Systems and programs

The system automates the classification of product development functions using a large language model to provide unbiased results, addressing the bias in manual classification and reducing costs and delays.

JP7830749B1Active Publication Date: 2026-03-16SOFTBANK CORPORATION
View PDF 4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-07-10
Publication Date
2026-03-16

AI Technical Summary

Technical Problem

The manual classification of product development functions using the MVP framework is prone to bias due to evaluator subjectivity, leading to increased development costs, delays, and missed market opportunities.

Method used

A system and program that utilize an acquisition unit, classification unit, and output unit to automatically classify product development functions based on basic guidelines, priority information, and functional information, using a large language model to provide unbiased classification results.

Benefits of technology

The system provides unbiased classification results, allowing for appropriate resource allocation and reducing development costs and delays, ensuring timely market entry.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007830749000001_ABST
    Figure 0007830749000001_ABST
Patent Text Reader

Abstract

The goal is to obtain unbiased classification results when classifying features planned for implementation in a product using the MVP framework. [Solution] The system (100) of this disclosure includes an acquisition unit (41) that acquires basic information indicating basic guidelines in a product development plan for developing a product, priority information indicating the priority of each of several development conditions that are in a trade-off relationship with each other, and functional information relating to several functions that are planned to be implemented in the product; a classification unit (42) that, based on the basic information, priority information, and functional information, classifies each of the several functions into one of several classifications indicating the necessity of development in the product development plan, using a learning model such as a large-scale language model; and an output unit (2) that outputs the results of the classification by the classification unit.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to a system and a program.

Background Art

[0002] Conventionally, in product development, in order to determine the scope of development and the priority of development, using the MVP (Minimum Viable Product) framework, the functions planned to be implemented in the product are classified into one of "Must (Essential)", "Need (Required)", and "Want (Desired)" according to the necessity of development (see Non-Patent Document 1).

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Means for Solving the Problems

[0004] A system according to an aspect of the present disclosure includes an acquisition unit that acquires basic information indicating basic guidelines in a product development plan for developing a product, ranking information indicating the priority of each of a plurality of development conditions that are in a trade-off relationship with each other, and function information regarding a plurality of functions planned to be implemented in the product; a classification unit that classifies each of the plurality of functions into one of a plurality of classifications indicating the necessity of development in the product development plan based on the basic information, the ranking information, and the function information; and an output unit that outputs the result of classification by the classification unit. It is provided with.

[0005] A program according to one form of this disclosure includes an acquisition process that acquires basic information indicating basic guidelines in a product development plan for developing a product, priority information indicating the priority of each of several development conditions that are in a trade-off relationship with each other, and functional information relating to several functions that are planned to be implemented in the product; a classification process that classifies each of the several functions into one of several classifications indicating the necessity of development in the product development plan, based on the basic information, the priority information, and the functional information; and an output process that outputs the results of the classification process. [Brief explanation of the drawing]

[0006] [Figure 1] This is a block diagram showing the functional configuration of a system according to an embodiment of this disclosure. [Figure 2] This figure shows an example of a prompt that the system according to the same embodiment inputs to a large-scale language model. [Figure 3] This flowchart shows an example of the processing flow performed by the system according to the present embodiment. [Figure 4] This is an image diagram showing an example of the output of the output unit of the system according to the same embodiment. [Figure 5] This is a block diagram showing the functional configuration of a system according to a modified example of the same embodiment. [Modes for carrying out the invention]

[0007] Hereinafter, one embodiment of this disclosure will be described in detail with reference to the drawings. For ease of understanding, the background and challenges of this disclosure will be described first, followed by a detailed description of the disclosure.

[0008] Traditionally, in product development, the Minimum Viable Product (MVP) framework has been used to classify features planned for implementation in the product into Must, Need, or Want categories according to their development necessity, in order to determine the scope and priority of development (see Patent Document 1).

[0009] The classification of each function has historically been done manually. As a result, the classification process could be biased due to the evaluator's subjectivity and organizational position. For example, evaluators tended to classify many functions as "Must," rendering the classification ineffective. Such cases can lead to significant business risks, such as increased development costs, delays in product development, and missed opportunities to bring products to market.

[0010] <System Overview> The system in this disclosure includes an acquisition unit that acquires basic information indicating the basic guidelines in a product development plan for developing a product, priority information indicating the priority of each of several development conditions that are in a trade-off relationship with each other, and functional information regarding several functions that are planned to be implemented in the product. A classification unit that classifies each of the multiple functions into one of the multiple classifications indicating the need for development in the product development plan, based on the basic information, the ranking information, and the function information. An output unit that outputs the results of the classification performed by the classification unit, It is equipped with.

[0011] With the above configuration, the classification unit automatically performs the classification of functions, resulting in unbiased classification results. Product development planners and others can then appropriately allocate resources to the development of each function based on the obtained classification results. As a result, the increase in product development costs can be suppressed, and development delays and the resulting loss of market entry timing can be prevented.

[0012] <Configuration of System 100> Hereinafter, a system 100 according to an embodiment of this disclosure will be described with reference to the drawings. Figure 1 is a block diagram showing the functional configuration of the system 100 according to this embodiment. As shown in Figure 1, the system 100 according to this embodiment is configured as a single user terminal. The user terminal may be, for example, a stationary PC (Personal Computer), a laptop PC, a mobile terminal such as a smartphone or tablet, an HMD (Head Mount Display), or a wearable terminal such as a smartwatch. The system 100 according to this embodiment includes an input unit 1, a display unit 2, a storage unit 3, and a control unit 4. Each of the units 1 to 4 is electrically connected to each other via a communication bus.

[0013] [Input section 1] Input unit 1 accepts various types of information input from the user. Input unit 1 is composed of, for example, a keyboard, a pointing device (mouse, etc.), a touch panel, etc. Input unit 1 may also be composed of a camera, a microphone, etc. In other words, input unit 1 may extract various types of information from images, audio, etc.

[0014] [Display section 2] The display unit 2 is the output unit in this embodiment. The display unit 2 is composed of a display. That is, the display unit 2 displays the classification results by the classification unit 42, which will be described later. The application for the display unit 2 to display the classification results may be a browser or an application dedicated to system 100. The display unit 2 may also be composed of a speaker, terminals, a communication module, etc. That is, the display unit 2 may output the classification results in the form of sound, signals, data, etc.

[0015] [Storage section 3] The storage unit 3 is for storing various data. The storage unit 3 according to the present embodiment is composed of, for example, a flash memory, an HDD (Hard Disc Drive), or a magneto-optical disk. The storage unit 3 according to the present embodiment stores the program 31. The storage unit 3 according to the present embodiment further stores the large language model 32. Note that the large language model 32 may be stored in another device provided outside the system 100 (it does not have to be incorporated into the system 100). Also, the storage unit 3 may be divided into a storage unit for storing the program 31 and a storage unit for storing the large language model 32.

[0016] (Program 31) The program 31 describes the operations of the control unit 4 of the system 100 (the content and execution order of various processes executed by the control unit 4).

[0017] (Large language model 32) The large language model 32 is constructed to output classification results for a plurality of functions when basic information, function information, and ranking information are input. The large language model 32 according to the present embodiment classifies each function into one of "Must (essential)", "Need (necessary)", and "Want (desired)". Note that the large language model 32 may classify each function into any of two or four or more classifications. Also, the prompt P is input to the large language model 32. As shown in FIG. 2, the prompt P describes the request content for the large language model 32 and the ranking information. If a description to output the reason for classification when classifying the function is included in the request content, the large language model 32 outputs the reason for classification of each function together when outputting the classification result. Also, when analyzing each function included in the function information and outputting the classification result, the large language model 32 outputs the prerequisite conditions between functions or the interdependent relationship between functions together.

[0018] 〔Control unit 4〕 The control unit 4 is composed of a processor and a memory. The processor executes the instructions described in the program 31. The processor is composed of an arithmetic unit, registers, and / or peripheral circuits, etc. The memory temporarily stores the program 31 and the data processed by the program 31, etc., and is, for example, a volatile memory such as a DRAM (Dynamic Random Access Memory). The control unit 4 includes an acquisition unit 41, a classification unit 42, and an output processing unit 43. The control unit 4 according to the present embodiment further includes a reception unit 44, a reception determination unit 45, and a condition determination unit 46.

[0019] FIG. 3 is a flowchart showing an example of the flow of various processes executed by the system 100 according to the present embodiment. The control unit 4 executes an acquisition process S1, a classification process S2, and an output process S3 in the flow shown in FIG. 3 according to the program 31. The control unit 4 according to the present embodiment further executes a reception process S4, a reception determination process S5, and a condition determination process S6.

[0020] (Reception process S4 · Reception unit 44) In the initial reception process S4, the reception unit 44 imports basic information and function information. The "basic information" is information indicating the basic guidelines in the product development plan for developing a product. The basic information may be, for example, a product charter, an inception deck, etc. The "function information" is information regarding a plurality of functions to be implemented in the product. Note that the reception unit 44 may be configured to be able to import new basic information and function information after importing the basic information and the function information. That is, the reception unit 44 may be configured to be able to receive changes to the basic information and the function information.

[0021] Furthermore, in the reception process S4, the reception unit 44 puts the system 100 into a state where it can accept input or modification of ranking information. "Ranking information" is information indicating the priority of each of the multiple development conditions. "Multiple development conditions" may be multiple conditions considered when proceeding with product development. "Multiple development conditions" may be, for example, adherence to the concept, quality, budget, time, security, etc. In the reception process S4 according to this embodiment, first, the reception unit 44 displays the interface screen 21 on the display unit 2. The interface screen 21 is for accepting the setting of ranking information. In the reception process S4 according to this embodiment, the reception unit 44 displays the interface screen 21 on the display unit 2, for example, as shown in Figure 4. The interface screen 21 includes a ranking setting area 211 and a result display area 212. The ranking setting area 211 according to this embodiment is a trade-off slider. That is, the ranking setting area 211 has a slider for setting the number indicating the priority of the development conditions for each of the multiple development conditions. The numbers indicating priority can be, for example, five numbers such as 1, 2, 3, 4, and 5. A higher number may indicate a condition that is considered more important in development. The result display area 212 will be described later. The ranking setting area 211 may also allow prioritization by means other than the trade-off slider (e.g., direct numerical input, buttons, etc.). The interface screen 21 may have an area for displaying basic information. The reception unit 44 can accept input or modification of ranking information at any time. That is, the user can operate any slider in the ranking setting area 211 at any time. Setting priorities does not necessarily mean that duplicate rankings for multiple development conditions are excluded. That is, the same ranking may be set for multiple development conditions when setting priorities. For example, as described later, the same condition constraints may be applied, resulting in the same ranking being set for multiple development conditions.

[0022] (Reception decision processing S5, reception decision unit 45) The reception determination process S5 is executed in parallel with the execution of the reception process S4. In the reception determination process S5, the reception determination unit 45 determines whether or not the ranking information has been entered or changed. The reception determination process S5 is repeated as long as the ranking information has not been entered or changed.

[0023] (Conditional judgment processing S6, conditional judgment unit 46) After the acceptance determination process S5 determines that the input or modification of basic information and ranking information has been accepted, the process moves to the condition determination process S6. In the condition determination process S6, the condition determination unit 46 determines whether the ranking information accepted in the acceptance process S4 satisfies predetermined constraints. "Constraints" include, for example, a sum constraint, a maximum selection constraint, a hierarchical constraint, a tie-rank constraint, or a combination thereof. If the condition determination unit 46 determines that the ranking information does not satisfy the predetermined constraints, it performs error control according to the content of the constraints.

[0024] A "total constraint" is a constraint condition in which the sum of the numbers indicating the priority of each of the multiple development conditions is less than or equal to a predetermined value. If the priority is on a 5-point scale from 1 to 5, and the development conditions are of the 5 types: adherence to concept, quality, budget, time, and security, the predetermined value can be, for example, 20. When the constraint condition is a total constraint, the condition judgment unit 46 calculates the sum of the numerical values ​​indicating the priority each time the reception unit 44 accepts input or change of priority, and compares it with the predetermined value. In addition, the condition judgment unit 46 performs at least one of the following as error control: for example, notification and setting restriction. "Notification" includes, for example, having the output unit output a notification that the predetermined value has been exceeded. "Setting restriction" includes, for example, forcibly lowering the most recently set priority so that its sum is less than or equal to a predetermined value, or restricting setting operations to raise the priority (only accepting operations to lower the priority and bring the sum below a predetermined value).

[0025] The "maximum selection constraint" is a constraint that limits the number of development conditions that can be set to a priority of 5 (most important item) to a predetermined number or less. If there are 5 levels of priority from 1 to 5 and 5 types of development conditions, the predetermined number can be, for example, 2. When the constraint is the maximum selection constraint, the condition judgment unit 46 counts the number of development conditions set to a priority of 5 each time the reception unit 44 accepts input or change of priority and compares it with the predetermined number. The condition judgment unit 46 also performs at least one of the following as error control: for example, notification and setting restriction. "Notification" includes having the output unit output a notification that the predetermined number has been exceeded. "Setting restriction" includes, for example, forcibly lowering the priority of the most recently set to 5 to 4 or less, or restricting the upper limit of the priority of development conditions other than those already set to a priority of 5 to 4 (screen control that prevents the slider from sliding beyond 4).

[0026] A "hierarchical constraint" is a constraint that limits the number of development conditions that can be set for each priority level to a predetermined number or less. If there are five priority levels from 1 to 5 and five types of development conditions, the predetermined number for priority level 5 (most important item) can be, for example, 2, for priority level 4 (semi-important item) it can be, for example, 2, and for priorities 1 to 3 it can be, for example, 1. When the constraint is a hierarchical constraint, the condition judgment unit 46 counts the number of development conditions set for each priority level each time the reception unit 44 receives input or a change in priority level and compares it with a predetermined value. The condition judgment unit 46 also performs error control, for example, by providing a notification. The "notification" may include, for example, a notification that a predetermined value has been exceeded, which is output to the output unit.

[0027] A "same priority constraint" is a constraint that prohibits setting the same priority for multiple development conditions. When the constraint is a same priority constraint, the condition determination unit 46 checks for duplicate priority settings each time the reception unit 44 receives input or changes to the priority settings. The condition determination unit 46 also performs at least one of the following as error control: for example, notification and setting restrictions. "Notification" includes having the output unit output a notification that a predetermined number has been exceeded. "Setting restrictions" include, for example, forcibly changing the value of the most recently set priority to a value that does not overlap with others, and performing screen control to restrict the number of settings to the same number as the priority already set for other development conditions.

[0028] (Acquisition process S1, acquisition unit 41) After the conditional judgment process S6, the process moves to the acquisition process S1. In the acquisition process S1, the acquisition unit 41 acquires basic information, ranking information, and function information. As described above, the reception unit 44 can accept changes to the ranking information at any time. Therefore, each time the acquisition unit 41 accepts a change to the ranking information in the reception process S4, it acquires the changed ranking information. Also, as described above, the control unit 4 in this embodiment executes the conditional judgment process S6 before executing the acquisition process S1. Therefore, each time the acquisition unit 41 accepts a change to the ranking information in the reception process S4 and determines in the conditional judgment process S6 that the changed ranking information satisfies the constraints, it acquires the changed ranking information. As will be described in detail later, the classification unit 42 in this embodiment causes the large-scale language model 32 to classify the functions. Therefore, in the acquisition process S1 in this embodiment, the acquisition unit 41 structures the acquired basic information into a format that the large-scale language model 32 can understand.

[0029] (Classification processing S2 / classification unit 42) After the acquisition process S1, the process moves to the classification process S2. In the classification process S2, the classification unit 42 classifies each of the multiple functions into one of several classifications indicating the necessity of development in the product development plan, based on the basic information, ranking information, and function information. As described above, the acquisition unit 41 acquires the changed ranking information each time it receives a change in the ranking information in the reception process S4 and determines in the condition judgment process S6 that the changed ranking information satisfies the constraints. Therefore, the classification unit 42 classifies each of the multiple functions each time the acquisition unit 41 acquires the changed ranking information. Also, as described above, the control unit 4 in this embodiment executes the condition judgment process S6 before executing the acquisition process S1. Therefore, in the classification process S2 in this embodiment, the classification unit 42 classifies each of the multiple functions based on the basic information, function information, and ranking information that has been determined to satisfy the constraints.

[0030] In classification processing S2, the classification unit 42 inputs functional information and structured basic information, along with prompt P, to the large-scale language model 32, causing it to classify multiple functions. As described above, when the large-scale language model 32 outputs the classification results, it also outputs the preconditions between functions or the interdependencies between functions for multiple functions. In classification processing S2 according to this embodiment, the classification unit 42 obtains the preconditions between functions or the interdependencies between functions for multiple functions from the large-scale language model 32. Also, as described above, when the large-scale language model 32 outputs the classification reason for each function, it also outputs the reason for classifying each function. For this reason, the classification unit 42 further obtains the reason for classifying each function from the large-scale language model 32.

[0031] (Output processing S3 · Output processing unit 43) After the classification process S2, the process moves to the output process S3. In the output process S3, the output processing unit 43 causes the output unit to output the classification results from the classification unit 42. As described above, the output unit in this embodiment is the display unit 2. Therefore, the display unit 2 displays the classification results from the classification unit 42. In the output process S3 in this embodiment, the output processing unit 43 displays the classification results in the result display area 212 of the interface screen 21. As shown in Figure 4, the result display area 212 (display unit 2) in this embodiment displays the classification results (MVP) in a list format. That is, the result display area 212 in this embodiment displays the names of the multiple functions that were classified (function names) and the classification (MVP evaluation) of each function in association with each other. In addition, the result display area 212 in this embodiment displays the category (major category) of each function, the details of each function, and the man-hours. The output processing unit 43 may be configured to display the result display area 212 on a screen different from the interface screen 21.

[0032] Furthermore, as described above, the classification unit 42 according to this embodiment obtains the classification reason for each function from the large-scale language model 32. For this reason, the output unit according to this embodiment, i.e., the result display area 212 of the display unit 2, displays the classification reason (MVP evaluation reason / basis) together with the classification result, as shown in Figure 4. Also, as described above, the classification unit 42 according to this embodiment obtains preconditions between functions or interdependencies between functions from the large-scale language model 32. For this reason, the display unit 2 according to this embodiment may output the preconditions or interdependencies together with the classification result. If the display unit 2 is composed of terminals, communication modules, etc., the display unit 2 may output the classification result data to another system that manages the product development plan (e.g., a project management tool) in a format that the other system can use.

[0033] <Effects and benefits of System 100> Because the pre-trained model automatically classifies the functions, unbiased classification results (not influenced by human subjectivity or organizational perspectives) can be obtained. Based on the resulting classification, product development planners can then narrow down the functions to be included in the product to only those that are truly necessary and allocate resources quickly and appropriately to the development of each function. As a result, the increase in development costs can be suppressed, and delays in product development and the resulting loss of market entry timing can be prevented.

[0034] <Variation>

[0035] This disclosure is not limited to the embodiments described above, and various modifications are possible within the scope of the claims.

[0036] For example, as shown in Figure 5, system 100 may include a user terminal 100a and a server 100b. In this case, the user terminal 100a and the server 100b communicate with each other via their respective communication units 5a and 5b, and a communication network (not shown). The communication network may be a mobile communication system such as 3G, 4G, 5G, or 6G, LTE (Long Term Evolution), Wi-Fi (registered trademark), or an in-house LAN. In this case, the display unit 2 of the user terminal 100a or the communication unit 5b of the server becomes the output unit.

[0037] Furthermore, the program 31 may be recorded on one or more computer-readable recording media, not temporary ones. These recording media may or may not be provided by the device. In the latter case, the program 31 may be supplied to the device via any wired or wireless transmission medium.

[0038] Furthermore, some or all of the functions of each of the control blocks 41 to 44 described above can also be realized by logic circuits. For example, an integrated circuit in which logic circuits functioning as each of the control blocks described above are formed is also included in the scope of this disclosure. In addition, it is also possible to realize the functions of each of the control blocks described above by, for example, a quantum computer.

[0039] <Summary> This disclosure includes at least the following aspects:

[0040] The system according to aspect 1 of this disclosure includes an acquisition unit that acquires basic information indicating basic guidelines in a product development plan for developing a product, priority information indicating the priority of each of a plurality of development conditions that are in a trade-off relationship with each other, and functional information relating to a plurality of functions that are planned to be implemented in the product; a classification unit that classifies each of the plurality of functions into one of a plurality of classifications indicating the necessity of development in the product development plan based on the basic information, the priority information, and the functional information; and an output unit that outputs the results of the classification by the classification unit.

[0041] According to the above configuration, when classifying the features to be implemented in a product using the MVP framework, an unbiased classification result can be obtained.

[0042] In the system according to Embodiment 2 of this disclosure, in Embodiment 1, the acquisition unit structures the received basic information into a format that can be understood by a large-scale language model, and the classification unit inputs the structured basic information, along with prompts, into the large-scale language model to classify the multiple functions.

[0043] With the above configuration, it becomes easier to reflect changes in ranking information by using a large-scale language model and prompts. Furthermore, with the above configuration, it becomes easier to output classification results in a manner desired by the user by using a large-scale language model and prompts. In addition, with the above configuration, supplementary information (such as classification reasons) of the classification results can also be output to the large-scale language model by using a large-scale language model and prompts.

[0044] In the system according to aspect 3 of this disclosure, in aspect 2, the classification unit obtains preconditions between functions or interdependencies between functions for the multiple functions from the large-scale language model, and the output unit outputs the preconditions or interdependencies together with the classification results.

[0045] According to the above configuration, the product development planner and others will be able to determine the development order of each function, taking into account the preconditions or the aforementioned interdependencies. As a result, efficient development with fewer rework issues will be possible.

[0046] In the system according to aspect 4 of the present disclosure, in aspect 2 or 3, the classification unit further obtains the classification reason for each function from the large-scale language model, and the output unit outputs the classification reason together with the classification result.

[0047] According to the above configuration, the resulting classification results will be highly transparent, accompanied by a clear process and logic.

[0048] In the system according to aspect 5 of this disclosure, in any of aspects 1 to 3, the acquisition unit acquires the changed ranking information each time it receives a change in the ranking information, and the classification unit classifies each of the multiple functions each time the acquisition unit acquires the changed ranking information.

[0049] With the above configuration, users can obtain classification results at their desired time. Furthermore, with the above configuration, it is possible to simulate classification while fine-tuning the priority settings over a short period of time.

[0050] The system according to aspect 6 of this disclosure further comprises a condition determination unit that determines whether the acquired ranking information satisfies predetermined constraints, and the classification unit classifies each of the plurality of functions based on the ranking information that is determined to satisfy the constraints.

[0051] According to the above configuration, in order to proceed with processing (classifying multiple functions), it becomes necessary to set the priority of each of the multiple development conditions so that the constraints are met. As a result, users will set priorities more carefully than before, and a more unbiased classification result can be obtained.

[0052] In the system according to aspect 7 of this disclosure, in any of aspects 1 to 6, the output unit displays the classification results in a list format.

[0053] According to the above configuration, users will be able to easily understand the functions required for the product and the development priorities.

[0054] In any of the embodiments 1 to 7, the system relating to embodiment 8 of this disclosure is configured such that the output unit outputs the data of the classification results to another system that manages the product development plan in a format usable by that other system.

[0055] According to the above configuration, the classification results can be used directly in processes performed by other systems, making it easier to manage product development plans.

[0056] The program according to aspect 9 of the present disclosure, in any of aspects 1 to 3, causes the processor to perform an acquisition process to acquire basic information indicating basic guidelines in a product development plan for developing a product, priority information indicating the priority of each of a plurality of development conditions that are in a trade-off relationship with each other, and functional information relating to a plurality of functions that are planned to be implemented in the product; a classification process that classifies each of the plurality of functions into one of a plurality of classifications indicating the necessity of development in the product development plan based on the basic information, the priority information, and the functional information; and an output process that outputs the result of the classification performed by the classification process.

[0057] According to the above configuration, similar to the system relating to Embodiment 1 of this disclosure, it is possible to obtain unbiased classification results when classifying functions to be implemented in a product using the MVP framework. [Explanation of Symbols]

[0058] 100 Systems 1 Input section 2. Display Unit (Output Unit) 21 Interface screen 211 Ranking area 212 Results display area 3 Storage section 31 Programs 32 Large-scale language models 4. Control Unit 41 Acquisition Department 42 Classification Department 43 Output Processing Unit 44 Reception Department 45 Reception and Decision Department 46 Condition judgment section 100a User Terminal 100b Server 5a,5b Communication Department S1 acquisition process S2 Classification Processing S3 Output Processing S4 Reception Processing S5 Reception Decision Processing S6 Conditional Judgment Processing

Claims

1. An acquisition unit that acquires basic information indicating the basic guidelines in a product development plan for developing a product, priority information indicating the priority of each of several development conditions that are in a trade-off relationship with each other, and functional information regarding several functions that are planned to be implemented in the product. A classification unit that classifies each of the multiple functions into one of the multiple classifications indicating the need for development in the product development plan, based on the basic information, the ranking information, and the function information. An output unit that outputs the results of the classification performed by the classification unit, Equipped with, system.

2. The acquisition unit structures the received basic information into a format that can be understood by a large-scale language model. The classification unit inputs the structured basic information along with prompts to a large-scale language model and causes it to classify the multiple functions. The system according to claim 1.

3. The classification unit obtains, from the large-scale language model, preconditions between functions or interdependencies between functions for the multiple functions, The output unit outputs the preconditions or interdependencies together with the classification results. The system according to claim 2.

4. The classification unit further obtains the reason for classifying each function from the large-scale language model, The output unit outputs the reason for classification along with the classification result. The system according to claim 2.

5. The acquisition unit acquires the changed ranking information each time it receives a change in the ranking information. The classification unit classifies each of the multiple functions each time the acquisition unit acquires the changed ranking information. The system according to claim 1.

6. The system further includes a condition determination unit that determines whether the acquired ranking information satisfies predetermined constraints. The classification unit classifies each of the multiple functions based on the ranking information which is determined to satisfy the constraints. The system according to claim 5.

7. The output unit displays the classification results in a list format. The system according to claim 1.

8. The output unit outputs the classification result data to another system that manages the product development plan, in a format usable by that other system. The system according to claim 1.

9. In the processor, An acquisition process to obtain basic information that shows the basic guidelines in a product development plan for developing a product, priority information that shows the priority of each of several development conditions that have a trade-off relationship with each other, and functional information regarding several functions that are planned to be implemented in the product, A classification process that classifies each of the multiple functions into one of the multiple classifications indicating the need for development in the product development plan, based on the basic information, the ranking information, and the function information. An output process that outputs the results of the classification process described above, To execute program.

Citation Information

Patent Citations

  • Resource management system and resource management method

    JP2016024697A

  • Evaluation device, evaluation method and evaluation program

    JP2017228014A

  • Evaluation support device, evaluation support method, and evaluation support program

    JP2018147190A

  • system

    JP2025058266A