Design assistance system

Abstract

Provided is a design assistance system that is suitable for changing an editing element requiring human judgment in order to be changed.　A drawing creation assistance device 100 includes a first element which is an editable editing element in design information and requires human judgment in order to be changed, and a second element which is affected by changing of the first element or affects changing of the first element. A request including a request to generate candidate information about one or a plurality of candidates for changing the first element is input to a large-scale language model, and candidate information output from the large-scale language model is acquired for the request. In this way, the designer can simply choose from among the change candidates, and it thus becomes easier to change the editing element requiring human judgment in order to be changed.

Description

Design Support System

[0001] The present invention relates to a design support system, and more particularly to a design support system suitable for changing edit elements that require human judgment to change.

[0002] 2. Description of the Related Art Conventionally, a technique for supporting the creation of architectural drawings using a large language model is known, for example, from Japanese Patent Laid-Open No. 2003-222999.

[0003] The technology described in Patent Document 1 transmits a request to a large-scale language model server, the request including an edit element and a request to generate related element information about related elements of the edit element and judgment necessity information about whether a change to the related element requires human judgment, and acquires the related element information and judgment necessity information output from the large-scale language model server in response to the request. If it is determined based on the acquired judgment necessity information that a change to the related element requires human judgment, the technology displays the related element in a specific manner based on the acquired related element information.

[0004] Patent No. 7341580 (

[0119] )

[0005] However, the technology described in Patent Document 1 simply displays editing elements that require human judgment to change in a specific manner, so it is unclear how to change editing elements that require human judgment to change, or what to consider when making changes.

[0006] Therefore, the present invention has been made with a focus on the unresolved problems of the conventional technology, and aims to provide a design support system that is suitable for changing editing elements that require human judgment to change.

[0007] [Invention 1] In order to achieve the above object, the design support system of Invention 1 comprises a request input means for inputting a request to a large-scale language model, the request including first element information on a first element that is an editable editing element in design information and whose modification requires human judgment, and second element information on a second element that is affected by or modifies the first element, the request including a request to generate candidate information on one or more candidates for modifying the first element, and a candidate information acquisition means for acquiring candidate information output from the large-scale language model in response to the request.

[0008] Here, the first element information can be configured as, for example, information for identifying the first element (e.g., link information such as a name, number, ID, code, or URL). The first element information can also be configured as, for example, characters, numbers, figures, codes, symbols, images, sounds, or other information. The first element information can also be configured as keywords related to the first element (e.g., one or more keywords indicating part of the name of the first element). The same applies to the design support systems of Inventions 2 to 8.

[0009] The second element information can be configured as, for example, information for identifying the second element (e.g., link information such as a name, number, ID, code, or URL). The second element information can be configured as, for example, characters, numbers, figures, codes, symbols, images, sounds, or other information. The second element information can be configured as keywords related to the second element (e.g., one or more keywords indicating part of the name of the second element). The same applies to the design support systems of Inventions 2 to 4 below.

[0010] The second element information includes, for example, element information relating to editable elements in the design information, or element information relating to elements other than the edit elements. The same applies to the design support systems of the second to fourth aspects below.

[0011] Furthermore, this system may be realized as a single device, terminal, or other equipment, or as a network system in which multiple devices, terminals, or other equipment are connected so that they can communicate with each other. In the latter case, each component may belong to any of the multiple devices, etc., as long as they are connected so that they can communicate with each other. The same applies to the design support systems of Inventions 2 to 8 below.

[0012] [Invention 2] Furthermore, the design support system of Invention 2 comprises a request input means for inputting a request to a large-scale language model, the request including first element information on a first element that is an editable editing element in design information and whose change requires human judgment, and second element information on a second element that is affected by or affects the change to the first element, the request including a request to generate impact information on one or more impacts when the first element is changed, and impact information acquisition means for acquiring impact information output from the large-scale language model in response to the request.

[0013] Here, the impact of changing the first element includes, for example, (1) edit elements that need to be changed or can be changed, or the number or amount of change thereof, (2) an increase or decrease in the asset value related to the design object, (3) the presence or absence of risk related to the design object, or an increase or decrease in risk, and (4) inconvenience to users of the design object. The same applies to the design support system of Invention 6 below.

[0014] Here, examples of cases where the first element is changed include cases where a specific change is made to the first element. Examples of specific changes include (1) changing a specific element among the elements that make up the first element, (2) changing the first element or some of the elements that make up the first element to specific content, and (3) changing the first element or some of the elements that make up the first element so that the relationship with the second element, other edit elements, or some of the elements that make up the first element becomes specific. The same applies to the design support systems of Inventions 3, 4, and 6 to 8.

[0015] [Invention 3] Furthermore, the design support system of Invention 3 comprises a request input means for inputting a request to a large-scale language model, the request including first element information on a first element that is an editable editing element in design information and whose change requires human judgment, and second element information on a second element that is affected by or affects the change to the first element, the request including a request to generate consideration information on one or more matters that should be taken into consideration when changing the first element, and a consideration information acquisition means for acquiring the consideration information output from the large-scale language model in response to the request.

[0016] Here, matters to be considered when changing the first element include, for example, precautions regarding planning, development, design, manufacturing, construction, operation, use, management, or maintenance when changing the first element. The same applies to the design support system of Invention 7 below.

[0017] [Invention 4] Furthermore, the design support system of Invention 4 comprises a request input means for inputting a request to a large-scale language model, the request including first element information on a first element that is an editable editing element in design information and requires human judgment to change, and second element information on a second element that is affected by or affects the change to the first element, and the request including a request to generate risk information on one or more risks when the first element is changed, and a risk information acquisition means for acquiring risk information output from the large-scale language model in response to the request.

[0018] Here, the risks involved in changing the first element include, for example, (1) the risk of not complying with or no longer complying with laws and standards, (2) the risk of the design object collapsing or being damaged due to insufficient strength, durability, earthquake resistance, etc., (3) the risk of the design object becoming obsolete due to deterioration over time, (4) the risk of increased costs, (5) the risk of reduced convenience for users of the design object, (6) the risk of fire, flood, wind damage, earthquake, and other natural disasters, (7) the risk of a decrease in the asset value related to the design object, and (8) the risk of causing social problems. The same applies hereinafter to the design support system of Invention 8.

[0019] [Invention 5] Furthermore, the design support system of Invention 5 comprises a request input means for inputting a request to a large-scale language model, the request including first element information about a first element that is an editable editing element in design information and requires human judgment to change, and a request to generate candidate information about one or more candidates for changing the first element, and a candidate information acquisition means for acquiring candidate information output from the large-scale language model in response to the request.

[0020] [Invention 6] Furthermore, the design support system of Invention 6 comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and a request for generating impact information relating to one or more impacts when the first element is changed, and an impact information acquisition means for acquiring impact information output from the large-scale language model in response to the request.

[0021] [Invention 7] Furthermore, the design support system of Invention 7 comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and the request including a request to generate consideration information relating to one or more matters that should be taken into consideration when changing the first element, and a consideration information acquisition means for acquiring the consideration information output from the large-scale language model in response to the request.

[0022] [Invention 8] Furthermore, the design support system of Invention 8 comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and the request including a request to generate risk information relating to one or more risks when changing the first element, and a risk information acquisition means for acquiring risk information output from the large-scale language model in response to the request.

[0023] [Invention 9] Furthermore, the design support system of Invention 9 is the design support system of any one of Inventions 1 and 5, wherein the information about the first element is information about the first element before the change, among the first elements before and after the change, when the first element is changed, and the candidate information is information about the first element after the change, among the first elements before and after the change.

[0024] [Invention 10] Furthermore, in the design support system of invention 10, in any one of the design support systems of inventions 1 to 8, the design information is design information for designing a building.

[0025] As described above, according to the design support system of inventions 1 or 5, candidate information regarding one or more candidates for changing the first element is obtained, making it easier to change edited elements that require human judgment to change.

[0026] Furthermore, according to the design support system of Inventions 2 or 6, impact information regarding one or more impacts when changing the first element is obtained, so that the impact can be taken into consideration when changing an edited element that requires human judgment to change.

[0027] Furthermore, according to the design support system of Inventions 3 or 7, information on considerations regarding one or more matters that should be taken into account when changing the first element is obtained, so that precautions, etc. can be taken into account when changing an edited element that requires human judgment to change.

[0028] Furthermore, according to the design support system of Inventions 4 or 8, risk information regarding one or more risks when changing the first element is obtained, so that risks can be taken into consideration when changing an edited element that requires human judgment to change.

[0029] 1 is a diagram showing the hardware configuration of a drawing creation support device 100. FIG. 1 is a diagram showing the structure of edit history data. FIG. 2 is a diagram showing the structure of CAD data of a plan drawing. A flowchart showing a trained model generation process. A block diagram showing the process of generating and using a trained model. A flowchart showing post-change dimension estimation process. A plan drawing before the dimensions of the PS are changed. A screen for inputting a second element. A screen for displaying change candidates for the first element. A plan drawing after the dimensions of the PS are changed. A plan drawing before it is assumed that a piano will be delivered. A screen for inputting a second element. A screen for displaying change candidates for the first element. A plan drawing after it is assumed that a piano will be delivered. FIG. 2 is a diagram showing the structure of influence data. A flowchart showing influence estimation process. A screen for displaying change candidates for the first element. A screen for displaying influences corresponding to the first change candidate. A screen for displaying change candidates for the first element. A screen for displaying influences corresponding to the first change candidate. FIG. 3 is a diagram showing the structure of considerations data. A flowchart showing considerations estimation process. A screen for displaying change candidates for the first element. A screen for displaying considerations corresponding to the first change candidate. A screen for displaying change candidates for the first element. A screen for displaying considerations corresponding to the first change candidate. FIG. 1 is a diagram showing the structure of risk data; FIG. 2 is a flowchart showing risk estimation processing; FIG. 3 is a screen displaying modification candidates for a first element; FIG. 4 is a screen displaying risks corresponding to the first modification candidates; FIG. 5 is a screen displaying modification candidates for the first element; FIG. 6 is a screen displaying risks corresponding to the first modification candidates.

[0030] [First Embodiment] A first embodiment of the present invention will be described below, with reference to Figures 1 to 14 showing this embodiment.

[0031] First, the configuration of this embodiment will be described. FIG. 1 is a diagram showing the hardware configuration of a drawing creation support device 100.

[0032] As shown in FIG. 1, the drawing creation support device 100 comprises a CPU (Central Processing Unit) 30 that controls calculations and the entire system based on a control program, a ROM (Read Only Memory) 32 that stores the control program and the like for the CPU 30 in advance in a predetermined area, a RAM (Random Access Memory) 34 that stores data read from the ROM 32 and the like and calculation results required in the calculation process of the CPU 30, and an I / F (Interface) 38 that mediates the input and output of data to and from external devices, and these are connected to each other and capable of sending and receiving data by a bus 39, which is a signal line for transferring data.

[0033] The I / F 38 is connected to external devices such as an input device 40 consisting of a keyboard, mouse, etc. that can input data as a human interface, a storage device 42 that stores data, tables, etc. as files, and a display device 44 that displays a screen based on an image signal.

[0034] CAD (Computer Aided Design) software and BIM (Building Information Modeling) software (hereinafter collectively referred to as "CAD software") are installed in the storage device 42. CAD software is software that assists in the creation of drawings in response to user operations. When a request is made to start the CAD software, the CPU 30 starts a CAD program stored in a predetermined area of ​​the ROM 32 and executes processing in accordance with the program. A designer can start the CAD software to create plan drawings, detailed floor plans, and other architectural drawings.

[0035] Next, we will explain the data structure of the storage device 42. Figure 2 shows the structure of the edit history data.

[0036] The storage device 42 stores edit history data as shown in Fig. 2. The edit history data includes element information 400 relating to a first element, which is an editable edit element in the design information and requires human judgment to change, element information 402 relating to a second element that is affected by or affects the change to the first element, and dimension information 404 relating to the dimensions after the change when the first element is changed.

[0037] Whether an edit element requires human judgment for element modification can be determined, for example, by the techniques described in Patent Documents 1

[0076] to

[0089] . That is, edit elements requiring human judgment for modification may be set by human judgment. Alternatively, edit elements with two or more change patterns before and after modification may be extracted based on the pre-modification CAD data and the post-modification CAD data, and the extracted edit elements may be set as edit elements requiring human judgment for modification. Based on the results, judgment necessity data is created in which edit elements are registered in association with information on whether or not human judgment is required for modification, and the data is stored in the storage device 42. Then, edit elements determined to require human judgment for modification based on the judgment necessity data in the storage device 42 are treated as first elements in the edit history data.

[0038] The second element includes an editable edit element in the design information or another element (e.g., furniture to be installed in the room or other virtual elements). The edit history data can be created, for example, by recording a change made by the designer to the first element in relation to the second element as edit history data.

[0039] The first and second lines in Figure 2 represent one edit history. The first and second lines in Figure 2 register a "700mm wide left closet" and a "700mm wide center closet" as the first element, a "270mm wide PS" as the second element, and "650" and "650" as the changed dimensions, respectively. This edit history indicates that the 700mm wide left closet and the 700mm wide center closet are elements that require human judgment to change, the 270mm wide PS (pipe shaft) is an element that affects the change to the first element, and that in changing the PS dimension to 270mm, the designer changed the width of the left closet from 700mm to 650mm and the width of the center closet from 700mm to 650mm.

[0040] The third line in Figure 2 is an edit history. The first element in the third line is a "left closet with a width of 700 mm," the second element is a "PS with a width of 270 mm," and the new dimension is "600." This edit history indicates that the designer changed the width of the left closet from 700 mm to 600 mm when changing the PS dimension to 270 mm.

[0041] The fourth line in Figure 2 is an edit history. The first element in the fourth line is a "700mm wide central closet," the second element is a "270mm wide PS," and the new dimension is "600." This edit history indicates that the designer changed the central closet's width from 700mm to 600mm when changing the PS dimension to 270mm.

[0042] The fifth line in Figure 2 is one edit history. The fifth line in Figure 2 has "a toilet with a width of 920mm" registered as the first element, "a piano with a width of 120mm" registered as the second element, and "870" registered as the changed dimension. This edit history shows that the designer changed the width of the toilet from 920mm to 870mm to prevent interference between the piano and the toilet wall when bringing in the 120mm wide piano.

[0043] The sixth line in Figure 2 is one edit history. The sixth line in Figure 2 has the first element "a toilet with a width of 920 mm" registered, the second element "a piano with a width of 130 mm" registered, and the changed dimensions "855" registered. This edit history shows that the designer changed the width of the toilet from 920 mm to 855 mm to prevent interference between the piano and the toilet wall when bringing in the 130 mm wide piano.

[0044] The seventh line in Figure 2 is one edit history. The seventh line in Figure 2 has "a toilet with a width of 920mm" registered as the first element, "a piano with a width of 140mm" registered as the second element, and "840" registered as the changed dimension. This edit history shows that the designer changed the width of the toilet from 920mm to 840mm to prevent interference between the piano and the toilet wall when bringing in a 140mm wide piano.

[0045] 3 is a diagram showing the structure of CAD data of a rectangular plan. As shown in FIG. 3, the storage device 42 stores CAD data of a rectangular plan and BIM data (hereinafter collectively referred to as "CAD data").

[0046] A rectangular plan is a detailed drawing of a building's cross section. CAD data for the plan is created by a designer using CAD software. The designer creates the plan in the CAD software by adding, editing, or deleting editable elements in the plan. In the example of Figure 3, the floor, wall, and ceiling edit elements are arranged in the area labeled "internal corridor," and the floor, wall, and ceiling edit elements are arranged in the area labeled "windbreak."

[0047] The storage device 42 stores CAD data of other architectural drawings and data of the trained model. The trained model is trained based on element information 400 related to a first element, element information 402 related to a second element, and dimension information 404 related to one or more post-change dimensions when the first element is changed. Since the editing history data is used for AI training, the storage device 42 stores a large amount of editing history data created in the past. This editing history data indicates that each edited element in the architectural drawing conforms to the specifications, and serves as training data for AI training.

[0048] Next, the operation of this embodiment will be described. Fig. 4 is a flowchart showing the trained model generation process.

[0049] 5 is a block diagram showing the steps of generating and using a trained model. The trained model generation process is a process executed to generate a trained model, and when executed by the CPU 30, the process proceeds to step S100, where an edit history data analysis process is executed, as shown in FIG. 4. In the edit history data analysis process, edit history data is read from the storage device 42, and element information 400, 402 and dimension information 404 are extracted from the read edit history data.

[0050] Next, the process proceeds to step S102. In step S102, as shown in FIG. 5 , a training dataset is generated based on the information extracted in step S100. The process then proceeds to step S104, where the generated training dataset is input into a training program and a trained model is generated by the training program. The training program includes pre-training parameters and hyperparameters, and performs training based on the input training dataset and hyperparameters to update the pre-training parameters. The trained model is then output as the training result.

[0051] The trained model includes trained parameters in which pre-trained parameters have been updated through training, and an inference program. The inference program inputs a first element and a second element, estimates one or more post-modification dimensions from the input first element and second element based on the trained parameters, and outputs the estimated post-modification dimensions. Note that the relationship between the input first element and second element and the post-modification dimensions is determined by AI training, and therefore, although it shows a similar trend to the content of past editing history data, there is ambiguity in that it does not necessarily match exactly. However, this ambiguity can be reduced by adjusting the amount of training data and the training accuracy.

[0052] 6 is a flowchart showing the post-change dimension estimation process. The post-change dimension estimation process is executed in response to a request from the user, and when executed by the CPU 30, as shown in FIG. 6, the process first proceeds to step S150.

[0053] In step S150, a second element is acquired. For example, the second element can be acquired from any of the edit elements in the architectural drawing (e.g., the selected edit element), or can be acquired by inputting a virtual element (e.g., furniture to be installed in the room) or other element.

[0054] Next, the process proceeds to step S152, where the CAD software acquires first elements that are affected by the acquired second elements, such as first elements that are within a predetermined range from the second element or first elements that have been changed in the past in relation to the second element.

[0055] Next, the process proceeds to step S154, where one or more changed dimensions are estimated from the acquired first element and second element using the trained model in the storage device 42. If multiple first elements or multiple second elements are acquired, the changed dimensions are estimated for each first element or each second element. The estimation is performed by inputting the first element and the second element into the trained model and acquiring the changed dimensions output from the trained model.

[0056] Next, the process proceeds to step S156, where the estimated post-change dimensions are displayed on the display device 44 as change candidates, and then the process proceeds to step S158.

[0057] In step S158, it is determined whether any of the displayed changed dimensions has been selected, and if it is determined that a changed dimension has been selected (YES), the process proceeds to step S160, where the dimension of the first element obtained in step S152 is changed to the changed dimension selected in step S158, and the series of processes is terminated.

[0058] On the other hand, if it is determined in step S158 that the changed dimensions have not been selected (NO), the process waits in step S158 until the changed dimensions are selected.

[0059] [When changing the dimensions of the PS] Next, the operation when changing the dimensions of the PS will be described. Fig. 7 is a plan view of the PS before the dimensions are changed.

[0060] Fig. 8 shows a screen for inputting the second element, and Fig. 9 shows a screen for displaying modification candidates for the first element.

[0061] FIG. 10 shows a plan view after changing the dimensions of PS. When a designer selects the edit element "PS" while editing the plan view of FIG. 7 in CAD software, the screen shown in FIG. 8 appears. To change the dimension of the edit element "PS" to 270 mm, the designer enters "PS with a width of 270 mm" on the screen shown in FIG. 8. Then, by clicking the "Proposal" button, steps S150 to S154 are performed to obtain "PS with a width of 270 mm" as the second element, and "left closet with a width of 700 mm" and "center closet with a width of 700 mm" as the first elements. One or more post-change dimensions are estimated from the obtained first and second elements using the trained model. The estimated post-change dimensions are displayed as change candidates for the first element, as shown in FIG. 9, via step S156. In the example of FIG. 9, three change candidates are displayed. The first change candidate is to change the width of the left closet from 700 [mm] to 650 [mm] and the width of the center closet from 700 [mm] to 650 [mm]. The second change candidate is to change the width of the left closet from 700 [mm] to 600 [mm]. The third change candidate is to change the width of the center closet from 700 [mm] to 600 [mm]. Here, if the designer selects the first change candidate, for example, the width of the left closet is changed from 700 [mm] to 650 [mm] and the width of the center closet is changed from 700 [mm] to 650 [mm], as shown in FIG. 10 , through step S160.

[0062] [When a Piano is to be Carried In] Next, an operation when a piano is to be carried in will be described.

[0063] Fig. 11 is a plan view of the piano before it is brought in. Fig. 12 is a screen for inputting the second element.

[0064] Fig. 13 is a screen displaying modification candidates for the first element. Fig. 14 is a plan view after the piano is brought in.

[0065] When a designer selects a hallway while editing the plan view of FIG. 11 in CAD software, the screen shown in FIG. 12 appears. To consider the possibility of bringing in a piano, the designer enters "piano with a width of 120 mm or more" in the screen shown in FIG. 12. Then, by clicking the "Proposal" button, steps S150 to S154 are performed, where "piano with a width of 120 mm or more" is acquired as the second element and "toilet with a width of 920 mm" is acquired as the first element. Using the trained model, one or more post-change dimensions are estimated from the acquired first and second elements. The estimated post-change dimensions are displayed as change candidates for the first element, as shown in FIG. 13, in step S156. In the example shown in FIG. 13, three change candidates are displayed. The first change candidate is to change the width of the toilet from 920 mm to 870 mm so that the 120-mm piano does not interfere with the toilet wall when being brought in. The second change candidate is to change the width of the toilet from 920 mm to 855 mm so that a 130 cm wide piano will not interfere with the toilet wall when being brought in. The third change candidate is to change the width of the toilet from 920 mm to 840 mm so that a 140 cm wide piano will not interfere with the toilet wall when being brought in. Here, if the designer selects the first change candidate, for example, the width of the toilet will be changed from 920 mm to 870 mm through step S160, as shown in FIG.

[0066] Next, the effects of this embodiment will be described. In this embodiment, a first element, which is an editable editing element in design information and whose change requires human judgment, and a second element that is affected by or affects the change to the first element are acquired, and one or more changed dimensions are estimated from the acquired first element and second element using a trained model that has been trained based on element information 400 regarding the first element, element information 402 regarding the second element, and dimension information 404 regarding one or more changed dimensions when the first element is changed.

[0067] This allows the designer to simply select from among the change candidates, making it easier to change edit elements that require human judgment to change.

[0068] Second Embodiment Next, a second embodiment of the present invention will be described. Figures 15 to 20 are diagrams illustrating this embodiment. Figures 7, 8, 10, 11, 12, and 14 are also used.

[0069] This embodiment differs from the first embodiment in that the influence of changing the first element is estimated.

[0070] First, the configuration of this embodiment will be described. Fig. 15 is a diagram showing the structure of the influence data.

[0071] The storage device 42 stores impact data, as shown in FIG. 15 . The impact data includes element information 406 regarding a first element, which is an editable edit element in the design information and requires human judgment to change; element information 408 regarding a second element that is affected by or impacts the change to the first element; and impact information 410 regarding the impact of changing the first element. Whether an edit element requires human judgment to change is determined as in the first embodiment. An edit element determined to require human judgment to change based on the judgment necessity data in the storage device 42 is treated as a first element in the impact data. The second element includes an editable edit element in the design information or other elements (e.g., furniture to be installed in a room or other virtual elements). The impact data can be configured, for example, as data corresponding to each edit history in the edit history data. When a designer inputs the impact of changing the first element, this can be recorded as impact data.

[0072] 15, the first element is registered as "700mm wide left closet" and "700mm wide center closet"; the second element is registered as "270mm wide PS"; and the effect is registered as "at least one of the left closet and center closet needs to be changed." This indicates that the 700mm wide left closet and the 700mm wide center closet are elements that require human judgment to change, the 270mm wide PS is an element that affects the change of the first element, and if the PS dimension is changed to 270mm, at least one of the left closet and center closet needs to be changed.

[0073] In the second line of Figure 15, the first element is "left closet with a width of 700mm" and "center closet with a width of 700mm", the second element is "PS with a width of 270mm", and the effect is "both the left closet and the center closet can be changed". This indicates that if the PS dimension is changed to 270mm, both the left closet and the center closet can be changed.

[0074] The third line in Figure 15 has "700mm wide left closet" as the first element, "270mm wide PS" as the second element, and "If only the left closet is changed, the storage capacity and convenience of the left closet will decrease by XX%" as the impact. This indicates that if only the left closet is changed when the PS dimension is changed to 270mm, the storage capacity and convenience of the left closet will decrease by XX%.

[0075] The fourth line in Figure 15 has "700mm wide central closet" as the first element, "270mm wide PS" as the second element, and "If only the central closet is changed, the storage capacity and convenience of the central closet will decrease by △△%" as the impact. This indicates that if only the central closet is changed when the PS dimension is changed to 270mm, the storage capacity and convenience of the central closet will decrease by △△%.

[0076] The fifth line in Figure 15 has the first element "700mm wide left closet" and "700mm wide center closet," the second element "270mm wide PS," and the impact "When both the left closet and center closet are changed, the storage capacity and convenience will decrease by XX%." This indicates that when both the left closet and center closet are changed to change the PS dimension to 270mm, the storage capacity and convenience will decrease by XX%.

[0077] In the sixth line of Figure 15, the first element is "a toilet with a width of 920 mm," the second element is "a piano with a width of 120 mm or more," and the impact is "the toilet width needs to be changed." This indicates that the width of the toilet needs to be changed to prevent interference between the piano and the toilet wall when a piano with a width of 120 mm or more is brought in.

[0078] In the seventh line of Figure 15, the first element is "a toilet with a width of 920 mm," the second element is "a piano with a width of 120 mm or more," and the effect is "there are no other elements that can be changed other than the width of the toilet." This indicates that there are no other elements that can be changed other than the width of the toilet to prevent interference between the piano and the toilet wall when bringing in a piano with a width of 120 mm or more.

[0079] In the eighth line of Figure 15, the first element is "a toilet with a width of 920mm," the second element is "a piano with a width of 120mm," and the impact is "If the width of the toilet is changed to 870mm, the convenience of the toilet will decrease by XX%." This indicates that if the width of the toilet is changed to 870mm to prevent interference between the piano and the toilet wall when bringing in a piano with a width of 120mm, the convenience of the toilet will decrease by XX%.

[0080] In the ninth line of Figure 15, the first element is "a toilet with a width of 920mm," the second element is "a piano with a width of 130mm," and the impact is "If the width of the toilet is changed to 855mm, the convenience of the toilet will decrease by △△%." This indicates that if the width of the toilet is changed to 855mm to prevent interference between the piano and the toilet wall when bringing in a piano with a width of 130mm, the convenience of the toilet will decrease by △△%.

[0081] In the tenth line of Figure 15, the first element is "a toilet with a width of 920mm," the second element is "a piano with a width of 140mm," and the impact is "If the width of the toilet is changed to 840mm, the convenience of the toilet will decrease by XX%." This indicates that if the width of the toilet is changed to 840mm to prevent interference between the piano and the toilet wall when bringing in a piano with a width of 140mm, the convenience of the toilet will decrease by XX%.

[0082] The storage device 42 stores data of the second trained model. The second trained model is trained based on element information 406 related to the first element, element information 408 related to the second element, and impact information 410 related to one or more impacts when the first element is changed. The method for generating the second trained model is the same as that of the first embodiment (FIGS. 4 and 5). Since the impact data is used for AI training, the storage device 42 stores a large amount of impact data created in the past. This impact data serves as training data for AI training.

[0083] Next, the operation of this embodiment will be described with reference to a flowchart of FIG.

[0084] The influence estimation process is a process that is executed in response to a request from a user. When executed by the CPU 30, as shown in FIG. 16, first, the same processes as steps S150 to S156 in the first embodiment are performed, and then the process proceeds to step S162.

[0085] In step S162, it is determined whether or not any of the effect buttons displayed for each post-change dimension has been clicked, and if it is determined that an effect button has been clicked (YES), the process proceeds to step S164.

[0086] In step S164, one or more influences are estimated from the first element and the second element acquired in steps S150 and S152 using the second trained model in the storage device 42. If multiple first elements or multiple second elements are acquired, the influence is estimated for each first element or each second element. The estimation is performed by inputting the first element and the second element into the second trained model and acquiring the influence output from the second trained model.

[0087] Next, the process proceeds to step S166, where the estimated influence is displayed on the display device 44, and the same processes as steps S158 and S160 in the first embodiment are carried out, thereby completing the series of processes.

[0088] On the other hand, if it is determined in step S158 that the changed dimensions have not been selected (NO), the process proceeds to step S162.

[0089] On the other hand, if it is determined in step S162 that the influence button has not been clicked (NO), the process proceeds to step S158.

[0090] [When Changing the Dimensions of PS] Next, the operation when changing the dimensions of PS will be described. Fig. 17 shows a screen displaying change candidates for the first element.

[0091] FIG. 18 is a screen displaying the impact of the first change candidate. When a designer selects the edit element "PS" while editing the plan view of FIG. 7 in CAD software, the screen of FIG. 8 is displayed. To change the dimension of the edit element "PS" to 270 mm, the designer enters "PS with a width of 270 mm" on the screen of FIG. 8. Then, by clicking the "Proposal" button, steps S150 to S154 are performed to obtain "PS with a width of 270 mm" as the second element, and "left closet with a width of 700 mm" and "center closet with a width of 700 mm" as the first elements. One or more post-change dimensions are estimated from the obtained first and second elements using the trained model. The estimated post-change dimensions are displayed as change candidates for the first element, as shown in FIG. 17, after step S156. In the example of FIG. 17, three change candidates are displayed. The contents of these change candidates are the same as those in FIG. 9. Each proposed change also has an impact button displayed for displaying the impact corresponding to that proposed change.

[0092] Here, if the designer clicks, for example, an impact button corresponding to the first change candidate, one or more impacts are estimated from the acquired first and second elements using the second trained model in step S164. The estimated impacts are displayed in step S166 as the impacts of adopting the first change candidate, as shown in FIG. 18 . In the example of FIG. 18 , three impacts are displayed. The first impact indicates that at least one of the left closet and the center closet needs to be changed. The parentheses indicate that the number of edit elements that need to be changed is one. The second impact indicates that both the left closet and the center closet can be changed. The parentheses indicate that the number of edit elements that can be changed is two. The third impact indicates that if both the left closet and the center closet are changed, storage capacity and convenience will decrease by XX%.

[0093] Clicking on the impact button corresponding to the second proposed change similarly displays the impact of adopting the second proposed change, and clicking on the impact button corresponding to the third proposed change similarly displays the impact of adopting the third proposed change.

[0094] After checking the impact, the designer selects, for example, the first change candidate. After step S160, the width of the left closet is changed from 700 mm to 650 mm, and the width of the center closet is changed from 700 mm to 650 mm, as shown in FIG. 10.

[0095] [When a Piano is to be Carried In] Next, an operation when a piano is to be carried in will be described.

[0096] Figure 19 is a screen displaying a modification candidate for the first element, and Figure 20 is a screen displaying the impact corresponding to the first modification candidate.

[0097] When a designer selects a hallway while editing the plan view of Figure 11 in CAD software, the screen shown in Figure 12 appears. To consider the delivery of a piano, the designer enters "piano with a width of 120 or more" in the screen shown in Figure 12. Then, by clicking the "Proposal" button, steps S150 to S154 are performed, where "piano with a width of 120 or more" is acquired as the second element and "toilet with a width of 920" is acquired as the first element. Using the trained model, one or more post-change dimensions are estimated from the acquired first and second elements. The estimated post-change dimensions are displayed as change candidates for the first element, as shown in Figure 19, in step S156. In the example shown in Figure 19, three change candidates are displayed. The contents of these change candidates are the same as those shown in Figure 13. Each change candidate also has an impact button that displays the impact of the change candidate.

[0098] Here, if the designer clicks, for example, an impact button corresponding to the first change candidate, one or more impacts are estimated from the acquired first and second elements using the second trained model in step S164. The estimated impacts are displayed in step S166 as the impacts of adopting the first change candidate, as shown in FIG. 20. In the example of FIG. 20, three impacts are displayed. The first impact indicates that the width of the toilet needs to be changed. The parentheses indicate that there is only one edit element that needs to be changed. The second impact indicates that there are no other edit elements that can be changed other than the width of the toilet. The parentheses indicate that there is only one edit element that can be changed. The third impact indicates that if the width of the toilet is changed to 870, the convenience of the toilet will decrease by XX%.

[0099] Clicking on the impact button corresponding to the second proposed change similarly displays the impact of adopting the second proposed change, and clicking on the impact button corresponding to the third proposed change similarly displays the impact of adopting the third proposed change.

[0100] After checking the effect, the designer selects, for example, the first change candidate, and then, through step S160, the width of the toilet is changed from 920 [mm] to 870 [mm] as shown in FIG.

[0101] Next, the effects of this embodiment will be described. In this embodiment, a first element, which is an editable edit element in design information and whose change requires human judgment, and second element information about a second element that is affected or affects the change to the first element are acquired, and one or more effects are estimated from the acquired first element and second element using a second trained model that has been trained based on element information 406 about the first element, element information 408 about the second element, and impact information 410 about one or more impacts when the first element is changed.

[0102] This allows the designer to check the impact of changing the first element, and therefore allows the designer to take the impact into consideration when changing an edit element that requires human judgment to change.

[0103] [Third Embodiment] Next, a third embodiment of the present invention will be described. Figures 21 to 26 are diagrams illustrating this embodiment. Figures 7, 8, 10, 11, 12, and 14 are also used.

[0104] This embodiment differs from the first embodiment in that it estimates the matters to be considered when changing the first element.

[0105] First, the configuration of this embodiment will be described. Fig. 21 is a diagram showing the structure of consideration data.

[0106] As shown in FIG. 21 , the storage device 42 stores consideration data. The consideration data includes element information 412 related to a first element, which is an editable edit element in the design information and requires human judgment to change; element information 414 related to a second element that is affected by or affects the change to the first element; and consideration data 416 related to one or more factors to be considered when changing the first element. Whether an edit element requires human judgment to change is determined as in the first embodiment. An edit element determined to require human judgment to change based on the judgment necessity data in the storage device 42 is treated as a first element in the consideration data. The second element includes an editable edit element in the design information or other elements (e.g., furniture to be installed in a room or other virtual elements). The consideration data can be configured, for example, as data corresponding to each edit history in the edit history data. When a designer inputs factors to be considered when changing the first element, the factors can be recorded as consideration data.

[0107] 21, the first line has "700mm wide left closet" and "700mm wide center closet" as the first element, "270mm wide PS" as the second element, and "When changing both the left closet and center closet, consider the minimum space, size of stored items, balance of stored items, closet purpose, convenience, and room layout" as the consideration. This indicates that the left closet with a width of 700 [mm] and the center closet with a width of 700 [mm] are elements that require human judgment for their change, the PS with a width of 270 [mm] is an element that affects the change of the first element, and when changing both the left closet and center closet when changing the PS dimension to 270 [mm], the minimum space, size of stored items, balance of stored items, closet purpose, convenience, and room layout should be considered.

[0108] The second line in Figure 21 has "left closet with a width of 700mm" and "center closet with a width of 700mm" as the first element, "PS with a width of 270mm" as the second element, and "When changing either the left closet or the center closet, only one of them will be extremely small compared to when both are changed, so consider the above points even more carefully" as a consideration. This indicates that when changing either the left closet or the center closet to change the PS dimension to 270mm, only one of them will be extremely small compared to when both are changed, so consider the above points even more carefully.

[0109] In the third line of Figure 21, the first element is "a toilet with a width of 920 mm," the second element is "a piano with a width of 120 mm," and the consideration is "if the width of the toilet is changed to 870 mm, consider whether barrier-free specifications can be added." This indicates that if the width of the toilet is changed to 870 mm to prevent interference between the piano and the toilet wall when bringing in a 120 mm wide piano, it is necessary to consider whether barrier-free specifications can be added.

[0110] In the fourth line of Figure 21, the first element is "a toilet with a width of 920 mm," the second element is "a piano with a width of 130 mm," and the consideration is "if the width of the toilet is changed to 855 mm, consider whether a hand-washing counter can be added." This indicates that if the width of the toilet is changed to 855 mm to prevent interference between the piano and the toilet wall when a 130 mm wide piano is brought in, consideration should be given to whether a hand-washing counter can be added.

[0111] In the fifth line of Figure 21, the first element is "a toilet with a width of 920mm," the second element is "a piano with a width of 140mm," and the consideration is "consider whether the minimum space can be secured when changing the width of the toilet to 840mm." This indicates that when changing the width of the toilet to 840mm to prevent interference between the piano and the toilet wall when bringing in a 140mm wide piano, it is necessary to consider whether the minimum space can be secured.

[0112] The storage device 42 stores data of the second trained model. The second trained model is trained based on element information 412 related to the first element, element information 414 related to the second element, and consideration information 416 related to one or more considerations when changing the first element. The method for generating the second trained model is the same as that of the first embodiment (FIGS. 4 and 5). Since the consideration data is used for AI training, the storage device 42 stores a large amount of consideration data created in the past. This consideration data serves as training data for AI training.

[0113] Next, the operation of this embodiment will be described with reference to a flowchart of FIG.

[0114] The consideration item estimation process is a process that is executed in response to a request from a user, and when executed by the CPU 30, as shown in FIG. 22, first, the same processes as steps S150 to S156 in the first embodiment are performed, and then the process proceeds to step S168.

[0115] In step S168, it is determined whether any of the consideration buttons displayed for each post-change dimension has been clicked, and if it is determined that a consideration button has been clicked (YES), the process proceeds to step S170.

[0116] In step S170, one or more considerations are estimated from the first element and the second element acquired in steps S150 and S152 using the second trained model in the storage device 42. If multiple first elements or multiple second elements are acquired, considerations are estimated for each first element or each second element. The estimation is performed by inputting the first element and the second element into the second trained model and acquiring considerations output from the second trained model.

[0117] Next, the process proceeds to step S172, where the estimated consideration items are displayed on the display device 44, and the same processes as steps S158 and S160 in the first embodiment are performed, thereby completing the series of processes.

[0118] On the other hand, if it is determined in step S158 that the changed dimensions have not been selected (NO), the process proceeds to step S168.

[0119] On the other hand, if it is determined in step S168 that the consideration button has not been clicked (NO), the process proceeds to step S158.

[0120] [When Changing the Dimensions of PS] Next, the operation when changing the dimensions of PS will be described. Fig. 23 shows a screen displaying change candidates for the first element.

[0121] FIG. 24 is a screen displaying considerations corresponding to the first change candidate. When a designer selects the edit element "PS" while editing the plan view of FIG. 7 in CAD software, the screen of FIG. 8 is displayed. To change the dimension of the edit element "PS" to 270 mm, the designer enters "PS with a width of 270 mm" on the screen of FIG. 8. Then, by clicking the "Proposal" button, steps S150 to S154 are performed to obtain "PS with a width of 270 mm" as the second element, and "left closet with a width of 700 mm" and "center closet with a width of 700 mm" as the first elements. One or more post-change dimensions are estimated from the obtained first and second elements using the trained model. The estimated post-change dimensions are displayed as change candidates for the first element, as shown in FIG. 23, after step S156. In the example of FIG. 23, three change candidates are displayed. The contents of these change candidates are the same as those in FIG. 9. Each change candidate is also displayed with a consideration button for displaying the considerations corresponding to that change candidate.

[0122] Here, if the designer clicks, for example, a consideration button corresponding to the first change candidate, one or more considerations are estimated from the acquired first and second elements using the second trained model in step S170. The estimated considerations are displayed in step S172 as considerations to be taken into account when adopting the first change candidate, as shown in Fig. 24. In the example of Fig. 24, it is displayed that when both the left closet and the center closet are changed, the minimum space, the size of the stored items, the balance of the stored items, the purpose of the closet, convenience, and the room layout should be taken into account.

[0123] Clicking the considerations button corresponding to the second proposed change similarly displays the considerations that should be considered if the second proposed change is adopted. Clicking the considerations button corresponding to the third proposed change similarly displays the considerations that should be considered if the third proposed change is adopted.

[0124] After checking the considerations, the designer selects, for example, the first change candidate. After step S160, the width of the left closet is changed from 700 mm to 650 mm, and the width of the center closet is changed from 700 mm to 650 mm, as shown in FIG. 10.

[0125] [When a Piano is to be Carried In] Next, an operation when a piano is to be carried in will be described.

[0126] Fig. 25 is a screen displaying modification candidates for the first element, and Fig. 26 is a screen displaying considerations corresponding to the first modification candidate.

[0127] When a designer selects a hallway while editing the plan view of Figure 11 in CAD software, the screen shown in Figure 12 appears. To consider the delivery of a piano, the designer enters "piano with a width of 120 or more" in the screen shown in Figure 12. Then, by clicking the "Proposal" button, steps S150 to S154 are performed, where "piano with a width of 120 or more" is acquired as the second element and "toilet with a width of 920" is acquired as the first element. Using the trained model, one or more post-change dimensions are estimated from the acquired first and second elements. The estimated post-change dimensions are displayed as change candidates for the first element, as shown in Figure 25, in step S156. In the example shown in Figure 25, three change candidates are displayed. The contents of these change candidates are the same as those shown in Figure 13. Each change candidate also includes a consideration button for displaying the considerations associated with that change candidate.

[0128] Here, when the designer clicks, for example, the consideration button corresponding to the first change candidate, one or more considerations are estimated from the acquired first and second elements using the second trained model in step S170. The estimated considerations are displayed in step S172 as considerations to be taken into account when adopting the first change candidate, as shown in Figure 26. In the example of Figure 26, it is displayed that when the width of the toilet is changed to 870 mm, consideration should be given to whether barrier-free specifications can be added.

[0129] Clicking the considerations button corresponding to the second proposed change similarly displays the considerations that should be considered if the second proposed change is adopted. Clicking the considerations button corresponding to the third proposed change similarly displays the considerations that should be considered if the third proposed change is adopted.

[0130] After checking the considerations, the designer selects, for example, the first change candidate, and then, through step S160, the width of the toilet is changed from 920 [mm] to 870 [mm] as shown in FIG.

[0131] Next, the effects of this embodiment will be described. In this embodiment, second element information about a first element that is an editable edit element in design information and whose change requires human judgment and a second element that is affected by or affects the change to the first element is acquired, and one or more considerations are estimated from the acquired first element and second element using a second trained model that has been trained based on element information 412 about the first element, element information 414 about the second element, and consideration information 416 about one or more considerations when changing the first element.

[0132] This allows the designer to check the matters that should be taken into consideration when changing the first element, and therefore allows the designer to take into consideration important points to be noted when changing an edit element that requires human judgment to change.

[0133] [Fourth embodiment] Next, a fourth embodiment of the present invention will be described. Figures 27 to 32 are diagrams illustrating this embodiment. Figures 7, 8, 10, 11, 12, and 14 are also used.

[0134] This embodiment differs from the first embodiment in that the risk of changing the first element is estimated.

[0135] First, the configuration of this embodiment will be described. Fig. 27 is a diagram showing the structure of risk data.

[0136] The storage device 42 stores risk data, as shown in FIG. 27 . The risk data includes element information 418 regarding a first element, which is an editable edit element in the design information and requires human judgment to change; element information 420 regarding a second element that is affected by or affects the change to the first element; and risk information 422 regarding the risk of changing the first element. Whether an edit element requires human judgment to change is determined as in the first embodiment. An edit element determined to require human judgment to change based on the judgment necessity data in the storage device 42 is treated as a first element in the risk data. The second element includes an editable edit element in the design information or other elements (e.g., furniture or other virtual elements to be installed in a room). The risk data can be configured, for example, as data corresponding to each edit history in the edit history data. When a designer inputs the risk of changing the first element, this can be recorded as risk data.

[0137] In the first line of Figure 27, the first element is "700mm wide left closet" and "700mm wide center closet," the second element is "270mm wide PS," and the risks are "If both the left closet and center closet are changed, there is a legal risk of violating the Building Standards Act and Housing Quality Assurance Act, a disaster prevention risk that will hinder evacuation in an emergency, a risk of reduced convenience due to difficulty in taking items in and out, and a risk of a decrease in asset value that will lower the value of the property." This indicates that the left closet, which is 700 mm wide, and the center closet, which is also 700 mm wide, are elements that require human judgment to change, and the PS, which is 270 mm wide, is an element that affects changes to the first element, and if both the left closet and the center closet are changed to change the PS dimension to 270 mm, there are anticipated legal risks of violating the Building Standards Act and the Housing Quality Assurance Act, disaster prevention risks that will hinder emergency evacuation, risks of reduced convenience due to difficulty in taking stored items in and out, and risks of a decrease in asset value that will lower the value of the property.

[0138] The second line in Figure 27 has "left closet with a width of 700mm" and "center closet with a width of 700mm" registered as the first element, "PS with a width of 270mm" registered as the second element, and "When changing either the left closet or the center closet, the risk is even greater than when both are changed because only one of them becomes extremely small." This indicates that when changing the PS dimension to 270mm and either the left closet or the center closet is changed, the risk is even greater than when both are changed because only one of them becomes extremely small.

[0139] The third line in Figure 27 has "a toilet with a width of 920mm" as the first element, "a piano with a width of 120mm" as the second element, and "if the width of the toilet is changed to 870mm, there is a risk that convenience will be reduced, such as restricted movement during use and difficulty in cleaning" registered as a risk. This indicates that if the width of the toilet is changed to 870mm to prevent interference between the piano and the toilet wall when bringing in a 120mm wide piano, there is a risk that convenience will be reduced, such as restricted movement during use and difficulty in cleaning.

[0140] The fourth line in Figure 27 has "a toilet with a width of 920mm" as the first element, "a piano with a width of 130mm" as the second element, and "if the width of the toilet is changed to 855mm, there is a safety risk that if a piano with a width of 130mm is brought in and the piano does not interfere with the toilet wall, there is a safety risk that if the door opens inwards, it will not open."

[0141] In the fifth line of Figure 27, the first element is "a toilet with a width of 920mm," the second element is "a piano with a width of 140mm," and the risk is "If the width of the toilet is changed to 840mm, there is a legal risk of violating the Building Standards Act, the Housing Quality Assurance Act, etc." This indicates that if the width of the toilet is changed to 840mm to prevent interference between the piano and the toilet wall when bringing in a 140mm wide piano, there is a legal risk of violating the Building Standards Act, the Housing Quality Assurance Act, etc.

[0142] The storage device 42 stores data of the second trained model. The second trained model is trained based on element information 418 related to the first element, element information 420 related to the second element, and risk information 422 related to one or more risks when the first element is changed. The method for generating the second trained model is the same as that of the first embodiment (FIGS. 4 and 5). Since the risk data is used for AI training, the storage device 42 stores a large amount of risk data created in the past. This risk data serves as training data for AI training.

[0143] Next, the operation of this embodiment will be described with reference to the flowchart of FIG.

[0144] The risk estimation process is a process that is executed in response to a request from a user, and when executed by the CPU 30, as shown in FIG. 28, first, processes similar to steps S150 to S156 in the first embodiment described above are performed, and then the process proceeds to step S174.

[0145] In step S174, it is determined whether or not any of the risk buttons displayed for each post-change dimension has been clicked, and if it is determined that a risk button has been clicked (YES), the process proceeds to step S176.

[0146] In step S176, one or more risks are estimated from the first element and the second element acquired in steps S150 and S152 using the second trained model in the storage device 42. If multiple first elements or multiple second elements are acquired, a risk is estimated for each first element or each second element. The estimation is performed by inputting the first element and the second element into the second trained model and acquiring the risk output from the second trained model.

[0147] Next, the process proceeds to step S178, where the estimated risk is displayed on the display device 44, and the same processes as steps S158 and S160 in the first embodiment are carried out, thereby completing the series of processes.

[0148] On the other hand, if it is determined in step S158 that the changed dimensions have not been selected (NO), the process proceeds to step S174.

[0149] On the other hand, if it is determined in step S174 that the risk button has not been clicked (NO), the process proceeds to step S158.

[0150] [When Changing the Dimensions of PS] Next, the operation when changing the dimensions of PS will be described. Fig. 29 shows a screen displaying change candidates for the first element.

[0151] FIG. 30 shows a screen displaying the risks associated with the first change candidate. When a designer selects the edit element "PS" while editing the plan view of FIG. 7 in CAD software, the screen shown in FIG. 8 appears. To change the dimension of the edit element "PS" to 270 mm, the designer enters "PS with a width of 270 mm" in the screen shown in FIG. 8. Then, by clicking the "Proposal" button, steps S150 to S154 are performed to obtain "PS with a width of 270 mm" as the second element, and "left closet with a width of 700 mm" and "center closet with a width of 700 mm" as the first elements. One or more post-change dimensions are estimated from the obtained first and second elements using the trained model. The estimated post-change dimensions are displayed as change candidates for the first element, as shown in FIG. 29, via step S156. In the example of FIG. 29, three change candidates are displayed. The contents of these change candidates are the same as those in FIG. 9. In addition, a risk button for displaying the risk associated with each change candidate is displayed for each change candidate.

[0152] Here, if the designer clicks, for example, a risk button corresponding to the first change candidate, one or more risks are estimated from the acquired first and second elements using the second trained model in step S176. The estimated risks are displayed as risks when the first change candidate is adopted in step S178, as shown in Figure 30. The example in Figure 30 displays that, when both the left closet and the center closet are changed, there are expected legal risks such as violations of the Building Standards Act and the Housing Quality Assurance Act, disaster prevention risks such as interference with emergency evacuation, risks of reduced convenience due to difficulty in accessing stored items, and risks of a decrease in asset value due to a decrease in the value of the property.

[0153] Clicking on the risk button corresponding to the second proposed change similarly displays the risks associated with adopting the second proposed change, and clicking on the risk button corresponding to the third proposed change similarly displays the risks associated with adopting the third proposed change.

[0154] After identifying the risks, the designer selects, for example, the first change candidate. In step S160, the width of the left closet is changed from 700 mm to 650 mm, and the width of the center closet is changed from 700 mm to 650 mm, as shown in FIG. 10.

[0155] [When a Piano is to be Carried In] Next, an operation when a piano is to be carried in will be described.

[0156] Fig. 31 is a screen displaying modification candidates for the first element, and Fig. 32 is a screen displaying risks associated with the first modification candidates.

[0157] When a designer selects a hallway while editing the plan view of Figure 11 in CAD software, the screen shown in Figure 12 appears. To consider the delivery of a piano, the designer enters "piano with a width of 120 or more" in the screen shown in Figure 12. Then, by clicking the "Proposal" button, steps S150 to S154 are performed, where "piano with a width of 120 or more" is acquired as the second element and "toilet with a width of 920" is acquired as the first element. Using the trained model, one or more post-change dimensions are estimated from the acquired first and second elements. The estimated post-change dimensions are displayed as change candidates for the first element, as shown in Figure 31, in step S156. In the example shown in Figure 31, three change candidates are displayed. The contents of these change candidates are the same as those shown in Figure 13. Each change candidate also displays a risk button to display the risk associated with that change candidate.

[0158] Here, if the designer clicks, for example, a risk button corresponding to the first change candidate, one or more risks are estimated from the acquired first and second elements using the second trained model in step S176. The estimated risks are displayed as risks when the first change candidate is adopted in step S178, as shown in Figure 32. In the example of Figure 32, it is displayed that changing the width of the toilet to 870 mm is likely to result in reduced convenience, such as restricted movement during use and difficulty in cleaning.

[0159] Clicking on the risk button corresponding to the second proposed change similarly displays the risks associated with adopting the second proposed change, and clicking on the risk button corresponding to the third proposed change similarly displays the risks associated with adopting the third proposed change.

[0160] After identifying the risks, the designer selects, for example, the first change candidate, and then, through step S160, the width of the toilet is changed from 920 [mm] to 870 [mm], as shown in FIG.

[0161] Next, the effects of this embodiment will be described. In this embodiment, second element information about a first element that is an editable edit element in design information and whose change requires human judgment and a second element that is affected by or affects the change to the first element is acquired, and one or more risks are estimated from the acquired first element and second element using a second trained model that has been trained based on element information 418 about the first element, element information 420 about the second element, and risk information 422 about one or more risks when the first element is changed.

[0162] This allows the designer to confirm the risks involved in changing the first element, and therefore allows the designer to take risks into consideration when changing an edit element that requires human judgment to change.

[0163] [Modification] In the first embodiment and its modification, a trained model trained based on element information 400 regarding the first element, element information 402 regarding the second element, and dimension information 404 regarding one or more changed dimensions when the first element is changed is employed. However, this is not limited to this. A trained model trained based on element information 400 regarding the first element and dimension information 404 regarding one or more changed dimensions when the first element is changed can also be employed. In this case, the processing of step S150 in FIG. 6 is unnecessary. Furthermore, in step S154 in FIG. 6, one or more changed dimensions are estimated from the acquired first element using the trained model in the storage device 42. If multiple first elements are acquired, the changed dimensions are estimated for each first element. The estimation is performed by inputting the first element to the trained model and acquiring the changed dimensions output from the trained model.

[0164] Furthermore, in the second embodiment and its modified example, a trained model trained based on element information 406 related to the first element, element information 408 related to the second element, and impact information 410 related to one or more impacts when the first element is changed is employed. However, this is not limited to this, and a trained model trained based on element information 406 related to the first element and impact information 410 related to one or more impacts when the first element is changed can also be employed. In this case, the processing of step S150 in FIG. 16 is unnecessary. Furthermore, in step S164 in FIG. 16, one or more impacts are estimated from the acquired first element using the trained model in the storage device 42. If multiple first elements are acquired, an impact is estimated for each first element. The estimation is performed by inputting the first element into the trained model and acquiring the impact output from the trained model.

[0165] Furthermore, in the third embodiment and its modified examples, a trained model trained based on the element information 412 related to the first element, the element information 414 related to the second element, and the consideration information 416 related to one or more considerations when changing the first element is employed. However, this is not limited to this, and a trained model trained based on the element information 412 related to the first element and the consideration information 416 related to one or more considerations when changing the first element can also be employed. In this case, the processing of step S150 in FIG. 22 is unnecessary. Furthermore, in step S170 in FIG. 22, one or more considerations are estimated from the acquired first element using the trained model in the storage device 42. If multiple first elements are acquired, a consideration is estimated for each first element. The estimation is performed by inputting the first element into the trained model and acquiring the considerations output from the trained model.

[0166] Furthermore, in the fourth embodiment and its modified examples, a trained model trained based on element information 418 related to the first element, element information 420 related to the second element, and risk information 422 related to one or more risks when changing the first element is employed. However, this is not limited to this, and a trained model trained based on element information 418 related to the first element and risk information 422 related to one or more risks when changing the first element can also be employed. In this case, the processing of step S150 in FIG. 28 is unnecessary. Furthermore, in step S176 in FIG. 28, one or more risks are estimated from the acquired first element using the trained model in the storage device 42. If multiple first elements are acquired, a risk is estimated for each first element. The estimation is performed by inputting the first element into the trained model and obtaining the risk output from the trained model.

[0167] In addition, in the first to fourth embodiments and their modifications, a trained model is used, but this is not limiting, and a configuration using a large language model can also be adopted. Specifically, for example, the following configuration can be adopted.

[0168] The drawing creation support device 100 is connected via a network to a large-scale language model server having a large-scale language model. The large-scale language model server has a hardware configuration similar to that of the drawing creation support device 100. The large-scale language model is a deep learning model that pre-trains a language model, which models human spoken language based on its occurrence probability, from a massive amount of data. Upon receiving a request, the large-scale language model server uses the large-scale language model to statistically estimate the probability of generating the next word from the sentence included in the received request and transmits the estimation result to the requestor. For example, known technologies described in the internet sites "https: / / chatgpt-lab.com / n / n418d3aa56f0b" and "https: / / agirobots.com / chatgpt-mechanism-and-problem / " can be used as the large-scale language model.

[0169] The following inventions A1 to A8 can be applied to the drawing creation support device 100.

[0170] [Invention A1] The invention comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and whose modification requires human judgment, and second element information relating to a second element that is affected by or affects the modification of the first element, and the request including a request to generate candidate information relating to one or more candidates for modifying the first element; and a candidate information acquisition means for acquiring candidate information output from the large-scale language model in response to the request.

[0171] This allows for obtaining candidate information regarding one or more candidates for changing the first element, making it easier to change an edit element that requires human judgment to change.

[0172] An embodiment of invention A1 will be described as a modification of the first embodiment, in which the process of step S154 is replaced with a process of acquiring dimension information from a large-scale language model server.

[0173] When the designer inputs "PS with a width of 270" on the screen of FIG. 8 and clicks the "Proposal" button, "PS with a width of 270" is acquired as the second element, and "Left closet with a width of 700" and "Center closet with a width of 700" are acquired as the first elements through steps S150 to S154, and a request is sent to the large-scale language model server. The request includes the acquired first element and second element, and also includes a request to generate dimensional information regarding one or more post-change dimensions when the first element is changed in relation to the second element. In response to this request, the large-scale language model server outputs dimensional information. Upon receiving the dimensional information, the drawing creation support device 100 displays, through step S156, modification candidates for the first element based on the received dimensional information.

[0174] In this embodiment, step S154 corresponds to the request input means of the first invention and the candidate information acquisition means of the first invention.

[0175] [Invention A2] The invention comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and whose modification requires human judgment, and second element information relating to a second element that is affected or affects the modification of the first element, the request including a request to generate impact information relating to one or more impacts when the first element is modified; and impact information acquisition means for acquiring impact information output from the large-scale language model in response to the request.

[0176] This provides impact information regarding one or more impacts when changing the first element, allowing the impacts to be taken into consideration when changing an edit element that requires human judgment to change.

[0177] An embodiment of invention A2 will be described as a modification of the second embodiment, in which the process of step S164 is replaced with a process of acquiring influence information from a large-scale language model server.

[0178] When the designer inputs "PS with a width of 270" on the screen of FIG. 8 and clicks the impact button corresponding to the first change candidate, "PS with a width of 270" is acquired as the second element, and "left closet with a width of 700" and "center closet with a width of 700" are acquired as the first elements through steps S150, S152, and S164, and a request is sent to the large-scale language model server. The request includes the acquired first element and second element and a request to generate impact information regarding one or more impacts when the first element is changed in relation to the second element. In response to this request, the large-scale language model server outputs the impact information. Upon receiving the impact information, the drawing creation support device 100 displays the impact of adopting the first change candidate based on the received impact information through step S166.

[0179] In this embodiment, step S164 corresponds to the request input means of invention 2 and the effect information acquisition means of invention 2.

[0180] [Invention A3] The invention comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and whose modification requires human judgment, and second element information relating to a second element that is affected by or affects the modification of the first element, and the request including a request to generate consideration information relating to one or more matters that should be taken into consideration when modifying the first element; and a consideration information acquisition means for acquiring the consideration information output from the large-scale language model in response to the request.

[0181] This provides consideration information regarding one or more items that should be taken into account when changing the first element, allowing consideration of precautions, etc. when changing an edit element that requires human judgment to change.

[0182] An embodiment of Invention A3 will be described as a modification of the third embodiment, in which the process of step S170 is replaced with a process of acquiring consideration matter information from a large-scale language model server.

[0183] When the designer inputs "PS with a width of 270" on the screen of FIG. 8 and clicks the considerations button corresponding to the first change candidate, "PS with a width of 270" is acquired as the second element, and "left closet with a width of 700" and "center closet with a width of 700" are acquired as the first elements through steps S150, S152, and S170, and a request is sent to the large-scale language model server. The request includes the acquired first element and second element and a request to generate considerations information regarding one or more considerations when changing the first element in relation to the second element. In response to this request, the large-scale language model server outputs considerations information. Upon receiving the considerations information, the drawing creation support device 100 displays considerations when adopting the first change candidate based on the received considerations information through step S172.

[0184] In this embodiment, step S170 corresponds to the request input means of invention 3 and the consideration information acquisition means of invention 3.

[0185] [Invention A4] The invention comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and whose modification requires human judgment, and second element information relating to a second element that is affected by or affects the modification of the first element, and the request including a request to generate risk information relating to one or more risks when the first element is modified; and a risk information acquisition means for acquiring the risk information output from the large-scale language model in response to the request.

[0186] This provides risk information regarding one or more risks when changing the first element, allowing risks to be taken into consideration when changing an edit element that requires human judgment to change.

[0187] An embodiment of invention A4 will be described as a modification of the fourth embodiment, in which the process of step S176 is replaced with a process of acquiring risk information from a large-scale language model server.

[0188] When the designer inputs "PS with a width of 270" on the screen of FIG. 8 and clicks the risk button corresponding to the first change candidate, "PS with a width of 270" is acquired as the second element, and "left closet with a width of 700" and "center closet with a width of 700" are acquired as the first elements through steps S150, S152, and S176, and a request is sent to the large-scale language model server. The request includes the acquired first element and second element and a request to generate risk information regarding one or more risks associated with changing the first element in relation to the second element. In response to this request, the large-scale language model server outputs the risk information. Upon receiving the risk information, the drawing creation support device 100 displays the risk of adopting the first change candidate based on the received risk information through step S178.

[0189] In this embodiment, step S176 corresponds to the request input means of invention 4 and the risk information acquisition means of invention 4.

[0190] [Invention A5] The invention comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and a request for generating candidate information relating to one or more candidates for changing the first element, and a candidate information acquisition means for acquiring candidate information output from the large-scale language model in response to the request.

[0191] This allows for obtaining candidate information regarding one or more candidates for changing the first element, making it easier to change an edit element that requires human judgment to change.

[0192] An embodiment of Invention A5 will be described as a modification of the first embodiment, in which the process of step S154 is replaced with a process of acquiring dimension information from a large-scale language model server.

[0193] When the designer inputs "PS with a width of 270" on the screen of FIG. 8 and clicks the "Propose" button, "Left closet with a width of 700" and "Center closet with a width of 700" are acquired as first elements through steps S152 and S154, and a request is sent to the large-scale language model server. The request includes the acquired first element and a request to generate dimensional information regarding one or more post-change dimensions when the first element is changed. In response to this request, the large-scale language model server outputs the dimensional information. When the drawing creation support device 100 receives the dimensional information, it displays change candidates for the first element based on the received dimensional information through step S156.

[0194] In this embodiment, step S154 corresponds to the request input means of invention 5 and the candidate information acquisition means of invention 5.

[0195] [Invention A6] The invention comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and a request for generating impact information relating to one or more impacts when the first element is changed, and an impact information acquisition means for acquiring the impact information output from the large-scale language model in response to the request.

[0196] This provides impact information regarding one or more impacts when changing the first element, allowing the impacts to be taken into consideration when changing an edit element that requires human judgment to change.

[0197] An embodiment of Invention A6 will be described as a modification of the second embodiment, in which the process of step S164 is replaced with a process of acquiring influence information from a large-scale language model server.

[0198] When the designer inputs "PS with a width of 270" on the screen of FIG. 8 and clicks the impact button corresponding to the first change candidate, "left closet with a width of 700" and "center closet with a width of 700" are acquired as the first element through steps S152 and S164, and a request is sent to the large-scale language model server. The request includes the acquired first element and a request to generate impact information regarding one or more impacts when the first element is changed. In response to this request, the large-scale language model server outputs the impact information. Upon receiving the impact information, the drawing creation support device 100 displays the impact of adopting the first change candidate based on the received impact information through step S166.

[0199] In this embodiment, step S164 corresponds to the request input means of invention 6 and the effect information acquisition means of invention 6.

[0200] [Invention A7] The invention comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and a request for generating consideration information relating to one or more matters that should be taken into consideration when changing the first element; and a consideration information acquisition means for acquiring the consideration information output from the large-scale language model in response to the request.

[0201] This provides consideration information regarding one or more items that should be taken into account when changing the first element, allowing consideration of precautions, etc. when changing an edit element that requires human judgment to change.

[0202] An embodiment of Invention A7 will be described as a modification of the third embodiment, in which the process of step S170 is replaced with a process of acquiring consideration matter information from a large-scale language model server.

[0203] When the designer inputs "PS with a width of 270" on the screen of FIG. 8 and clicks the considerations button corresponding to the first change candidate, "left closet with a width of 700" and "center closet with a width of 700" are acquired as the first element through steps S152 and S170, and a request is sent to the large-scale language model server. The request includes the acquired first element and a request to generate considerations information regarding one or more considerations when changing the first element. In response to this request, the large-scale language model server outputs the considerations information. Upon receiving the considerations information, the drawing creation support device 100 displays considerations when adopting the first change candidate based on the received considerations information through step S172.

[0204] In this embodiment, step S170 corresponds to the request input means of the seventh invention and the consideration information acquisition means of the seventh invention.

[0205] [Invention A8] The invention comprises a request input means for inputting a request to a large-scale language model, the request including first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and a request for generating risk information relating to one or more risks when changing the first element, and a risk information acquisition means for acquiring risk information output from the large-scale language model in response to the request.

[0206] This provides risk information regarding one or more risks when changing the first element, allowing risks to be taken into consideration when changing an edit element that requires human judgment to change.

[0207] An embodiment of Invention A8 will be described as a modification of the fourth embodiment, in which the process of step S176 is replaced with a process of acquiring risk information from a large-scale language model server.

[0208] When the designer inputs "PS with a width of 270" on the screen of FIG. 8 and clicks the risk button corresponding to the first change candidate, "left closet with a width of 700" and "center closet with a width of 700" are acquired as the first element through steps S152 and S176, and a request is sent to the large-scale language model server. The request includes the acquired first element and a request to generate risk information regarding one or more risks associated with changing the first element. In response to this request, the large-scale language model server outputs the risk information. Upon receiving the risk information, the drawing creation support device 100 displays the risk of adopting the first change candidate based on the received risk information through step S178.

[0209] In this embodiment, step S176 corresponds to the request input means of invention 8 and the risk information acquisition means of invention 8.

[0210] Furthermore, in the first to fourth embodiments and their modifications, a trained model or a large-scale language model is used, but this is not limited to this, and the following inventions B1 to B8 can be applied to the drawing creation support device 100.

[0211] [Invention B1] The invention comprises an element information acquisition means for acquiring first element information relating to a first element which is an editable editing element in design information and which requires human judgment to change, and second element information relating to a second element which is affected by or affects the change to the first element, and a search means for searching, from a storage means for storing candidate information relating to one or more candidates for changing the first element in association with information relating to the first element and information relating to the second element, for the candidate information corresponding to the first element information and second element information acquired by the element information acquisition means.

[0212] This allows for obtaining candidate information regarding one or more candidates for changing the first element, making it easier to change an edit element that requires human judgment to change.

[0213] An embodiment of Invention B1 will be described as a variation of the first embodiment. The storage device 42 stores an edit history information table. One or more records are registered in the edit history information table. Each record includes a field for registering element information 400 of FIG. 2, a field for registering element information 402 of FIG. 2, and a field for registering dimension information 404 of FIG. 2. Then, similar to the processing in steps S150 and S152 of FIG. 6, a first element and a second element are acquired, and one or more pieces of dimension information 404 corresponding to the acquired first element and second element are searched for in the edit history information table. Subsequent processing is performed similarly to the processing in steps S156 to S160 of FIG. 6.

[0214] [Invention B2] The invention comprises an element information acquisition means for acquiring first element information relating to a first element which is an editable editing element in design information and which requires human judgment to change, and second element information relating to a second element which is affected or affects the change to the first element, and a search means for searching impact information corresponding to the first element information and second element information acquired by the element information acquisition means from a storage means which stores impact information relating to one or more effects when the first element is changed, in association with information relating to the first element and information relating to the second element.

[0215] This provides impact information regarding one or more impacts when changing the first element, allowing the impacts to be taken into consideration when changing an edit element that requires human judgment to change.

[0216] An embodiment of Invention B2 will be described as a variation of the second embodiment. The storage device 42 stores an impact information table. One or more records are registered in the impact information table. Each record includes a field for registering element information 406 of FIG. 15, a field for registering element information 408 of FIG. 15, and a field for registering impact information 410 of FIG. 15. Then, similar to the processing of steps S150 and S152 of FIG. 16, a first element and a second element are acquired, and one or more pieces of impact information 410 corresponding to the acquired first element and second element are searched for in the impact information table. Subsequent processing is performed similarly to the processing of steps S166, S158, and S160 of FIG. 16.

[0217] [Invention B3] The system comprises an element information acquisition means for acquiring first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and second element information relating to a second element that is affected by or affects the change to the first element, and a search means for searching, from a storage means that stores consideration information relating to one or more matters that should be taken into consideration when changing the first element, in association with information relating to the first element and information relating to the second element, for the consideration information corresponding to the first element information and second element information acquired by the element information acquisition means.

[0218] This provides consideration information regarding one or more items that should be taken into account when changing the first element, allowing consideration of precautions, etc. when changing an edit element that requires human judgment to change.

[0219] An embodiment of Invention B3 will be described as a variation of the third embodiment. The storage device 42 stores a consideration information table. One or more records are registered in the consideration information table. Each record includes a field for registering element information 412 of FIG. 21, a field for registering element information 414 of FIG. 21, and a field for registering consideration information 416 of FIG. 21. Then, similar to the processing in steps S150 and S152 of FIG. 22, a first element and a second element are acquired, and one or more pieces of consideration information 416 corresponding to the acquired first element and second element are searched for in the consideration information table. Subsequent processing is performed similarly to the processing in steps S172, S158, and S160 of FIG. 22.

[0220] [Invention B4] The system comprises an element information acquisition means for acquiring first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and second element information relating to a second element that is affected by or affects the change to the first element, and a search means for searching for risk information corresponding to the first element information and second element information acquired by the element information acquisition means from a storage means that stores risk information relating to one or more risks when changing the first element in association with information relating to the first element and information relating to the second element.

[0221] This provides risk information regarding one or more risks when changing the first element, allowing risks to be taken into consideration when changing an edit element that requires human judgment to change.

[0222] An embodiment of Invention B4 will be described as a variation of the fourth embodiment. The storage device 42 stores a risk information table. One or more records are registered in the risk information table. Each record includes a field for registering element information 418 of FIG. 27, a field for registering element information 420 of FIG. 27, and a field for registering risk information 422 of FIG. 27. Then, similar to the processing of steps S150 and S152 of FIG. 28, a first element and a second element are acquired, and one or more pieces of risk information 422 corresponding to the acquired first element and second element are searched for in the risk information table. Subsequent processing is performed similarly to the processing of steps S178, S158, and S160 of FIG. 28.

[0223] [Invention B5] The invention comprises an element information acquisition means for acquiring first element information relating to a first element which is an editable editing element in design information and requires human judgment to change, and a search means for searching, from a storage means for storing candidate information relating to one or more candidates for changing the first element in association with information about the first element, for the candidate information corresponding to the first element information acquired by the element information acquisition means.

[0224] This allows for obtaining candidate information regarding one or more candidates for changing the first element, making it easier to change an edit element that requires human judgment to change.

[0225] An embodiment of Invention B5 will be described as a variation of the first embodiment. The storage device 42 stores an edit history information table. One or more records are registered in the edit history information table. Each record includes a field for registering the element information 400 of FIG. 2 and a field for registering the dimension information 404 of FIG. 2. Then, a first element is acquired in the same manner as in step S152 of FIG. 6, and one or more pieces of dimension information 404 corresponding to the acquired first element are searched for in the edit history information table. Subsequent processing is performed in the same manner as in steps S156 to S160 of FIG. 6.

[0226] [Invention B6] The invention comprises an element information acquisition means for acquiring first element information relating to a first element which is an editable editing element in design information and requires human judgment to change, and a search means for searching impact information relating to one or more effects when changing the first element from a storage means which stores the impact information in association with information relating to the first element, and the impact information corresponding to the first element information acquired by the element information acquisition means.

[0227] This provides impact information regarding one or more impacts when changing the first element, allowing the impacts to be taken into consideration when changing an edit element that requires human judgment to change.

[0228] An embodiment of Invention B6 will be described as a variation of the second embodiment. The storage device 42 stores an impact information table. One or more records are registered in the impact information table. Each record includes a field for registering the element information 406 of FIG. 15 and a field for registering the impact information 410 of FIG. 15. Then, a first element is acquired in the same manner as in step S152 of FIG. 16, and one or more pieces of impact information 410 corresponding to the acquired first element are searched for in the impact information table. Subsequent processing is performed in the same manner as in steps S166, S158, and S160 of FIG. 16.

[0229] [Invention B7] The invention comprises an element information acquisition means for acquiring first element information relating to a first element that is an editable editing element in design information and requires human judgment to change, and a search means for searching for consideration information relating to one or more matters that should be taken into consideration when changing the first element, in correspondence with information about the first element, from a storage means for storing the consideration information corresponding to the first element information acquired by the element information acquisition means.

[0230] This provides consideration information regarding one or more items that should be taken into account when changing the first element, allowing consideration of precautions, etc. when changing an edit element that requires human judgment to change.

[0231] An embodiment of Invention B7 will be described as a variation of the third embodiment. The storage device 42 stores a consideration information table. One or more records are registered in the consideration information table. Each record includes a field for registering the element information 412 of FIG. 21 and a field for registering the consideration information 416 of FIG. 21. Then, a first element is acquired in the same manner as in step S152 of FIG. 22, and one or more pieces of consideration information 416 corresponding to the acquired first element are searched for in the consideration information table. Subsequent processing is performed in the same manner as in steps S172, S158, and S160 of FIG. 22.

[0232] [Invention B8] The invention comprises an element information acquisition means for acquiring first element information relating to a first element which is an editable editing element in design information and requires human judgment to change, and a search means for searching for risk information relating to one or more risks when changing the first element from a storage means which stores the risk information in association with information relating to the first element, and the risk information corresponding to the first element information acquired by the element information acquisition means.

[0233] This provides risk information regarding one or more risks when changing the first element, allowing risks to be taken into consideration when changing an edit element that requires human judgment to change.

[0234] An embodiment of invention B8 will be described as a variation of the fourth embodiment. The storage device 42 stores a risk information table. One or more records are registered in the risk information table. Each record includes a field for registering the element information 418 of FIG. 27 and a field for registering the risk information 422 of FIG. 27. Then, a first element is acquired in the same manner as in step S152 of FIG. 28, and one or more pieces of risk information 422 corresponding to the acquired first element are searched for in the risk information table. Subsequent processing is performed in the same manner as in steps S178, S158, and S160 of FIG. 28.

[0235] Furthermore, in the first to fourth embodiments and their modifications, a configuration including any one of an estimation process using a trained model, an acquisition process from a large-scale language model, and a search process using a table is employed. However, this is not limited to this, and a configuration including multiple of these processes can be employed. In this case, it is possible to control which process is prioritized. For example, it is possible to employ (1) a configuration in which, among multiple processes, a process with a low current load is prioritized; (2) a configuration in which, among multiple processes, a process whose results have been adopted by designers to a high degree (referring to the number of times, percentage, or other degree of adoption) is prioritized; or (3) a configuration in which, among multiple processes, a process whose results have been used by designers to a high degree (referring to the number of times, percentage, or other degree of adoption) is prioritized.

[0236] In the first embodiment and its modifications, one or more dimensions after modification are estimated when the first element is modified, but this is not limiting. Learning can be performed based on candidate information regarding one or more candidates for modifying the first element, and one or more candidates can be estimated from the acquired first element and second element. For example, the following configuration can be adopted.

[0237] The first configuration can estimate candidates for the weight, material, unit price, and other specifications as attributes other than the dimensions of the first element.

[0238] The second configuration uses a trained model trained based on element information 402 about the first element before and after a change to the first element and element information 402 about the second element to estimate one or more candidates (changed first elements) from the acquired first and second elements. Here, information about the changed first element corresponds to candidate information. In the example of FIG. 2 , instead of the dimension information 404 "650" in the first and second lines, the changed first elements "left closet with a width of 650" and "center closet with a width of 650" are registered. Note that the second configuration can also be similarly applied to the above-described modified example that employs a trained model trained based on element information 400 about the first element and dimension information 404 about one or more changed dimensions when the first element is changed.

[0239] In the first to fourth embodiments and their modifications, the data structure of Fig. 2 is used as the edit history data, but the data structure is not limited to this and may include information on other attributes of the first and second elements (for example, ID, specifications, coordinates in the drawing, location in the building). The same applies to the impact data in the second embodiment, the consideration data in the third embodiment, and the risk data in the fourth embodiment.

[0240] Furthermore, in the second to fourth embodiments and their variations, the dimensions after the change when the first element is changed are estimated and displayed, but this is not limited to this, and a configuration can be adopted in which the dimensions after the change when the first element is changed are not estimated and displayed.

[0241] Furthermore, in the first to fourth embodiments and their variations, modification candidates for the first element are displayed in step S156. However, this is not limited to this. If there is only one estimated modification candidate, it is possible to adopt a configuration in which the modification candidate is not displayed, and the dimensions of the first element obtained in step S152 are changed to the modified dimensions estimated in step S154.

[0242] Furthermore, in the first to fourth embodiments and their variations, the dimensions of the first element obtained in step S152 are changed to the changed dimensions selected in step S158, but this is not limiting, and a configuration in which the first element is not changed can also be adopted.

[0243] Furthermore, in the above first to fourth embodiments and their variations, the post-change dimensions, impacts, considerations, or risks are estimated for the first element, which is an editable editing element in the design information and requires human judgment to change; however, it is not limited to this, and the post-change dimensions, impacts, considerations, or risks can be estimated for editing elements other than editing elements that require human judgment to change (including editing elements that do not require human judgment to change), virtual elements, and other elements.

[0244] Furthermore, in the first to fourth embodiments and their variations, the first element and the second element are acquired in steps S150 and S152, but this is not limited to this. Alternatively, it is possible to adopt a configuration in which (1) the designer's voice is input and analyzed to acquire the first element or the second element, or (2) an architectural drawing is input and analyzed to acquire the first element or the second element.

[0245] In the first to fourth embodiments and their modifications, the device is implemented as a single device, but the present invention is not limited to this and can also be implemented as a network system. As an example of a network system, some or all of the functions of the drawing creation support device 100 can be configured as a virtual server on a server that provides cloud computing services.

[0246] Furthermore, in the above first to fourth embodiments and their variations, the drawing creation support device 100 is configured to use the storage device 42, but this is not limited to this, and it can also be configured to use an external storage device such as a database server.

[0247] Furthermore, in the above first to fourth embodiments and their variations, when executing the processes shown in the flowcharts of Figures 4, 6, 16, 22 and 28, the case where a program pre-stored in ROM 32 is executed has been described. However, this is not limited to this, and the program showing these procedures may be read into RAM 34 from a storage medium on which the program is stored and executed.

[0248] Here, storage media refers to semiconductor storage media such as RAM and ROM, magnetic storage storage media such as FD and HD, optically readable storage media such as CD, CDV, LD and DVD, and magnetic storage / optically readable storage media such as MO, and includes any storage media that can be read by a computer, regardless of the reading method, whether electronic, magnetic, optical, etc.

[0249] Furthermore, the first to fourth embodiments and their modifications can be applied to each other. Furthermore, the present invention is not limited to the first to fourth embodiments and their modifications, and can be applied to other cases without departing from the spirit of the present invention. For example, the present invention can be applied to a wide range of design cases, such as automobile design, machine design, and circuit design.

[0250] 100...Drawing creation support device, 30...CPU, 32...ROM, 34...RAM, 38...I / F, 39...bus, 40...input device, 42...storage device, 44...display device, 400, 402, 406, 408, 412, 414, 418, 420...element information, 404...dimension information, 410...influence information, 416...consideration information, 422...risk information

Claims

1. A design support system comprising: request input means for inputting a request including first element information regarding a first element which is an editable editing element in design information and for which human judgment is required for its change, and second element information regarding a second element which is affected by or affects the change of the first element, and including a request for generating candidate information regarding one or more candidates for changing the first element, into a large language model; and candidate information acquisition means for acquiring the candidate information output from the large language model in response to the request.

2. A design support system comprising: request input means for inputting a request including first element information regarding a first element which is an editable editing element in design information and for which human judgment is required for its change, and second element information regarding a second element which is affected by or affects the change of the first element, and including a request for generating impact information regarding one or more impacts when the first element is changed, into a large language model; and impact information acquisition means for acquiring the impact information output from the large language model in response to the request.

3. A design support system comprising: request input means for inputting a request including first element information regarding a first element which is an editable editing element in design information and for which human judgment is required for its change, and second element information regarding a second element which is affected by or affects the change of the first element, and including a request for generating consideration information regarding one or more considerations to be taken into account when the first element is changed, into a large language model; and consideration information acquisition means for acquiring the consideration information output from the large language model in response to the request.

4. A design support system comprising: request input means for inputting a request including first element information regarding a first element which is an editable editing element in design information and for which human judgment is required for its change, and second element information regarding a second element which is affected by or affects the change of the first element, and including a request for generating risk information regarding one or more risks when the first element is changed, into a large language model; and risk information acquisition means for acquiring the risk information output from the large language model in response to the request.

5. A design support system comprising: a request input means for inputting a request including first element information regarding a first editable element in design information that requires human judgment for its change and including a request for generating candidate information regarding one or more candidates for changing the first element into a large language model; and a candidate information acquisition means for acquiring candidate information output from the large language model in response to the request.

6. A design support system comprising: a request input means for inputting a request including first element information regarding a first editable element in design information that requires human judgment for its change and including a request for generating impact information regarding one or more impacts when the first element is changed into a large language model; and an impact information acquisition means for acquiring impact information output from the large language model in response to the request.

7. A design support system comprising: a request input means for inputting a request including first element information regarding a first editable element in design information that requires human judgment for its change and including a request for generating consideration item information regarding one or more items to be considered when the first element is changed into a large language model; and a consideration item information acquisition means for acquiring consideration item information output from the large language model in response to the request.

8. A design support system comprising: a request input means for inputting a request including first element information regarding a first editable element in design information that requires human judgment for its change and including a request for generating risk information regarding one or more risks when the first element is changed into a large language model; and a risk information acquisition means for acquiring risk information output from the large language model in response to the request.

9. The design support system according to any one of claims 1 and 5, wherein the information regarding the first element is information regarding the first element before change among the first elements before and after the change when the first element is changed, and the candidate information is information regarding the first element after change among the first elements before and after the change.

10. The design support system according to any one of claims 1 to 8, wherein the design information is design information for performing architectural design.

Images