Information processing device, information processing method, and program

The information processing device and method address the challenge of applying data analysis predictions to optimization problems by using templates to generate customized optimization models, effectively solving complex optimization issues in real-world operations.

JP7593484B2Active Publication Date: 2024-12-03NEC CORP
View PDF 13 Cites 0 Cited by

Patent Information

Application Number
JP2023516288
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2021-04-19
Filing Date
2022-01-28
Publication Date
2024-12-03
Estimated Expiration
2042-01-28

AI Technical Summary

Technical Problem

Existing technologies face challenges in efficiently and effectively applying prediction results from data analysis to complex optimization problems in real-world operations, particularly in work sites, as they lack methods to facilitate accurate and efficient prediction and optimization.

Method used

An information processing device and method that generates an optimization model using templates defining objective functions and constraints, with item and algorithm definition information, allowing users to specify data items and algorithms for optimization problems, and includes features for customization and integration of prediction results.

Benefits of technology

Enables efficient and effective solving of complex optimization problems by generating tailored optimization models that align with business needs and user intentions, enhancing operational efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007593484000001
    Figure 0007593484000001
  • Figure 0007593484000002
    Figure 0007593484000002
  • Figure 0007593484000003
    Figure 0007593484000003
Patent Text Reader

Abstract

An information processing device (100) comprises a generation unit (102) that generates an optimization model (80) by using template information (10) that defines an objective function serving as an indicator of an optimization problem and a constraint condition therefor, the template information (10) including item definition information (12) that specifies a data item to be input to the constraint condition and the objective function, and algorithm definition information (14) that defines algorithms of the objective function and the constraint condition.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program, and more particularly to an information processing device, an information processing method, and a program for performing data analysis. [Background technology]

[0002] Technologies have been developed that predict future demand and abnormal occurrences by analyzing past data and generating a prediction model. For example, Patent Document 1 discloses a technology that predicts demand for an object that is the target of a business operator's business and corresponds to the attributes of the user who made a reservation for the area in which the business operator is located, based on the attributes of the user. Patent Document 8 also describes a device that creates appropriate work shifts.

[0003] It is not easy to apply predictions based on data analysis to the workplace. Various analysis systems have been proposed (Patent Documents 4 to 7, etc.), and there are, for example, various specific methods for generating a prediction model. Therefore, it is necessary to select an appropriate method from among these various methods.

[0004] In this regard, Patent Document 2 discloses a technology that uses templates to facilitate the analysis of business data. However, Patent Document 2 is intended to make it easier to understand past performance by statistically analyzing past data, and does not mention making predictions. Therefore, it does not disclose a technology that makes it easier to make predictions through data analysis. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2019-053737 A [Patent Document 2] JP 2000-285128 A [Patent Document 3] US Patent Application Publication No. 2014 / 0222741A1 [Patent Document 4] International Publication No. 2014 / 010071 [Patent Document 5] JP 2020-191114 A [Patent Document 6] JP 2019-160291 A [Patent Document 7] JP 2017-142800 A [Patent Document 8] JP 2012-181719 A [Non-patent literature]

[0006] [Non-Patent Document 1] Kenji Fukuda, "AI Transforms Financial Services," NEC Technical Report, vol. 69, No. 2, 2016, pp. 16-19 Summary of the Invention [Problem to be solved by the invention]

[0007] When applying the results of such data analysis to actual business operations, it is necessary to select the combination of variables (objective function) that will maximize a given indicator from among many options under various constraints in the business. Most It is desirable to efficiently and effectively solve so-called optimization problems, which seek a solution that maximizes or minimizes a certain parameter.

[0008] The present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide a technique for solving complex optimization problems efficiently and effectively. [Means for solving the problem]

[0009] In order to solve the above-mentioned problems, each aspect of the present invention employs the following configuration.

[0010] The first aspect relates to an information processing device. An information processing device according to a first aspect includes: A generating means for generating an optimization model by using a template that defines an objective function and a constraint condition that are indexes of an optimization problem, The template includes item definition information that specifies data items to be input into the objective function and the constraint conditions, and algorithm definition information that defines algorithms for the objective function and the constraint conditions.

[0011] A second aspect relates to an information processing method implemented by at least one computer. An information processing method according to a second aspect of the present invention comprises: An information processing device, generating an optimization model using a template that defines an objective function and constraint conditions that are indicators of the optimization problem; The template includes item definition information that specifies data items to be input into the objective function and the constraint conditions, and algorithm definition information that defines algorithms for the objective function and the constraint conditions.

[0012] Another aspect of the present invention may be a program for causing at least one computer to execute the method of the second aspect, or a computer-readable recording medium having such a program recorded thereon. This recording medium includes a non-transitory tangible medium. The computer program includes computer program code which, when executed by a computer, causes the computer to carry out the information processing method on an information processing device.

[0013] Any combination of the above components, and any transformation of the present invention into a method, device, system, recording medium, computer program, etc., are also effective as aspects of the present invention.

[0014] Furthermore, the various components of the present invention do not necessarily have to be independent entities, but may be formed as a single member by multiple components, one component may be formed from multiple components, one component may be a part of another component, or part of one component may overlap with part of another component, etc.

[0015] In addition, although the method and computer program of the present invention describe a number of steps in a sequential order, the order does not limit the order in which the steps are executed. Therefore, when implementing the method and computer program of the present invention, the order of the steps can be changed as long as it does not affect the content.

[0016] Furthermore, the steps of the method and computer program of the present invention are not limited to being executed at different timings, and therefore a step may occur while another step is being executed, or the execution timing of a step may overlap partially or entirely with the execution timing of another step, etc. Effect of the Invention

[0017] According to each of the above aspects, it is possible to provide a technique for efficiently and effectively solving a complex optimization problem. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 illustrates an example of an information processing system. [Diagram 2] 1 is a functional block diagram logically illustrating a configuration of an information processing apparatus according to an embodiment. [Diagram 3] 3 is a block diagram illustrating a hardware configuration of a computer that realizes the information processing device in FIG. 2. [Figure 4] 10 is a flowchart illustrating an example of an operation of the information processing apparatus according to the embodiment. [Diagram 5] FIG. 13 is a diagram illustrating an example of an index list screen of an optimization template. [Figure 6]FIG. 6 is a diagram showing an example of a data structure of template information of the template in FIG. 5. [Figure 7] 1 is a functional block diagram logically illustrating a configuration of an information processing apparatus according to an embodiment. [Figure 8] 10 is a flowchart illustrating an example of an operation of the information processing apparatus according to the embodiment. [Figure 9] FIG. 13 is a diagram showing an example of an optimization execution screen. [Figure 10] FIG. 13 is a diagram showing an example of a parameter setting screen. [Figure 11] FIG. 13 illustrates an example of an import screen. [Figure 12] FIG. 12 is a diagram showing an example of a data file imported in FIG. 11. [Figure 13] 10 is a flowchart illustrating an example of an operation of the information processing apparatus according to the embodiment. [Figure 14] 10 is a diagram showing another example of the optimization execution screen of FIG. 9. [Figure 15] FIG. 13 is a diagram illustrating an example of a data structure of template information of a template for executing intention learning. [Figure 16] 7 is a diagram showing another example of the index list screen of FIG. 5. [Figure 17] FIG. 13 is a diagram illustrating an example of an output of an analysis result. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] Hereinafter, the embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same components are given the same reference numerals, and the description will be omitted as appropriate. In addition, in each of the following drawings, the configuration of the parts that are not related to the essence of the present invention is omitted and is not shown.

[0020] In the embodiment, "acquisition" includes at least one of the following: a device goes to retrieve data or information stored in another device or storage medium (active acquisition), and inputs data or information output from another device to the device (passive acquisition). Examples of active acquisition include making a request or inquiry to another device and receiving a reply, and accessing and reading information from another device or storage medium. Examples of passive acquisition include receiving information that is distributed (or transmitted, push notification, etc.). Furthermore, "acquisition" may mean selecting and acquiring data or information from received data or information, or selecting and receiving distributed data or information.

[0021] (First embodiment) <System Overview> An information processing system 1 according to an embodiment of the present invention is a system for solving various optimization problems in business. The information processing system 1 is realized using one or more computers. FIG. 1 is a diagram illustrating an example of an implementation of the information processing system 1.

[0022] 1, the information processing system 1 includes an information processing device 100 (FIG. 2) that includes a front-end server 30 and a back-end server 40. In this manner, the information processing device 100 may be realized by at least two computers, or may be realized by one computer.

[0023] For example, the front-end server 30 provides the user U with a website for using the information processing system 1. The user U who wants to use the information processing system 1 first accesses the front-end server 30 using the user terminal 20. The front-end server 30 provides the user terminal 20 with template information 10 (FIG. 2) and a web page for specifying input data. The user U uses the provided web page on the user terminal 20 to perform operations such as selecting a template 5 and inputting data.

[0024] The user terminal 20 is realized by a computer such as a tablet terminal, and has a display device (not shown) such as a liquid crystal display, and an input device (not shown) such as a touch panel. The Web page provided to the user terminal 20 by the front-end server 30 is displayed on the display (not shown) of the user terminal 20, and the user U can input or select by operating a GUI (Graphical User Interface) displayed on the screen, for example, by operating the touch panel of the user terminal 20.

[0025] To use a Web page using the user terminal 20, for example, a user U who uses this system uses an application such as a browser to access a URL (Uniform Resource Locator) and is a user who has been registered in advance. a By logging in using the account, it becomes possible to use the Web page on the user terminal 20.

[0026] The front-end server 30 uses the template 5 and input data selected by the user U to cause the back-end server 40 to execute optimization. For example, the front-end server 30 causes the back-end server 40 to execute optimization by sending to the back-end server 40 a predetermined command including information received from the user terminal 20, such as identification information for the template 5. The back-end server 40 executes optimization in response to this instruction, thereby generating the optimized model 80.

[0027] Details of the processing indicated by the dashed arrows in Figure 1 will be described in the embodiment described later, but the front-end server 30 receives information representing the optimization results (information relating to the optimization model 80) from the back-end server 40, generates display information using view definition information corresponding to the template 5, and provides it to the user terminal 20 for display.

[0028] The realization form of the information processing system 1 is not limited to the above-mentioned example. For example, the front-end server 30 and the back-end server 40 may be realized by one computer. In addition, for example, the user terminal 20 may have the same functions as the front-end server 30. That is, the user terminal 20 may have a function of accepting the designation of the template information 10 and input data, a function of instructing the back-end server 40 to execute optimization, a function of receiving the optimization result from the back-end server 40, and a function of generating display information from the received optimization result (that is, an application that realizes the function of interacting with the back-end server 40 is installed in the user terminal 20). In addition, for example, both the functions of the front-end server 30 and the back-end server 40 may be provided in the user terminal 20. That is, the information processing system 1 may be realized by a computer operated by the user U (an application that realizes all the functions of the information processing system 1 is installed in the user terminal 20).

[0029] <Example of functional configuration> 2 is a functional block diagram logically showing the configuration of the information processing device 100 of this embodiment. The information processing device 100 includes a generating unit 102. The generating unit 102 generates an optimization model 80 by using template information 10 of a template 5 that defines an objective function and constraint conditions that are indices of an optimization problem.

[0030] The template information 10 includes item definition information 12 and algorithm definition information 14. The item definition information 12 and the algorithm definition information 14 include information defined by a user U. The item definition information 12 includes information that identifies data items to be input into an objective function and a constraint condition. The algorithm definition information 14 includes information that defines algorithms for the objective function and the constraint condition.

[0031] A plurality of templates 5, each of which defines an objective function and constraint conditions that are indices for an optimization problem, are prepared for each optimization problem. The templates 5 are defined by template information 10. The user U can select which template 5 to use depending on the business content, the purpose and content of the optimization to be solved, and the like. It is preferable that the templates 5 are prepared according to the purpose of optimization, the content of optimization, the target industry, the target business content, and the like. For example, when creating a work shift, the user U may select a template 5 based on the type of business to be optimized, such as whether to target a product factory or a retail store such as a convenience store, the number of shift allocations, the type of business content, and the like, as selection considerations. However, the template 5 may be prepared for each user U, specialized for the type of business or business content.

[0032] Furthermore, multiple objective functions and constraint conditions, which are indices of the optimization problem, are also prepared. The user U can specify which of the objective functions and constraint conditions to use to execute the optimization. Furthermore, each index of the optimization problem has a parameter section that the user U can change, allowing it to be specified to suit the actual work. In this way, business Template 5 can be customized according to the content, optimization purpose, content, and conditions that are emphasized.

[0033] The constraints of an optimization problem include essential conditions that must be satisfied (hereinafter also referred to as hard constraints) and optional conditions that should be satisfied if possible (hereinafter also referred to as soft constraints). In the template information 10, the soft constraints are incorporated as elements of the objective function, and the hard constraints are incorporated as constraints.

[0034] <Hardware configuration example> FIG. 3 is a block diagram illustrating a hardware configuration of a computer 1000 that realizes the above-described information processing device 100 (FIG. 2). In this embodiment, the information processing device 100 is front1 is realized by the end server 30 and the back-end server 40, each of which is realized by a computer 1000. The user terminal 20 in FIG.

[0035] Each functional component of the information processing device 100 may be realized by hardware that realizes each functional component (e.g., a hardwired electronic circuit, etc.), or may be realized by a combination of hardware and software (e.g., a combination of an electronic circuit and a program that controls it, etc.). Below, a further description will be given of the case where each functional component of the information processing device 100 is realized by a combination of hardware and software.

[0036] The information processing device 100 is realized by using one or more computers. Fig. 3 is a diagram illustrating a computer 1000 for realizing the information processing device 100. The computer 1000 is any computer. For example, the computer 1000 is a stationary computer such as a PC (Personal Computer) or a server machine. Alternatively, for example, the computer 1000 is a portable computer such as a smartphone, a tablet terminal, or a notebook PC.

[0037] The computer 1000 may be a dedicated computer designed to realize the information processing device 100, or may be a general-purpose computer. In the latter case, for example, a predetermined application is installed in the computer 1000, whereby at least a part of the functions of the information processing device 100 is realized by the computer 1000. The application is an application configured by a program for realizing any one or more of the functional components of the information processing device 100.

[0038] For example, as described later, the information processing device 100 may be configured by a back-end server 40 that generates an optimization model, and a front-end server 30 that functions as an interface between the user terminal 20 and the back-end server 40 (see FIG. 1). In this case, the front-end server 30 and the back-end server 40 are realized by different computers 1000. In this case, an application for realizing functions to be provided to the front-end server 30 among the functions of the information processing device 100 is installed in the computer 1000 that realizes the back-end server 40. On the other hand, an application for realizing functions to be provided to the back-end server 40 among the functions of the information processing device 100 is installed in the computer 1000 that realizes the back-end server 40.

[0039] The computer 1000 has a bus 1010, a processor 1020, a memory 1030, a storage device 1040, an input / output interface 1050, and a network interface 1060.

[0040] The bus 1010 is a data transmission path for transmitting and receiving data among the processor 1020, memory 1030, storage device 1040, input / output interface 1050, and network interface 1060. However, the method of connecting the processor 1020 and the like to each other is not limited to a bus connection.

[0041] The processor 1020 is implemented by a central processing unit (CPU) or a graphics processing unit (GPU).

[0042] The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.

[0043] The storage device 1040 is an auxiliary storage device realized by a hard disk drive (HDD), a solid state drive (SSD), a memory card, a read only memory (ROM), or the like. The storage device 1040 stores program modules that realize each function of the information processing device 100 (e.g., a generating unit 102 and a receiving unit 104 described below). The processor 1020 loads each of these program modules onto the memory 1030 and executes them to realize each function corresponding to the program module. The storage device 1040 may also store each data of the template information storage device 60 (e.g., template information 10), input data, the generated optimization model 80, and other data.

[0044] The program module may be recorded in a recording medium. The recording medium for recording the program module may include a non-transitory tangible medium usable by the computer 1000, and the program code readable by the computer 1000 (processor 1020) may be embedded in the medium.

[0045] The input / output interface 1050 is an interface for connecting the computer 1000 to various input / output devices. The input / output interface 1050 also functions as a communication interface for performing short-range wireless communication such as Bluetooth (registered trademark) and NFC (Near Field Communication).

[0046] The network interface 1060 is an interface for connecting the computer 1000 to a communication network. This communication network is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network). The method for connecting the network interface 1060 to the communication network may be a wireless connection or a wired connection.

[0047] The computer 1000 then connects to necessary equipment (e.g., the front-end server 30, the back-end server 40, and the user terminal 20's display, touch panel, keyboard, mouse, operation switches, speakers, microphones, cameras, printers (all not shown), etc.) via the input / output interface 1050 or the network interface 1060.

[0048] <Example of operation> 4 is a flowchart showing an example of the operation of the information processing device 100 of this embodiment. First, an optimization template 5 is prepared (step S101). Preparation of the template 5 includes the selection of indices to be used for optimization and the acceptance and storage of parameter settings (changes). The template 5 in which the indices are selected and parameters are set is stored in the template information storage device 60. This parameter setting process will be described in detail in the embodiment described later. In this embodiment, a configuration is adopted in which a template 5 in which parameters have already been set is used.

[0049] Then, the generating unit 102 generates the optimized model 80 using the template 5 (step S103).

[0050] In this way, the generation unit 102 optimizes the input data to generate the optimization model 80, and outputs information about the generated optimization model 80. Here, the method for generating the optimization model 80 and the mode in which information about the generated optimization model 80 is to be output are defined in advance as a template 5. Hereinafter, information representing this template 5 will be referred to as template information 10.

[0051] As described above, the template information 10 includes the item definition information 12 and the algorithm definition information 14. The item definition information 12 is information that specifies the items of each input data used to generate the optimization model 80. For example, assume that information on past shift performance and future shift preferences for each employee, and information on attributes such as the work that each employee can handle are used to generate the optimization model 80 that optimizes part-time shifts. In this case, in the template information 10 for generating the optimization model 80, the item definition information 12 includes items equivalent to "information on employee shift performance" (such as "work shift performance"), items equivalent to "information on employee vacation and work preferences" (such as "vacation and work preferences"), and items equivalent to "information on employee attributes" (such as "employee attributes").

[0052] The algorithm definition information 14 specifies an algorithm for generating the optimization model 80. For example, assume that multiple types of AI engines are prepared as program modules that embody the algorithm for generating the optimization model 80. In this case, the algorithm definition information 14 indicates information that identifies one of the multiple types of AI engines (AI engine identification information). However, the algorithm definition information 14 may include the AI ​​engine itself, rather than the AI ​​engine identification information. In addition, the embodiment of the algorithm used to generate the optimization model 80 is not limited to an AI engine.

[0053] FIG. 5 is a diagram showing an example of an index list screen 200 of a certain optimization template 5. The index list screen 200 includes an objective function list 210, a constraint condition list 220, an execute button 240, and an interrupt button 242. The objective function list 210 includes a plurality of objective functions, which are provided for each item to be optimized, that is, for each index of the optimization problem to be solved. Furthermore, the constraint condition list 220 also includes a plurality of constraint conditions, which are provided for each item to be optimized, that is, for each index of the optimization problem to be solved. For example, in the case of a retail store, the items to be optimized include operating costs such as employee allowances, the total working hours of all employees, the employee attendance request achievement rate, the customer satisfaction expectation value, the number of expected visitors, and the like.

[0054] In Template 5, a set of multiple indices is provided for each optimization objective. Examples of optimization objectives (optimization problems) include optimizing work shifts in a manufacturing factory or retail store, and optimizing the matching of the attributes of personnel required for a process on the schedule with the required human resources.

[0055] The objective function list 210 includes an objective function name column, an index description column, a use column indicating whether the objective function is used or not, an effect column, a weight column, and a parameter column. The constraint condition list 220 includes a constraint condition name column, a condition description column, a use column, and a parameter column. Details of each column will be described in the embodiments described later.

[0056] Furthermore, in the objective function list 210 and the constraint condition list 220, a setting button 230 is provided for each row, and by accepting the pressing of the setting button 230, the parameters, weights, etc. of each index can be set. The details of the setting of the parameters and weights will also be described in the embodiment described later. In the objective function list 210 and the constraint condition list 220, a "●" is displayed in the use column for an index selected for use by the user U. For example, by selecting the use column of the row of each index in the objective function list 210 and the constraint condition list 220, it may be possible to specify whether or not to use the index.

[0057] FIG. 5 shows an example of an index list screen 200 of an optimization template 5 for part-time worker work shifts in a certain store. In addition, consecutive shift constraints and consecutive shift combinations are set as objective functions. constraints , the minimum number of staff by day of the week and by shift, and employee cost are selected, and consecutive shift constraints, weekly shift constraints, consecutive shift combination constraints, and the minimum number of staff by day of the week and by shift are selected as constraint conditions. In the case of a product planogram optimization problem, other objective functions include the minimum number of items, the minimum number of products, the maximum product cost, and product item combination constraints. In this embodiment, it is assumed that the objective functions and constraint conditions to be used are preselected, and that the parameters of each index are also preconfigured.

[0058] FIG. 6 is a diagram showing an example of the data structure of template information 10 of template 5 in FIG. 5. An optimization AI engine is specified as the engine type. The optimization index includes the objective function and constraint conditions that can be specified in FIG. 5, and corresponds to algorithm definition information 14. The output value is information about the output value that is output as the optimization result. Here, a work schedule is output, and may also include the specification of the schedule start date and schedule period, and layout information of the output data file. The item definition indicates the data items to be input, and corresponds to item definition information 12.

[0059] In this way, the template information 10 includes view definition information 16 that specifies the display mode of information related to the analysis result by the optimization model 80. (not shown) For example, the view definition information 16 includes the type and structure of a chart to be used to represent information about the optimization model 80, or the arrangement of multiple charts.

[0060] The display information for the optimization model 80 generated by the generating unit 102 is generated in a display format defined by the view definition information 16 of the template information 10. For example, information on the optimization model 80 is displayed using a diagram that is visually easy to understand. Therefore, the view definition information 16 includes definitions of the type, structure, and the like for one or more diagrams included in the display information. The type of diagram may be any type, such as a table, a scatter plot, a line graph, or a bar graph. The structure of the table includes, for example, a definition of each column. The structure of the graph includes, for example, a definition of each axis. The view definition information 16 further includes information that determines the overall arrangement of multiple diagrams and other information.

[0061] The above-mentioned front-end server 30 can generate display information by processing the optimization model 80 using the view definition information 16. The front-end server 30 outputs the display information to the user terminal 20. For example, this display information is a Web page on which information on the optimization model 80 can be viewed in a display format defined in the view definition information 16 of the template information 10. Alternatively, for example, the display information may be provided as a file such as a PDF file.

[0062] Returning to FIG. 5, the index list screen 200 further includes an optimized interval week number designation field 202 and an optimized interval start date designation field 204. The optimized interval week number designation field 202 is a text box for accepting designation of the number of weeks of the period to be optimized, and the optimized interval start date designation field 204 is a text box for accepting designation of the start date of the period to be optimized. However, instead of designating the number of weeks, Optimal The end date of the conversion execution period may be specified.

[0063] When the execute button 240 is pressed by the user U, the generation unit 102 generates an optimized model 80 for the period specified in the optimization section, using the indices selected for use in advance in the template information 10 (FIG. 6) and the parameters preset for those indices. That is, in this example, the work shift schedules of part-time workers in the specified optimization section are optimized and generated. On the other hand, when the interrupt button 242 is pressed by the user U, the optimization is stopped and the index list screen 200 is closed in response to the press.

[0064] As described above, according to this embodiment, the generation unit 102 generates the optimization model 80 using the template 5 that defines the objective function and constraint conditions that serve as indices for the optimization problem, so that it becomes possible to generate the optimization model 80 using the template 5 specialized for a task by having the user U define the indices. This makes it possible to generate the optimization model 80 that efficiently and effectively solves a complex optimization problem.

[0065] Second embodiment <Example of functional configuration> Fig. 7 is a functional block diagram logically showing the configuration of an information processing device 100 of this embodiment. The information processing device 100 of Fig. 7 includes the same generating unit 102 as the information processing device 100 of Fig. 2, and further includes a receiving unit 104. The information processing device 100 of this embodiment is the same as the information processing device 100 of Fig. 2 except that it further includes a configuration for receiving designation of parameters for changing at least a part of the indexes of the optimization problem and generating an optimization model 80 using the parameters. In addition, the information processing device 100 of this embodiment may be combined with at least one of the configurations of the other embodiments to the extent that no contradiction occurs.

[0066] The receiving unit 104 receives designation of parameters for changing at least a part of the indexes of the optimization problem. The generating unit 102 generates the optimization model 80 using the template 5 that has been modified using the received parameters.

[0067] <Example: Parameter setting> Fig. 8 is a flowchart showing an example of the operation of the information processing device 100 of this embodiment. The flowchart in Fig. 8 includes steps S111 and S113 instead of step S101 in the flowchart in Fig. 4. Fig. 9 is a diagram showing an example of an optimization execution screen 400. First, a user U Most When performing optimization, the optimization execution screen 400 is displayed on the user terminal 20, and the optimization template 5 is selected (step S111 in FIG. 8).

[0068] 9 includes a template list 402 and an optimization execution button 406. The template list 402 includes, for each template 5, identification information (template ID) of the template 5, the name of the template 5 (template name), and the classification of the template 5. When a template 5 is selected by the user U (for example, by clicking with the mouse over the relevant line), the relevant line 404 is displayed, for example, inverted in response to the selection.

[0069] When the optimization execution button 406 is pressed, the generation unit 102 displays the index list screen 200 (FIG. 5) of the selected template 5. The reception unit 104 receives the selection of the index to be used designated by the user U from among the multiple indexes (objective functions and constraint conditions) displayed on the index list screen 200 and the parameters set by the user U (step S113).

[0070] Specifically, when the receiving unit 104 receives a press of the setting button 230 of each row of the objective function list 210 and the constraint condition list 220 on the indicator list screen 200 of Fig. 5, the receiving unit 104 displays the indicator parameter setting screen 300 of Fig. 10. In this example, the parameter setting screen 300 is displayed when the setting button 230 of the constraint condition of the minimum number of staff by day of the week and by shift in the constraint condition list 220 is pressed.

[0071] In the following description, various screens are displayed on the display of the user terminal 20 by the front-end server 30. The front-end server 30 (acceptor 104) accepts operations by the user U of the GUI displayed on the screen, and transfers the accepted inputs such as data and instructions to the back-end server 40.

[0072] The parameter setting screen 300 includes a specification method explanation field 310, a parameter specification field 320, and a weight specification field 330. Furthermore, the parameter setting screen 300 includes a registration button 340 and a back button 342. The specification method explanation field 310 describes a parameter specification method for changing at least a part of the index. The parameter specification field 320 allows input in a text format in multiple lines. Parameters can be specified according to the specification method in the specification method explanation field 310. Since this index is a constraint condition of the minimum number of personnel by day of the week and by shift, the minimum number of personnel required by day of the week and shift unit such as morning, afternoon, and night shift can be specified as parameters for each line in the parameter specification field 320. The code indicating the shift is 1 for morning shift, 2 for afternoon shift, and 3 for night shift.

[0073] In the example shown in this figure, lines 1 to 4 specify that the minimum number of staff for the Monday morning shift is 2, the minimum number of staff for the Monday afternoon shift is 3, the minimum number of staff for the Monday night shift is 1, and the minimum number of staff for the Tuesday morning shift is 2.

[0074] Furthermore, in the example of this figure, a "weight" is specified as one of the parameters for changing the index, but the setting of the "weight" will be explained in the embodiment described later.

[0075] After the input of parameters in the parameter specification field 320 and the weight specification field 330 is completed, when the reception unit 104 receives a press of the registration button 340 by the user U, eachThe parameters are set based on the contents entered in the fields, and the template information 10 of the template 5 is updated. When the receiving unit 104 receives a press of the back button 342 by the user U, the parameter setting screen 300 is closed and the screen returns to the index list screen 200. When a parameter is changed and the register button 340 is pressed on the parameter setting screen 300, the changed parameters are displayed in the objective function list 210 and the constraint condition list 220 when the screen returns to the index list screen 200.

[0076] Then, the generating unit 102 generates the optimized model 80 using the changed template 5 (step S103 in FIG. 8).

[0077] In this manner, in this embodiment, the receiving unit 104 receives the specification of parameters for changing at least a part of the index of the optimization problem, and the generating unit 102 generates the optimization model 80 using the template 5 that has been changed using the received parameters. of By specifying the parameters, the parameters can be changed according to the business content, and the optimization model 80 can be generated using the changed parameters. Therefore, complex optimization problems can be solved efficiently and effectively according to the business content.

[0078] Third embodiment <Example of functional configuration> The information processing device 100 of this embodiment includes a generating unit 102 and a receiving unit 104 similar to those of the information processing device 100 of FIG. 7. The information processing device 100 of this embodiment differs from the information processing device 100 of FIG. 7 in that the generating unit 102 receives input of data defined in the item definition information 12, performs optimization using the specified index, and generates an optimization model 80. The information processing device 100 of this embodiment may be combined with at least one of the configurations of the other embodiments to the extent that no contradiction occurs. For example, it may be combined with the information processing device 100 of FIG. 2. In that case, since the information processing device 100 does not include the receiving unit 104, the generating unit 102 performs optimization using a template in which an index for an optimization problem is specified in advance.

[0079] The generation unit 102 inputs data of the data items defined in the item definition information 12 , and optimizes the input data using the template 5 to generate an optimized model 80 .

[0080] <Example: Data import> First, when executing optimization, the user U causes the user terminal 20 to display the optimization execution screen 400 of Fig. 9. Then, the generation unit 102 receives the template ID of the template 5 selected in the template list 402. Then, in response to pressing of an optimization execution button 406, the generation unit 102 imports data necessary for optimization using the selected template 5.

[0081] The generation unit 102 accepts the specification of a data file including data of the data items defined in the item definition information 12 of the selected template 5, and inputs the data from the data file. Specifically, the data file to be imported is specified on the import screen 500 as follows.

[0082] Fig. 11 is a diagram showing an example of the import screen 500. For example, when the optimization execution button 406 on the optimization execution screen 400 in Fig. 9 is pressed, the import screen 500 is displayed. However, the data file in which the data items required for optimization are stored may be specified in advance, in which case the import screen 500 does not need to be displayed. For example, the data file in which the data specified in the data items of the item definition information 12 are stored may be linked in advance.

[0083] The import screen 500 includes a file specification list 502 and an import execution button 506. The file specification list 502 includes information on data items required for optimization (e.g., data names) and a file storage location for the data that the user U selects. reference The file selection button 504 may also include an operation button (not shown) that allows reference to the layout of required data items. Since the layout of required data items can be referenced, a data file including required data items can be created and specified.

[0084] The data file is, for example, a Comma Separated Values ​​(CSV) file, but the data file format may be other formats, such as a text file written in Extensible Markup Language (XML), a Tab-Separated Values ​​(TSV) file, an Excel (registered trademark) file, or a SYMbolic LinK (SYLK) file, but is not limited to these.

[0085] Also, a check box for specifying that a data item is specified in the first line of the data file may be further included for each data item (each line) in the file specification list 502. Multiple data files may be specified, and a file add button (not shown) may be further included for each data item (each line).

[0086] Fig. 12 is a diagram showing examples of data files imported in Fig. 11. When the optimization template 5 for work shift optimization (part-time worker management) is selected, the data files imported are, for example, actual shift data 510 (Fig. 12(a)), desired vacation and work data 512 (Fig. 12(b)), and employee master 514 (Fig. 12(c)).

[0087] When the import of the data file is completed on the import screen 500, the index list screen 200 in FIG. 5 is displayed.

[0088] When the execute button 240 on the index list screen 200 is pressed, the generation unit 102 optimizes the imported data using the indexes (objective function and constraint conditions) of the optimization problem specified in the template 5, and generates an optimization model 80.

[0089] In this embodiment, the generation unit 102 inputs data of data items defined in the item definition information 12, and optimizes the input data using the template 5 to generate an optimization model 80. With this configuration, the optimization model 80 can be generated by prompting the user U to specify and import data required for optimization according to the item definition information 12 and algorithm definition information 14 of the template 5. Therefore, complex optimization problems can be solved efficiently and effectively in accordance with the business contents.

[0090] (Fourth embodiment) The information processing device 100 of this embodiment includes the same generating unit 102 and receiving unit 104 as the information processing device 100 of Fig. 7, and will be described with reference to Fig. 7. The information processing device 100 of this embodiment differs from any of the above-described information processing devices 100 in that the generating unit 102 derives a "weight" indicating a user's intention by intention learning using past performance, and changes an index of an optimization problem using the "weight". In addition, the information processing device 100 of this embodiment may be combined with at least one of the configurations of the other embodiments to the extent that no contradiction occurs.

[0091] The template information 10 includes a weight indicating the intention of the user U for each index of the optimization problem. Here, the user U is a store manager, a workplace manager, an expert, etc. The "weight" is derived by intention learning that learns the intention of the user U using past performance. The algorithm definition information 14 includes an intention learning algorithm for estimating this weight. The generation unit 102 calculates the weight using this intention learning algorithm. The generation unit 102 reflects the calculated weight in the template information 10 to generate the optimization model 80.

[0092] As described above, the weights indicate the intentions of the store manager or the like, so it is possible to optimize the system by reproducing, for example, what the store manager considers important.

[0093] In the example of the parameter setting screen 300 in FIG. 10, a "weight" indicating a penalty for exceeding the minimum number of staff for each shift can be specified. In this example, the "weight" is a penalty, but is not limited to this. In other examples, the "weight" may not be a penalty, but may be an evaluation, a level, or points. The "weight" is an index indicating the importance of an indicator, and in the case of a store, it may indicate the degree to which the store manager or the like considers it important, or it may indicate the degree of importance of an indicator that is important during a specific period such as a campaign.

[0094] In this example, a weight is specified with a value between 1 and 5, with the higher the value, the heavier the penalty. The method for specifying the weight is described in the specification method explanation field 310. Weights can also be entered in text format using multiple lines. A weight is specified for each shift, with one line for each shift. In the example shown in this figure, the weight for the morning shift (shift code 1) is specified as 2, the weight for the afternoon shift (shift code 2) as 2, and the weight for the night shift (shift code 3) as 5.

[0095] In this embodiment, the generation unit 102 can further change the designated weights through intention learning.

[0096] <Example of operation: Intention learning> Fig. 13 is a flowchart showing an example of the operation of the information processing device 100 of this embodiment. The flowchart in Fig. 13 includes the same steps S101 and S103 as Fig. 4, and also includes step S121 between steps S101 and S103. Note that step S101 in the flowchart in Fig. 13 may be replaced with steps S111 and S113 in Fig. 8.

[0097] FIG. 14 is a diagram showing another example of the optimization execution screen 400 of FIG. 9. The optimization execution screen 400 further includes an intention learning execution button 408. When template 5 is selected on the optimization execution screen 400 and the intention learning execution button 408 is pressed, the generation unit 102 calculates a weight by intention learning and reflects it in the weight of template 5 (step S121 in FIG. 13). FIG. 15 is a diagram showing an example of the data structure of template information 10 of template 5 for executing intention learning. In addition to the data structure of template information 10 in FIG. 6, an intention learning AI engine is added to the learning engine. Furthermore, Intention learning results As output value, Closed Free time and work request completion status has been added.

[0098] Fig. 16 is a diagram showing another example of the index list screen 200. Fig. 16 further includes a learning interval start date designation field 206, a learning interval end date designation field 207, a verification interval start date designation field 208, and a verification interval end date designation field 209 in addition to the index list screen 200 of Fig. 5.

[0099] Intention learning is performed in the learning interval specified in these fields. The validation interval is a period in which optimization is performed after the user U manually adjusts the intention-learned weights. Then, the generation unit 102 generates the optimized model 80 using the template 5 in which the intention of the user U is reflected by the intention learning (step S103 in FIG. 13).

[0100] The optimization model 80 includes, as a result of the intention learning, weights calculated by the intention learning, and their average value and standard deviation value for each index of the optimization problem. Furthermore, the optimization model 80 includes, as a result of solving the optimization problem using the weights calculated by the intention learning, average value and standard deviation value for each index of the optimization problem. Furthermore, the optimization model 80 may include, for each index, an evaluation result (e.g., OK or NG) against a pre-specified evaluation criterion for the result of the optimization problem.

[0101] In addition, the optimization model 80 may further include, for each index, a simulation result in which the weights calculated by the intention learning are further adjusted manually by the user U and the optimization problem is solved using the adjusted weights.

[0102] FIG. 17 is a diagram showing an example of the output of the analysis results. FIG. 17(a) shows a shift schedule 600, and FIG. 17(b) shows the vacation / work request achievement status 610. The shift schedule 600 includes index values ​​for each day of the shift (here, the number of staff during the busy lunch hour and the number of people available to work the cash register) and an evaluation result (e.g., OK or NG) of whether the index values ​​satisfy the evaluation criteria. The evaluation criteria are, for example, mandatory constraints.

[0103] It may be possible to drill down on the shift schedule 600 in Fig. 17(a) by person or by day. For example, it may be possible to refer to the attributes of employees working in shifts on a daily basis.

[0104] Vacation / work request achievement status 610 in FIG. 17(b) displays the achievement status (here, approved or not approved) of vacation / work requests on a per-person basis.

[0105] Thus, in this embodiment, the generation unit 102 generates the optimization model 80 using the weights calculated by the intention learning. With this configuration, the optimization model 80 can be generated with an index for the optimization problem that is in line with the intention of the user U, such as a store manager, that is, an index that the user U considers important. Therefore, a complex optimization problem can be effectively solved efficiently in line with the business content and the intention of the user U.

[0106] Fifth embodiment The information processing device 100 of this embodiment includes the same generating unit 102 and receiving unit 104 as the information processing device 100 of Fig. 7, and will be described with reference to Fig. 7. The information processing device 100 of this embodiment differs from any of the above-described information processing devices 100 in that the generating unit 102 uses a prediction result by a prediction model as one of the input data items of the optimization model 80. In addition, the information processing device 100 of this embodiment may be combined with at least one of the configurations of the other embodiments to the extent that no contradiction occurs.

[0107] The generation unit 102 uses the prediction result by the prediction model as one of the input data items of the optimization model 80.

[0108] A predictive model is generated by performing various predictive analyses on business operations by processing input data based on various algorithms. As an algorithm for generating a predictive model, various machine learning algorithms such as heterogeneous mixture learning (Patent Document 3), RAPID time series analysis (Non-Patent Document 1), neural network, or SVM (Support Vector Machine) can be used.

[0109] For example, for each of various machine learning algorithms, an AI engine, which is a program module that realizes the algorithm, is provided. The predictive model may also be generated using a template 5 similar to the template 5 used to generate the optimization model 80 of this embodiment. In this case, for example, the algorithm definition information 14 includes identification information that identifies one of the multiple AI engines. The predictive model is generated using the AI ​​engine identified by the identification information included in the algorithm definition information 14.

[0110] Here, the same AI engine may be used for multiple analysis types (regression, discrimination, etc.). In this case, the algorithm definition information 14 further includes information indicating the type of analysis to be performed (type of predictive model to be generated). For example, template information 10 for generating a predictive model that predicts product sales by heterogeneous mixture learning has "AI engine: heterogeneous mixture learning, analysis type: regression". On the other hand, template information 10 for generating a predictive model that predicts whether or not a device will break down in the future by heterogeneous mixture learning has "AI engine: heterogeneous mixture learning, analysis type: discrimination".

[0111] The algorithm definition information 14 also includes information representing the correspondence between the objective variables and explanatory variables of the AI ​​engine and the input data. For example, which subitem among the subitems specified in the item definition information 12 is to be used as the objective variable and which subitem is to be used as the explanatory variable is determined in the algorithm definition information 14. However, the objective variables and explanatory variables only need to have some relationship with one or more subitems specified in the item definition information 12, and do not need to completely match the subitems. For example, in predicting the sales volume of a product, the sales volume of the product can be included in the sales performance data (the subitem corresponding to the major item, the sales performance data, can include "sales volume"), and the objective variable can be set to "the difference from the moving average of the sales volume."

[0112] Hyperparameters to be set in the AI ​​engine may also be defined in the algorithm definition information 14. Examples of hyperparameters include the depth of a tree in heterogeneous mixture learning and the depth of a layer in a neural network.

[0113] Furthermore, information specifying pre-processing to be performed on the input data before inputting it into the AI ​​engine may be defined. When generating a predictive model in the AI ​​engine, the learning accuracy can be improved by performing scale conversion or the like instead of using the input data as is. Therefore, such pre-processing to be applied to the input data is defined in the item definition information 12. In addition, for example, a process of extracting only a part of the input data as a processing target is also defined as pre-processing. In addition, for example, a process of converting the format of the input data into a predetermined format (a format that can be interpreted by the AI ​​engine) defined for each AI engine is also defined as pre-processing.

[0114] The algorithm definition information 14 may include the program module itself that realizes the preprocessing, or may include identification information (such as a function name) and setting information (such as arguments) for calling a program that realizes the preprocessing. In the latter case, various preprocessing processes are provided in advance in the generation unit 102. Then, by defining the identification information of the preprocessing to be used and the setting information of the preprocessing in the algorithm definition information 14, the desired preprocessing can be executed by the generation unit 102.

[0115] Here, the prediction model (prediction target) generated by the analysis using one piece of template information 10 is not limited to one. For example, assume that template information 10 is prepared for predicting sales figures for each store and each product. In this case, the prediction target is the sales figures for each combination of "store, product." Therefore, when this template information 10 is used, a prediction model is generated for each store and each product.

[0116] As a simple example, assume that there are three types of products, G1 to G3, and two stores, S1 to S2. In this case, there are six prediction targets, so six prediction models are generated. G 3 A predictive model for each sales volume is generated.

[0117] What kind of prediction model will be created by analysis using one piece of template information 10 is defined in advance by algorithm definition information 14. That is, in the template information 10 for generating a prediction model that predicts sales figures for each store and each product, a target variable is defined in advance in the form of "sales for each store and each product." Therefore, the generation unit 102 generates a prediction model for each store and each product.

[0118] Here, it is preferable that the generation unit 102 not only generates a prediction model but also evaluates (verifies) its accuracy. In this case, for example, the generation unit 102 divides the input data into learning data and validation data. Then, the generation unit 102 generates a prediction model (learns a model) using the learning data, and verifies the prediction model using the validation data. Alternatively, for example, the generation unit 102 may perform so-called cross-validation. A specific method for dividing the input data in this way and generating and evaluating a model can utilize existing technology.

[0119] Furthermore, the generating unit 102 may perform prediction using the predictive model in addition to generating and verifying the predictive model. In this case, for example, the generating unit 102 divides the input data into test data used for prediction and other data (data used for learning and verification). Then, the generating unit 102 performs generation and verification of the predictive model with the latter, and then performs prediction using the test data. Note that existing technologies can be used as a specific method for dividing the input data in this way, generating and verifying the predictive model, and performing prediction.

[0120] However, the information processing device 100 does not necessarily need to generate a prediction model and immediately execute a prediction. For example, a user may first use the information processing device 100 to generate and verify a prediction model. The generated prediction model is stored in a storage device accessible from the information processing device 100. Thereafter, when a need arises for a prediction, the user uses the prediction model generated in advance to execute the prediction.

[0121] Here, the method of dividing the input data may be fixed and determined independently of the template information 10, may be determined by the template information 10, or may be specified by the user. For example, when dividing the input data by period, the user specifies the period of the input data to be used for each of the learning data, the verification data, and the test data.

[0122] Note that prediction using the prediction model does not necessarily have to be executed by the information processing device 100. For example, when the information processing device 100 is configured with the front-end server 30 and the back-end server 40 as described above, prediction using the prediction model may be executed by the user terminal 20. In this case, the prediction model generated by the information processing device 100 is stored in a storage device accessible from the user terminal 20.

[0123] In this embodiment, the generation unit 102 can use the prediction result of the prediction model as one of the items of input data for the optimization model. This allows the optimization model 80 to be generated based on various analysis contents, making it possible to generate a more advanced optimization model 80.

[0124] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various configurations other than those described above can also be adopted. For example, the template 5 itself used in the above embodiment may be configured to be customized. For example, the layout of the data file to be imported may be changeable. Furthermore, the objective function and constraint conditions included in the template 5 may be added and changed to create a different template 5. Furthermore, the selection of an AI engine for performing optimization or intent learning, and the method of specifying the interval for performing optimization or intent learning may be changeable.

[0125] Although the present invention has been described above with reference to the embodiments and examples, the present invention is not limited to the above-mentioned embodiments and examples. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. In the present invention, when information about a user is acquired and used, this shall be done lawfully.

[0126] A part or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes. 1. A generating means for generating an optimization model using a template that defines an objective function and constraint conditions that are indexes of an optimization problem; The template includes item definition information that identifies data items to be input into the objective function and the constraint conditions, and algorithm definition information that defines algorithms for the objective function and the constraint conditions. 2. In the information processing device according to 1., a receiving unit configured to receive a parameter specification for changing at least a part of the index of the optimization problem; The generating means generates the optimization model by using the template that has been modified using the received parameters. 3. In the information processing device according to 1. or 2., The generation means is an information processing device that inputs data of the data items defined by the item definition information, and optimizes the input data using the template to generate the optimized model. 4. In the information processing device according to any one of 1. to 3., The template includes view definition information that specifies a display mode of information regarding an analysis result based on the optimization model. 5. In the information processing device according to any one of 1. to 4., The information processing device, wherein the templates are provided for each item to be optimized by the optimization model. 6. In the information processing device according to any one of 1. to 5., The information processing device, wherein in the template, the sets of multiple indicators are provided for different optimization purposes. 7. In the information processing device according to any one of 1. to 6., The information processing device, wherein the generation means uses a prediction result by a prediction model as one of the data items of the input data of the optimization model. 8. In the information processing device according to any one of 1. to 7., The information processing device, wherein the template includes a weight indicating a user's intention for each index of the optimization problem. 9. In the information processing device according to 8., The algorithm definition information includes an intent learning algorithm for estimating the weights, The generation means calculates the weights using the intention learning algorithm and reflects the weights in the template to generate the optimized model.

[0127] 10. An information processing device, An optimization model is generated using a template that defines an objective function and constraint conditions that are indices for the optimization problem; An information processing method, wherein the template includes item definition information that identifies data items to be input into the objective function and the constraint conditions, and algorithm definition information that defines algorithms for the objective function and the constraint conditions. 11. In the information processing method according to 10., The information processing device further comprises: accepting designation of parameters for changing at least a portion of the index of the optimization problem; generating the optimization model using the template modified using the received parameters. 12. In the information processing method according to 10. or 11., The information processing device, An information processing method comprising: inputting data of the data items defined by the item definition information; and optimizing the input data using the template to generate the optimized model. 13. In the information processing method according to any one of 10. to 12., An information processing method, wherein the template includes view definition information that specifies a display mode of information relating to an analysis result using the optimization model. 14. In the information processing method according to any one of 10. to 13., An information processing method, wherein the template is provided for each item to be optimized by the optimization model. 15. In the information processing method according to any one of 10. to 14., An information processing method, wherein in the template, the sets of multiple indicators are provided for different optimization purposes. 16. In the information processing method according to any one of 10. to 15., The information processing device, An information processing method, comprising: using a prediction result by a prediction model as one of the data items of the input data of the optimization model. 17. In the information processing method according to any one of 10. to 16., The information processing method, wherein the template includes a weight indicating a user's intention for each index of the optimization problem. 18. In the information processing method according to 17., The algorithm definition information includes an intent learning algorithm for estimating the weights, The information processing device, The information processing method includes calculating the weights using the intention learning algorithm and reflecting the weights in the template to generate the optimized model.

[0128] 19. To the computer: A procedure for generating an optimization model using a template that defines an objective function and constraint conditions that are indicators of the optimization problem is executed; The template is a program including: item definition information that identifies data items to be input into the objective function and the constraint conditions; and algorithm definition information that defines algorithms for the objective function and the constraint conditions. 20. In the program according to paragraph 19, receiving a designation of parameters for changing at least a part of the indexes of the optimization problem; generating the optimization model using the template that has been modified using the received parameters. 21. In the program according to 19. or 20., A program for causing a computer to execute a procedure of inputting data of the data items defined by the item definition information, and optimizing the input data using the template to generate the optimized model. 22. In the program according to any one of 19. to 21., The template is a program including view definition information that specifies a display mode of information related to an analysis result obtained by the optimization model. 23. In the program according to any one of 19. to 22., The template is provided for each item to be optimized by the optimization model. 24. In the program according to any one of 19. to 23., A program, wherein in the template, the sets of multiple indicators are provided for different optimization purposes. 25. In the program according to any one of 19. to 24., A program for causing a computer to execute a procedure for using a prediction result by a prediction model as one of the data items of the input data of the optimization model. 26. In the program according to any one of 19. to 25., The template includes a weight indicating a user's intention for each index of the optimization problem. 27. In the program according to 26., The algorithm definition information includes an intent learning algorithm for estimating the weights, A program for causing a computer to execute a procedure of calculating the weights using the intention learning algorithm and reflecting the weights in the template to generate the optimization model.

[0129] This application claims priority based on Japanese Patent Application No. 2021-070593, filed on April 19, 2021, the disclosure of which is incorporated herein in its entirety. [Explanation of symbols]

[0130] 1. Information Processing Systems 5. Templates 10 Template Information 12 Item definition information 14 Algorithm definition information 16 View definition information 20 User terminal 30 Front-end Server 30 End Server 40 Backend Server 60 Template information storage device 80 Optimization Models 100 Information processing device 102 Generation part 104 Reception 200 Index list screen 202 Optimization interval week number specification field 204 Optimization section start date specification field 206 Study section start date specification field 207 Study section end date specification field 208 Verification section start date specification field 209 Verification section end date specification field 210 Objective Function List 220 List of constraints 230 Settings button 240 Run button 242 Pause button 300 Parameter setting screen 310 Specification method explanation column 320 Parameter specification field 330 Weight specification field 340 Registration button 342 Button 400 Optimization execution screen 402 Template List 404 lines 406 Optimization execution button 408 Intention learning execution button 500 Import screen 502 File Specification List 504 File selection button 506 Import execution button 510 Shift performance data 512 Vacation and work preference data 514 Employee Master 600 Shift Schedule 610 Vacation / Work Request Completion Status 1000 Computers 1010 Bus 1020 Processor 1030 Memory 1040 Storage Device 1050 Input / Output Interface 1060 Network Interface

Claims

1. A generating means for generating an optimization model by using a template that defines an objective function and a constraint condition that are indexes of an optimization problem, the template includes item definition information that identifies data items to be input into the objective function and the constraint condition, and algorithm definition information that defines algorithms for the objective function and the constraint condition; the template includes a weight indicating a user's intention for each index of the optimization problem; The algorithm definition information includes an intent learning algorithm for estimating the weights, The generation means calculates the weights using the intention learning algorithm and reflects the weights in the template to generate the optimized model.

2. 2. The information processing device according to claim 1, a receiving unit configured to receive a parameter specification for changing at least a part of the index of the optimization problem; The generating means generates the optimization model by using the template that has been modified using the received parameters.

3. 3. The information processing device according to claim 1, The generation means is an information processing device that inputs data of the data items defined by the item definition information, and optimizes the input data using the template to generate the optimized model.

4. 4. The information processing device according to claim 1, The template includes view definition information that specifies a display mode of information regarding an analysis result based on the optimization model.

5. 5. The information processing device according to claim 1, The information processing device, wherein the templates are provided for each item to be optimized by the optimization model.

6. 6. The information processing device according to claim 1, The information processing device, wherein in the template, the sets of multiple indicators are provided for different optimization purposes.

7. 7. The information processing device according to claim 1, The information processing device, wherein the generation means uses a prediction result by a prediction model as one of the data items of the input data of the optimization model.

8. An information processing device, An optimization model is generated using a template that defines an objective function and constraint conditions that are indices for the optimization problem; the template includes item definition information that identifies data items to be input into the objective function and the constraint condition, and algorithm definition information that defines algorithms for the objective function and the constraint condition; the template includes a weight indicating a user's intention for each index of the optimization problem; The algorithm definition information includes an intent learning algorithm for estimating the weights, The information processing method, wherein the information processing device calculates the weights using the intention learning algorithm and reflects the weights in the template to generate the optimized model.

9. On the computer, A procedure for generating an optimization model using a template that defines an objective function and constraint conditions that are indicators of the optimization problem is executed; the template includes item definition information that identifies data items to be input into the objective function and the constraint condition, and algorithm definition information that defines algorithms for the objective function and the constraint condition; the template includes a weight indicating a user's intention for each index of the optimization problem; The algorithm definition information includes an intent learning algorithm for estimating the weights, a program for generating the optimization model by calculating the weights using the intention learning algorithm and reflecting the weights in the template in the generating step.

Citation Information

Patent Citations

  • Water resource planning method based on simulation and considering climate change

    CN112241607A

  • Multi-target process parameter intelligent optimization method based on multi-algorithm fusion

    CN112380760A

  • Job analytic system

    JP2000285128A

  • Transportation planning system

    JP2008114960A

  • Schedule table creation apparatus and program

    JP2009042799A