or more). And, the first probability that the i-th item (i ∈ {1, ···, N}) is included in the pattern is denoted as "p i ", and the second probability that both the i-th item (i ∈ {1, ···, N}) and the j-th item (j ∈ {1, ···, N}) are included in the pattern is denoted as "p ij ".

[0037] Also, the selection probability that the i-th item is selected is denoted as "w i ", and the conditional probability that the j-th item is selected when the i-th item is selected is denoted as "c ij ".

[0038] In this case, the calculation unit 255 obtains the selection probability "w i " and the conditional probability "c ij " that satisfy the following simultaneous equations (mathematical formulas (1) to (6)).

Equation

Equation

Equation

Equation

Equation

Mathematics

[0039] When solving the system of simultaneous equations (mathematical expressions (1) to (6)), various methods can be used in the calculation unit 255. For example, the calculation unit 255 may use a non-linear equation solving algorithm such as the Newton-Raphson method.

[0040] Alternatively, the calculation unit 255 may solve an optimization problem that adopts an objective function based on the residuals (errors) of the mathematical expressions (1) to (6). At that time, some of the mathematical expressions (1) to (6) may be used as inequalities or constraint conditions.

[0041] In addition, the calculation unit 255 may perform non-negative weighting on the residuals of the mathematical expressions (1) and (2). For example, a large value of weighting may be given to the highly reliable first probability "p i " or the second probability "p ij ". If the first probability "p i " and the second probability "p ij " are estimators calculated from the patterns collected in the actual operation data, the number of patterns collected in the actual operation data may be used as the weighting as it is.

[0042] The mathematical expressions (1) to (6) do not necessarily have solutions. However, according to the method of solving the optimization problem, the calculation unit 255 can calculate an approximate solution that approximately satisfies the mathematical expressions (1) to (6).

[0043] In addition, the calculation unit 255 may obtain a plurality of solutions that approximately or exactly satisfy the mathematical expressions (1) to (6).

[0044] The generation unit 257 generates a pattern based on the first probability and the second probability. More specifically, the generation unit 257 obtains the selection probability and the conditional probability calculated based on the first probability and the second probability, and generates a pattern based on the selection probability and the conditional probability.

[0045] For example, based on the calculated selection probability, the generation unit 257 determines whether the first item related to the selection probability is included in the pattern. Then, when the first item is selected, the generation unit 257 determines whether to select the second item related to the conditional probability based on the conditional probability related to the selected first item. The generation unit 257 generates a pattern by selecting the determined first item and second item.

[0046] The process of the generation unit 257 will be described in more detail. First, the subscript i is randomly restored and extracted with the selection probability "w i " as the weight. Then, the pattern z = {z j}(j ∈ {1, ···, N}) to be obtained is determined by the following formula. Here, "U[0,1]" is a number generated by a uniform random number generator that outputs a value between 0 and 1. [Equation]

[0047] In addition, the generation unit 257 may generate a pattern by combining a plurality of solutions obtained by the calculation unit 255. The differences in the solutions represent the tendency included in the pattern, which cannot be expressed only by the first probability and the second probability. By generating patterns by combining different solutions, various patterns that cannot be obtained from the first probability, the second probability, or the input pattern can be generated. As a result, a more accurate layout design can be performed using the generated patterns.

[0048] The generation unit 257 outputs the generated pattern via the output unit 23.

[0049] [Processing Procedure of Data Generation Device] FIG. 2 is a flowchart showing the processing procedure of the data generation device according to the embodiment of the present disclosure.

[0050] In step S101, the first probability acquisition unit 251 acquires the first probability for each item. Further, the second probability acquisition unit 253 acquires the second probability for each combination of some or all of two different items.

[0051] In step S103, the calculation unit 255 calculates a selection probability and a conditional probability based on the first probability and the second probability.

[0052] In step S105, the generation unit 257 generates a pattern based on the selection probability and the conditional probability.

[0053] In step S107, the generation unit 257 determines whether a predetermined number of patterns have been generated. If the predetermined number of patterns have not been generated (NO in step S107), the process returns to step S105.

[0054] If the predetermined number of patterns have been generated (YES in step S107), in step S109, the generation unit 257 outputs the generated pattern via the output unit 23.

[0055] [Utilization examples of the generated patterns] By using a pattern for selecting one or more items from a plurality of items, it is possible to consider an appropriate layout design when arranging the items at a plurality of locations. For example, by arranging items that are likely to be selected simultaneously according to the pattern at the same location as much as possible, the movement cost between a plurality of locations during transportation can be reduced. In addition, by arranging items that are likely to be selected simultaneously according to the pattern within the same page such as an Internet site, the load associated with page transitions can be reduced.

[0056] However, generally, the pattern of selecting one or more items from a plurality of items can only be obtained by collecting actual operation data after actually arranging the items. When the combination of items is changed, or the location where the items are arranged is changed, etc., since there is no actual operation data, the pattern to be selected cannot be obtained, and an appropriate layout design cannot be considered. In order to perform a highly accurate layout design by simulation or the like, it is necessary to be able to generate an arbitrary number of patterns even in a situation where actual operation data cannot be collected.

[0057] An example of an item placement determination method will be described. For example, as "Process 1", when performing a highly accurate layout design by simulation or the like, the data generation device 20 described above generates the necessary number of patterns for selecting one or more items from a plurality of items.

[0058] As "Process 2", temporarily determine the placement of items at a plurality of locations to obtain a temporary placement. The placement of items may be determined randomly. Also, if there is a current placement of items, it may be used as the temporary placement.

[0059] As "Process 3", under the temporary placement, calculate an index for evaluating the efficiency of executing the delivery according to the plurality of patterns generated by the data generation device 20. As the index, the moving distance, the moving time, or the like may be adopted.

[0060] As "Process 4", create a plan for changing the placement of items in an appropriate manner. For example, a method such as swapping the placements of two randomly selected items can be considered. The updated placement of items based on the plan for changing the placement of items is defined as the updated placement.

[0061] As "Process 5", under the updated placement, calculate an index for evaluating the efficiency of executing the delivery according to the plurality of patterns generated by the data generation device 20.

[0062] As "Process 6", compare the indicators in the tentative layout with the indicators in the updated layout. If the updated layout is more efficient in executing the delivery, replace the tentative layout with the updated layout.

[0063] Then, repeat the above Processes 3 to 6 until sufficient efficiency regarding the delivery execution can be achieved. Thereby, a highly accurate layout design can be performed.

[0064] [Effects of the Embodiment] As described in detail above, the data generation device, data generation method, and data generation program according to the present disclosure relate to a controller connected to an input unit and an output unit. Regarding a pattern of selecting one or more items from a plurality of items, the first probability that an item of interest is included in the pattern is acquired for each item via the input unit. The second probability that a first item and a second item, which are two different items of interest, are both included in the pattern is acquired for each combination of some or all of the first item and the second item via the input unit. Then, a pattern is generated based on the first probability and the second probability, and the pattern is output via the output unit.

[0065] Thereby, it is possible to present a sufficient number of patterns regarding the selection of items, which are used when performing a highly accurate layout design by simulation or the like. In particular, even in a situation where actual operation data cannot be collected, any number of patterns can be generated. As a result, it is possible to provide the selection patterns necessary for considering the optimization of the item selection. And, by using a pattern of selecting one or more items from a plurality of items, an appropriate layout design such as when arranging items at a plurality of locations can be considered.

[0066] To confirm the effects of the embodiments of the present disclosure, 2,000 types of items were used, and 20,000 patterns related to item selection were generated. Then, the first probability and the second probability were estimated from the generated patterns and compared with the input first probability and second probability.

[0067] FIG. 3 is a diagram showing the correlation between the input first probability and the first probability estimated from the generated patterns. FIG. 4 is a diagram showing the correlation between the input second probability and the second probability estimated from the generated patterns. As shown in FIGS. 3 and 4, it can be seen that a good correlation is established between the input values and the estimated values. Therefore, it can be seen that the desired data is obtained according to the present disclosure.

[0068] The controller may calculate the selection probability that an item is selected by a pattern and the conditional probability that a second item is selected when a first item is selected by a pattern based on the first probability and the second probability. And it may generate a pattern based on the selection probability and the conditional probability. Thereby, parameters necessary for generating a pattern, namely, the selection probability and the conditional probability, can be obtained. As a result, a highly accurate selection pattern can be generated.

[0069] The controller may determine whether a first item related to the selection probability is included in the pattern based on the selection probability. When the first item is selected, it may determine whether to select a second item related to the conditional probability based on the conditional probability related to the selected first item. And it may generate a pattern based on the presence or absence of the selection of the first item and the presence or absence of the selection of the second item. Thereby, the relationship between the selection probability and the conditional probability and the generated pattern can be guaranteed. As a result, it can be guaranteed that the generated pattern reproduces the input first probability and second probability.

[0070] The controller may calculate the product of the selection probability that the first item is selected and the conditional probability that the second item is selected when the first item is selected. The controller may also change the first item over a plurality of items and sum the values of the products. The controller may change the selection probability and the conditional probability in a direction such that the total value obtained by summing the values of the products approaches the first probability that the second item is included in the pattern. Thereby, the relationship between the selection probability, the conditional probability, and the generated pattern can be guaranteed.

[0071] Alternatively, the controller may calculate the product (first product) of the selection probability that the first item is selected and the conditional probability that the second item is selected when the first item is selected. The controller may also calculate the product (second product) of the selection probability that the second item is selected and the conditional probability that the first item is selected when the second item is selected. Then, the controller may change the selection probability and the conditional probability in a direction such that the sum of the first product and the second product approaches the second probability that both the first item and the second item are included in the pattern. Thereby, the relationship between the selection probability, the conditional probability, and the generated pattern can be guaranteed. As a result, it can be guaranteed that the generated pattern reproduces the input first probability and second probability.

[0072] The items may be arranged at a plurality of different locations included in a warehouse or an area. And the pattern may describe the conveyance of items from a location or the conveyance of items to a location. Thereby, it is possible to consider an appropriate layout design when arranging items at a plurality of locations.

[0073] Based on the generated pattern, the controller may determine the arrangement of items at a location. Thereby, sufficient efficiency can be realized in the execution of delivery.

[0074] Each function shown in the above embodiments can be implemented by one or more processing circuits. The processing circuits include a programmed processor, an electric circuit, etc., and further include devices such as application-specific integrated circuits (ASICs), or circuit components arranged to execute the described functions.

[0075] According to the present disclosure, in warehouses, centers, etc. for storing and managing items, the efficiency of item shipping and storage is improved. As a result of the improved resource utilization efficiency, it can lead to infrastructure improvement and industrial improvement. Therefore, for example, it can contribute to Goal 9 of the Sustainable Development Goals (SDGs) led by the United Nations, "Build resilient infrastructure, promote inclusive and sustainable industrialization, and foster innovation."

[0076] Although several embodiments have been described, it is possible to modify or deform the embodiments based on the above disclosure. All the components of the above embodiments and all the features described in the claims may be individually extracted and combined as long as they do not conflict with each other.

Description of Reference Numerals

[0077] 20 Data generation device 21 Input unit 23 Output unit 25 Controller 27 Operation unit 29 Database 251 First probability acquisition unit 253 Second probability acquisition unit 255 Calculation unit 257 Generation unit

Claims

1. A data generation device comprising an input unit, an output unit, and a controller connected to the input unit and the output unit, wherein the controller regarding a pattern of selecting one or more of the items from a plurality of items, acquires, for each of the items, a first probability that the item of interest is included in the pattern via the input unit, acquires, for each combination of some or all of the first item and the second item, which are two different items of interest, a second probability that both the first item and the second item are included in the pattern via the input unit, generates the pattern based on the first probability and the second probability, and outputs the pattern via the output unit. A data generation device.

2. The controller calculates, based on the first probability and the second probability, a selection probability that the item is selected by the pattern and a conditional probability that the second item is selected when the first item is selected by the pattern, and generates the pattern based on the selection probability and the conditional probability. The data generation device according to claim 1.

3. The controller determines whether the first item related to the selection probability is included in the pattern based on the selection probability, and when the first item is selected, determines whether to select the second item related to the conditional probability based on the conditional probability related to the selected first item, and generates the pattern based on the presence or absence of selection of the first item and the presence or absence of selection of the second item. The data generation device according to claim 2.

4. The controller the value obtained by summing up the product of the selection probability that the first item is selected and the conditional probability that the second item is selected when the first item is selected, while varying the first item over a plurality of the items, is in a direction approaching the first probability that the second item is included in the pattern, and The sum of the product of the selection probability that the first item is selected and the conditional probability that the second item is selected when the first item is selected, and the product of the selection probability that the second item is selected and the conditional probability that the first item is selected when the second item is selected approaches the second probability that both the first item and the second item are included in the pattern, changing the selection probability and the conditional probability, The data generation device according to claim 2.

5. The items are arranged at a plurality of different locations included in a warehouse or an area, The pattern describes the conveyance of the items from the locations or the conveyance of the items to the locations, The data generation device according to any one of claims 1 to 4.

6. An item placement determination method for determining the placement of the items at the locations based on the pattern generated by the data generation device according to claim 5.

7. A data generation method for controlling a controller connected to an input unit and an output unit, The controller, Regarding a pattern of selecting one or more of the items from a plurality of items, acquiring, for each of the items, a first probability that the item of interest is included in the pattern via the input unit, acquiring, for each partial or total combination of the first item and the second item, which are two different items of interest, a second probability that both the first item and the second item are included in the pattern via the input unit, generating the pattern based on the first probability and the second probability, outputting the pattern via the output unit, A data generation method.

8. A data generation program to be executed by a controller connected to an input unit and an output unit, by the controller, Regarding a pattern of selecting one or more of the items from a plurality of items, acquiring, for each of the items, a first probability that the item of interest is included in the pattern via the input unit; acquiring, for each partial or total combination of the first item and the second item, which are two different items of interest, a second probability that both the first item and the second item are included in the pattern via the input unit; A step of generating the pattern based on the first probability and the second probability; A step of outputting the pattern via the output unit; A data generation program for causing the above to be executed.

Images