Design assistance system

Abstract

Provided is a design assistance system suitable for figuring out a plurality of elements to be successively edited and an editing order thereof.　A drawing creation assistance device 100 inputs, to an AI model 50, a request including an element ID of an already created or edited element and including an instruction to generate a plurality of editing elements edited successively after the editing element that has already been created or edited, and to generate editing order information related to the editing order thereof, and acquires information output from the AI model 50 in response to the request. Thus, the plurality of editing elements to be edited successively after the editing element that has already been created or edited, and the editing order information related to the editing order thereof, can be obtained, whereby the plurality of editing elements to be successively edited and the editing order thereof can be figured out.

Description

Design support system

[0001] The present invention relates to a system for supporting design, and particularly to a design support system suitable for grasping a plurality of continuously edited elements and their editing order.

[0002] Conventionally, as a technique for supporting design using AI (Artificial Intelligence), for example, the technique described in Patent Document 1 is known.

[0003] The technique described in Patent Document 1 assigns a reward R to a combination of a state S determined depending on the execution of a design activity and an action A that is an activity selectable under the state, and constructs a learned model by maximizing the value. Then, using the learned model, the action to be taken next from the current state is inferred.

[0004] Japanese Unexamined Patent Application Publication No. 2022-56238

[0005] In architectural design, there are cases where a plurality of related elements of a certain unit are continuously edited. For example, in the design of a toilet, the design of a toilet bowl and a washbasin are continuously carried out. However, in the technique described in Patent Document 1, since it is for inferring the action to be taken next from the current state, there is a problem that a plurality of continuously edited elements and their editing order cannot be grasped.

[0006] Therefore, the present invention has been made by paying attention to such an unsolved problem of the conventional technique, and aims to provide a design support system suitable for grasping a plurality of continuously edited elements and their editing order.

[0007] [Invention 1] In order to achieve the above object, the design support system of Invention 1 includes input means for inputting a request including element information regarding created or edited elements in design information and including a request for generating editing order information regarding a plurality of elements to be continuously edited next to the element and their editing order into the AI model, and acquisition means for acquiring information output from the AI model for the request.

[0008] In this configuration, the input means inputs a request to the AI ​​model that includes element information and a request to generate editing sequence information, and the acquisition means acquires the information output from the AI ​​model in response to the request.

[0009] Here, the input means include, for example, directly inputting a request into the AI ​​model, or indirectly inputting it into the AI ​​model through processing, functions, devices, networks, or other means.

[0010] Furthermore, means of acquisition include, for example, directly acquiring output information from an AI model, or indirectly acquiring output information from an AI model through processing, functions, devices, networks, or other means.

[0011] Furthermore, requests for generating editing order information include, for example, requests for generating answer information to questions regarding multiple elements and their editing order.

[0012] Furthermore, generation requests include, for example, those that explicitly request the generation of editing order information, and those that indirectly request the generation of editing order information. An explicit request would be, for example, "Please generate editing order information." An indirect request would be, for example, to request the generation of answer information to questions about multiple elements and their editing order, by making questions about multiple elements and their editing order the generation request. This is because the AI ​​model will generate the answer information if questions about multiple elements and their editing order are input as prompts.

[0013] Furthermore, element information and requests can be composed of, for example, vector data or any other format.

[0014] Furthermore, element information can consist of the element itself, as information for identifying the element (e.g., name, number, ID, code, URL or other link information), or as feature information relating to the element's overview, statistics, or other characteristics. Element information can also consist of characters, numbers, figures, codes, symbols, images, sounds, or other information. Additionally, element information can consist of keywords related to the element (e.g., one or more keywords indicating part of the element's name).

[0015] Furthermore, editing sequence information can consist of, for example, multiple elements and their editing sequence itself, or it can be composed of information for identifying elements and editing sequences (e.g., name, number, ID, code, URL or other link information), or it can be composed of characteristic information relating to the outline, statistics, or other features of elements and editing sequences. Editing sequence information can also be composed of, for example, characters, numbers, figures, codes, symbols, images, sounds, or other information. Editing sequence information can also be composed of keywords relating to elements and editing sequences (e.g., one or more keywords indicating part of the names of elements and editing sequences).

[0016] Furthermore, this system may be implemented as a single device, apparatus, terminal, or other device, or as a network system in which multiple devices, apparatus, terminals, or other devices are connected in a communicative manner. In the latter case, each component may belong to any of the multiple devices, as long as it is connected in a communicative manner.

[0017] [Invention 2] Furthermore, the design support system of Invention 2 includes a registration means for registering reference information, which includes element information relating to created or edited elements, and editing sequence information relating to a plurality of elements to be edited consecutively after the element and their editing order, in a knowledge base that the AI ​​model can access, and the acquisition means, in response to the request, obtains information output from the AI ​​model by referring to the reference information in the knowledge base.

[0018] In this configuration, the registration means registers reference information in the knowledge base. Then, the retrieval means retrieves the information output from the AI ​​model in response to a request. At this time, the AI ​​model references the reference information in the knowledge base.

[0019] Here, the knowledge base stores reference information by any means and at any time. The reference information may be stored in advance, or it may not be stored in advance, but rather stored through external input or other means during the operation of the system.

[0020] [Invention 3] Furthermore, in the design support system of Invention 3, the registration means registers in the knowledge base element information relating to created or edited elements, editing sequence information relating to a plurality of elements to be edited consecutively after the element and their editing order, and reference information including evaluation values ​​relating to the editing of the plurality of elements.

[0021] [Invention 4] Furthermore, the design support system of Invention 4 is the design support system of Invention 3, wherein the registration means registers in the knowledge base first reference information including the element information, the editing sequence information and the evaluation value based on a first index that indicates an index of the value of the evaluation value, and second reference information including the element information, the editing sequence information and the evaluation value based on a second index different from the first index, and the index information acquisition means acquires index information relating to the first index or the second index, and the input means inputs to the AI ​​model a request that includes a request to refer to either the first reference information or the second reference information of the knowledge base based on the index information acquisition means.

[0022] In this configuration, the registration means registers the first and second reference information in the knowledge base. Then, the indicator information acquisition means acquires the indicator information, and the input means inputs a request to the AI ​​model that includes a request to refer to either the first or second reference information in the knowledge base based on the acquired indicator information.

[0023] Here, the indicator information acquisition means may, for example, input indicator information from an input device, acquire or receive indicator information from an external terminal, read indicator information from a storage device or storage medium, or generate or calculate indicator information through information processing. Therefore, acquisition includes at least input, acquisition, reception, reading (including retrieval), generation, and calculation. The same applies hereafter to the design support system of Invention 7.

[0024] [Invention 5] Furthermore, in the design support system of Invention 5, the request includes, in the design support system of Invention 1, element information relating to an element that has been created or edited, and reference information relating to a plurality of elements to be edited consecutively after the element and the editing order thereof.

[0025] In this configuration, the input means inputs a request to the AI ​​model that includes element information and reference information, and a request to generate editing sequence information, and the acquisition means acquires the information output from the AI ​​model in response to the request.

[0026] [Invention 6] Furthermore, in the design support system of Invention 6, the request includes element information relating to an element that has been created or edited, editing sequence information relating to a plurality of elements to be edited consecutively after the element and the editing order thereof, and reference information including evaluation values ​​relating to the editing of the plurality of elements.

[0027] [Invention 7] Furthermore, the design support system of Invention 7 is equipped with an indicator information acquisition means that acquires indicator information relating to a first indicator that shows an indicator of the value of the evaluation value or a second indicator different from the first indicator, in addition to the design support system of Invention 6. The input means inputs the request to the AI ​​model, based on the indicator information acquired by the indicator information acquisition means, which includes either first reference information including the element information, the editing sequence information and the evaluation value based on the first indicator, or second reference information including the element information, the editing sequence information and the evaluation value based on the second indicator.

[0028] In this configuration, the indicator information acquisition means acquires indicator information, and the input means inputs a request to the AI ​​model that includes either the first reference information or the second reference information based on the acquired indicator information.

[0029] [Invention 8] Furthermore, the design support system of Invention 8 is a design support system of any one of Inventions 1 to 7, comprising a identification means for identifying the plurality of elements and their editing order as related groups if their occurrence frequency in the design information is predetermined or greater, and the request includes a request to generate the editing order information relating to the plurality of elements and their editing order as related groups identified by the identification means.

[0030] In this configuration, a specific means identifies multiple elements and their editing order that appear more than a predetermined number of times as related groups, and an input means inputs a request to the AI ​​model that includes a request to generate editing order information regarding the multiple elements and their editing order as identified related groups.

[0031] [Invention 9] Furthermore, the design support system of Invention 9 is a design support system of any one of Inventions 1 to 7, wherein the plurality of elements are elements whose setting or modification requires human judgment, and whose setting or modification affects other elements.

[0032] [Invention 10] Furthermore, the design support system of Invention 10 is a design support system of any one of Inventions 1 to 7, wherein the design information is design information for designing buildings.

[0033] As explained above, the design support system of Invention 1 allows you to obtain editing order information regarding multiple elements to be edited consecutively after an already created or edited element, and the order in which they are edited. Therefore, you can understand multiple elements to be edited consecutively and the order in which they are edited.

[0034] Furthermore, according to the design support system of invention 3 or 6, it is possible to obtain editing order information relating to multiple elements with high evaluation values ​​and their editing order.

[0035] Furthermore, according to the design support system of invention 4 or 7, it is possible to obtain editing order information relating to multiple elements with high evaluation values ​​based on the first or second indicator, and their editing order.

[0036] Furthermore, according to the design support system of Invention 8, editing can be performed while visualizing related groups.

[0037] Furthermore, according to the design support system of Invention 9, it is possible to obtain editing order information relating to multiple elements and their editing order, taking into account the relationship between elements that require human judgment for setting or changing and other elements that are affected by that setting or changing.

[0038] This is a diagram showing the hardware configuration of the drawing creation support device 100. This is a diagram showing the structure of the CAD data of the cross section drawing. This is a diagram showing the structure of the editing history data. This is a flowchart showing the training data generation process. This is a diagram showing the structure of the training data. This is a flowchart showing the trained model generation process. This is a block diagram showing the process of generating and using the trained model. This is a flowchart showing the editing order information estimation process. This is a cross section drawing assuming the delivery of a piano. This is a diagram showing the structure of the editing history data. This is a diagram showing the structure of the training data. This is a flowchart showing the editing order information estimation process. This is a block diagram showing the configuration of the network system according to this embodiment. This is a functional block diagram of the generation AI server 120. This is a flowchart showing the training data registration process. This is a flowchart showing the editing order information acquisition process. This is a flowchart showing the editing order information acquisition process.

[0039] [First Embodiment] The first embodiment of the present invention will be described below. Figures 1 to 9 show this embodiment.

[0040] [Configuration of this embodiment] First, the configuration of this embodiment will be described. Figure 1 is a diagram showing the hardware configuration of the drawing creation support device 100.

[0041] As shown in FIG. 1, the drawing creation support device 100 includes a CPU (Central Processing Unit) 30 that controls operations and the entire system based on a control program, a ROM (Read Only Memory) 32 that stores in advance the control program of the CPU 30 and the like in a predetermined area, a RAM (Random Access Memory) 34 that stores data read from the ROM 32 and the like and calculation results necessary in the calculation process of the CPU 30, and an I / F (InterFace) 38 that mediates data input / output to / from an external device. These are connected to each other via a bus 39, which is a signal line for transferring data, so that data can be exchanged between them.

[0042] Connected to the I / F 38 as external devices are an input device 40 composed of a keyboard, a 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.

[0043] The storage device 42 is installed with CAD (Computer Aided Design) software and BIM (Building Information Modeling) software (hereinafter collectively referred to as "CAD software"). CAD software is software that supports the creation of drawings according to the operations of a designer. When the startup of CAD software is requested, the CPU 30 starts the program for CAD software stored in a predetermined area of the ROM 32 and executes processing according to that program. The designer can start the CAD software and create orthographic drawings, plan details, and other architectural drawings.

[0044] Next, the data structure of the storage device 42 will be described. The storage device 42 stores CAD data of orthographic drawings, plan details, and other architectural drawings.

[0045] FIG. 2 is a diagram showing the structure of CAD data of a floor plan drawing. The CAD data of the floor plan drawing is data constituting a drawing that details the cross-section of a building, and is configured as data including one or more creatable or editable elements (hereinafter referred to as "editable elements"). The CAD data of the floor plan drawing is created by a designer using CAD software. The designer creates a floor plan drawing by creating, setting, changing, or deleting (hereinafter referred to as "editing") editable elements in the CAD software. In the example of FIG. 2, each editable element of the floor, wall, and ceiling in the area labeled "Internal Corridor" and each editable element of the floor, wall, and ceiling in the area labeled "Vestibule" are respectively arranged.

[0046] The same applies to CAD data of detailed floor plans and other architectural drawings, which are configured as data including one or more editable elements.

[0047] The storage device 42 stores edit history data indicating the history of editing editable elements for each CAD data. The final editing result related to the edit history data is reflected in the CAD data.

[0048] FIG. 3 is a diagram showing the structure of the edit history data. As shown in FIG. 3, the edit history data includes, for each edited editable element, element information 400 regarding the editable element and the editing time 402 required for editing the editable element, in the order of editing. The element information 400 includes an element ID for identifying the editable element, the area to be edited of the editable element, and the editable element. The editing time 402 can be calculated, for example, by subtracting the editing start time from the editing end time.

[0049] In the example of FIG. 3, as a first related group, it shows that the editable elements of the toilet, "toilet bowl", "washbasin counter", "mirror", and "towel rack" are continuously edited in that order. Element IDs "52", "53", "54", and "55" are assigned to these editable elements, and the editing times respectively require 35, 38, 70, and 94 minutes.

[0050] Furthermore, a second related group is shown, indicating that the editing elements of the toilet—"ventilation fan," "lighting," "storage," and "outlet"—were edited in that order consecutively. These editing elements were assigned element IDs "56," "57," "58," and "59," and the editing times were 94, 48, 42, and 74 minutes, respectively.

[0051] Furthermore, a third related group is shown, indicating that the closet editing elements "shelf board," "hanger pipe," "drawer," and "basket" are edited in that order consecutively. These editing elements are assigned element IDs "60," "61," "62," and "63," and the editing times required are 96, 74, 84, and 22 minutes, respectively.

[0052] Furthermore, a fourth related group is shown, indicating that the kitchen editing elements "flooring," "wall covering," "countertop," and "sink" are edited in that order consecutively. These editing elements are assigned element IDs "64," "65," "66," and "67," and the editing times were 42, 44, 35, and 32 minutes, respectively.

[0053] Since the editing history data is used to create training data, the storage device 42 stores a large amount of editing history data that has been created in the past.

[0054] [Operation of this embodiment] Next, the operation of this embodiment will be described. [Training data generation process] Figure 4 is a flowchart of the training data generation process.

[0055] The training data generation process is a process that generates training data, and when it is executed on the CPU 30, it proceeds to step S100, as shown in Figure 4.

[0056] In step S100, the unprocessed editing history data is retrieved from the storage device 42, the process moves to step S102, the variable n is set to "2", and the process moves to step S104.

[0057] In steps S104 to S108, the element ID of the created or edited editing element, the n editing elements (the number indicated by the value of variable n) that were edited consecutively after that editing element, their editing order, and their editing time are obtained from the editing history data acquired in step S100. Figure 3 is an example and explains the case when the value of variable n is "2". In the editing history data in Figure 3, each row is arranged in editing order.

[0058] When the second row is targeted, the element IDs "01" to "51" of the previously edited elements are obtained as created or edited element IDs. Since the value of variable n is "2", "toilet" and "handwashing counter" are obtained as edited elements, and "35" and "38" are obtained as editing times.

[0059] When the third row is targeted, the element IDs "01" to "52" of the previously edited elements are obtained as created or edited element IDs. Since the value of variable n is "2", "Handwashing Counter" and "Mirror" are obtained as edited elements, and "38" and "70" are obtained as editing times.

[0060] Next, the process moves to step S110, where the evaluation value is calculated by multiplying the sum of the editing times obtained in step S108 by "-1". In the example in the second line above, editing times of "35" and "38" are obtained, so the evaluation value is calculated as (35 + 38) × -1 = -73. The reason for multiplying by "-1" is to set a higher evaluation value for shorter editing times.

[0061] Next, the process moves to step S112, where the element IDs, editing elements and their editing order obtained in steps S104 to S108, and the evaluation values ​​calculated in step S110 are associated and registered as learning data. Then, the process moves to step S114, where "1" is added to the value of variable n, and finally, the process moves to step S116.

[0062] In step S116, it is determined whether the value of variable n is greater than "4". If it is determined to be less than or equal to "4" (NO), the process proceeds to step S104. Then, steps S104 to S114 are repeated until the value of variable n becomes "4".

[0063] Using Figure 3 as an example, the process from steps S104 to S110 will be explained for the case where the value of variable n is "3".

[0064] When the second row is targeted, the element IDs "01" to "51" of the previously edited elements are obtained as created or edited element IDs. Since the value of variable n is "3", "toilet", "handwashing counter", and "mirror" are obtained as edited elements, and "35", "38", and "70" are obtained as editing times. The evaluation value is calculated as (35 + 38 + 70) × -1 = -143.

[0065] When the third row is targeted, the element IDs of previously edited elements, "01" to "52", are obtained as created or edited element IDs. Since the value of variable n is "3", "handwashing counter", "mirror", and "towel rack" are obtained as edited elements, and "38", "70", and "94" are obtained as editing times. The evaluation value is calculated as (38 + 70 + 94) × -1 = -202.

[0066] Furthermore, using Figure 3 as an example, the process of steps S104 to S110 will be explained for the case where the value of variable n is "4".

[0067] When the second row is targeted, the element IDs "01" to "51" of previously edited elements are obtained as created or edited element IDs. Since the value of variable n is "4", "toilet", "handwashing counter", "mirror", and "towel rack" are obtained as edited elements, and "35", "38", "70", and "94" are obtained as editing times. The evaluation value is calculated as (35 + 38 + 70 + 94) × -1 = -237.

[0068] When the third row is targeted, the element IDs "01" to "52" of the previously edited elements are obtained as created or edited element IDs. Since the value of variable n is "4", "handwashing counter", "mirror", "towel rack", and "ventilation fan" are obtained as edited elements, and "38", "70", "94", and "94" are obtained as editing times. The evaluation value is calculated as (38 + 70 + 94 + 94) × -1 = -296.

[0069] On the other hand, if it is determined in step S116 that the value of variable n is greater than "4" (YES), the process proceeds to step S118 to determine whether the processing in steps S100 to S116 has been completed for all edit history data. If it is determined that the processing has been completed for all edit history data (YES), the process proceeds to step S120.

[0070] In step S120, the learning data in which the element ID and other information were registered in step S112 is stored in the storage device 42.

[0071] Figure 5 shows the structure of the training data. As shown in Figure 5, each row of the training data contains the created or edited element ID 410, editing order information 412, and evaluation value 414. The editing order information 412 includes the area to be edited for the editing element, the editing element, and its editing order.

[0072] The second row of Figure 5 shows two editing elements that were edited consecutively after the editing element in the first row of Figure 3, along with their editing order and evaluation value. The third row of Figure 5 shows two editing elements that were edited consecutively after the editing element in the second row of Figure 3, along with their editing order and evaluation value. The fourth row of Figure 5 shows two editing elements that were edited consecutively after the editing element in the third row of Figure 3, along with their editing order and evaluation value. Furthermore, the sixth row of Figure 5 shows two editing elements that were edited consecutively after the editing element in the fifth row of Figure 3, along with their editing order and evaluation value. The seventh row of Figure 5 shows two editing elements that were edited consecutively after the editing element in the sixth row of Figure 3, along with their editing order and evaluation value. The eighth row of Figure 5 shows two editing elements that were edited consecutively after the editing element in the seventh row of Figure 3, along with their editing order and evaluation value.

[0073] Furthermore, the 10th row of Figure 5 shows the three editing elements that were edited consecutively after the editing element in the 1st row of Figure 3, along with their editing order and evaluation value. The 11th row of Figure 5 shows the three editing elements that were edited consecutively after the editing element in the 2nd row of Figure 3, along with their editing order and evaluation value. Additionally, the 13th row of Figure 5 shows the three editing elements that were edited consecutively after the editing element in the 5th row of Figure 3, along with their editing order and evaluation value. The 14th row of Figure 5 shows the three editing elements that were edited consecutively after the editing element in the 6th row of Figure 3, along with their editing order and evaluation value.

[0074] Furthermore, row 16 of Figure 5 shows the four editing elements that were edited consecutively after the editing element in row 1 of Figure 3, along with their editing order and evaluation values. Similarly, row 18 of Figure 5 shows the four editing elements that were edited consecutively after the editing element in row 5 of Figure 3, along with their editing order and evaluation values.

[0075] [Trained Model Generation Process] Figure 6 is a flowchart showing the trained model generation process.

[0076] Figure 7 is a block diagram showing the process of generating and using a trained model. The trained model generation process is a process performed to generate a trained model, and when executed on the CPU 30, it proceeds to step S200, as shown in Figure 6, to execute the training data analysis process. In the training data analysis process, training data is read from the storage device 42, and the created or edited element ID 410, editing order information 412, and evaluation value 414 are extracted from the read training data.

[0077] Next, the process moves to step S202. In step S202, as shown in Figure 7, a training dataset is generated based on the information extracted in step S200. The process then moves to step S204, where the generated training dataset is input into the training program, and the training program generates a trained model. The training program includes pre-training parameters and hyperparameters, and performs training based on the input training dataset and hyperparameters, updating the pre-training parameters. As a training method, for example, reinforcement learning can be employed. In reinforcement learning, evaluation values ​​are assigned to combinations of created or edited element IDs and editing order information relating to multiple editing elements that were edited consecutively after that edited element, and training is performed to maximize the evaluation values. Finally, a trained model is output as the training result.

[0078] The trained model is trained to maximize the evaluation value 414 based on the created or edited element ID 410, editing order information 412, and evaluation value 414. The trained model comprises trained parameters, which are updated from pre-training parameters, and an inference program. The inference program takes the created or edited element ID as input, estimates the editing order information from the input created or edited element ID based on the trained parameters, and outputs the estimated editing order information. Note that the relationship between the input element ID and the output editing order information is determined by the AI's training, so while it shows a similar trend to the content of past training data, there is an ambiguity that prevents it from being an exact match. However, this ambiguity can be reduced by increasing the amount of training data and the training accuracy.

[0079] Next, the process moves to step S206, where the trained model generated in step S204 is stored in the storage device 42, and the series of processes ends.

[0080] [Editing Order Information Estimation Process] Figure 8 is a flowchart showing the editing order information estimation process.

[0081] The editing order information estimation process is performed in response to requests from designers or other users, and when it is executed on the CPU 30, it first proceeds to step S300, as shown in Figure 8.

[0082] In step S300, the element IDs of created or edited elements are obtained from the CAD data currently being edited, and the process proceeds to step S302.

[0083] In step S302, the trained model in the storage device 42 is used to estimate the editing order information from the created or edited element IDs obtained in step S300. The estimation is performed by inputting the created or edited element IDs into the trained model and obtaining the editing order information output from the trained model.

[0084] Here, methods such as Monte Carlo tree search can be used to estimate a larger number of editing elements and their editing order with higher accuracy. In the example in Figure 3, priority is given to estimating three editing elements and their editing order rather than two, and four editing elements rather than three. This allows the designer to perform editing while visualizing larger editing units.

[0085] Furthermore, instead of prioritizing the longest sequence, editing elements and their editing order can be estimated as related groups. For example, A, B, C, D, E, F, G, and H can be estimated as the next editing elements to be edited consecutively from the current editing state. If A-D and E-H are related groups, A-D will be estimated instead of A-H. The determination of whether a group is related can be made, for example, by identifying two or more editing elements and their editing order from A-H that appear more than a predetermined number of times in the training data. This allows the designer to edit while visualizing related groups.

[0086] Next, the process moves to step S304, where the editing order information estimated by the trained model is displayed on the display device 44, and the series of processes is completed.

[0087] [When considering piano delivery] Next, we will explain the procedures when considering piano delivery.

[0088] Figure 9 is a cross-section drawing that assumes the delivery of a piano. If the designer wants to install a piano with a width of 120 mm in the cross-section drawing of Figure 9 using CAD software, they need to shorten the width of the toilet to prevent interference between the piano and the toilet wall when delivering the piano. However, changing the width of the toilet necessitates editing other editing elements of the toilet. Therefore, if the designer requests the estimation of the editing elements to be edited for the toilet and their editing order after changing the width of the toilet, the system will go through steps S300 to S304, retrieve the IDs of created or edited elements from the currently edited CAD data, and estimate and display the editing order information from the retrieved created or edited element IDs. For example, if the editing order information for row 16 of Figure 5 is estimated, "toilet bowl → hand washing counter → mirror → towel rack" will be displayed. If these editing elements are already included in the CAD data as shown in Figure 9, these editing elements will be highlighted (for example, displayed in a specific color or pattern) and connected by arrows to display the editing order.

[0089] [Effects of this embodiment] Next, the effects of this embodiment will be described. In this embodiment, created or edited element IDs are obtained, and edit order information is estimated from the obtained created or edited element IDs using a trained model that has been trained on training data including created or edited element IDs, and editing order information relating to a plurality of editing elements that were edited consecutively after the edited element and their editing order.

[0090] This allows you to obtain editing order information regarding multiple editing elements that are edited consecutively after an already created or edited editing element, and the order in which they are edited. Therefore, you can understand the sequence of editing of multiple editing elements that are edited consecutively and their editing order.

[0091] Furthermore, in this embodiment, the trained model is trained to maximize the evaluation value based on training data that includes the IDs of created or edited elements, editing order information relating to a plurality of editing elements that were edited consecutively after that editing element, and evaluation values ​​relating to the editing of those plurality of editing elements.

[0092] This allows us to obtain editing order information regarding multiple editing elements with high evaluation values ​​and their editing order.

[0093] Furthermore, in this embodiment, multiple editing elements and their editing order are identified as related groups if they appear more than a predetermined number of times, and editing order information is estimated for the multiple editing elements and their editing order based on the identified related groups.

[0094] This allows editing to be performed while visualizing related groups. Furthermore, in this embodiment, an evaluation value is assigned to the editing of multiple editing elements for each combination of an already created or edited element ID and editing order information relating to the multiple editing elements edited consecutively after that editing element, and a trained model is generated by training to maximize the evaluation value.

[0095] This allows us to obtain a trained model that provides editing order information regarding multiple editing elements with high evaluation values ​​and their editing order.

[0096] [Second Embodiment] Next, a second embodiment of the present invention will be described. Figures 10 to 12 show this embodiment. Figures 3, 5 and 9 will also be referenced.

[0097] This embodiment differs from the first embodiment in that it performs estimation using a first pre-trained model and a second pre-trained model, each trained using evaluation values ​​based on different indicators. Below, only the parts that differ from the first embodiment will be described, and the overlapping parts will be omitted.

[0098] [Configuration of this embodiment] First, the configuration of this embodiment will be described. Figure 10 is a diagram showing the structure of the editing history data.

[0099] The storage device 42 stores, for each CAD data, the editing history data shown in Figure 3, as well as the editing history data shown in Figure 10.

[0100] As shown in Figure 10, the editing history data includes, for each edited editing element, element information 404 related to that editing element and the number of editing items 406 required to edit that editing element, in the order of editing.

[0101] In the example in Figure 10, the first related group of editing elements for the toilet, "toilet bowl," "handwashing counter," "mirror," and "towel rack," are edited in that order consecutively. These editing elements are assigned element IDs "52," "53," "54," and "55," and the number of editable items is 8, 5, 6, and 9, respectively.

[0102] Furthermore, a second related group is shown, indicating that the editing elements for the toilet—"ventilation fan," "lighting," "storage," and "outlet"—are edited in that order consecutively. These editing elements are assigned element IDs "56," "57," "58," and "59," with the number of editable items being 3, 7, 8, and 6 respectively.

[0103] Furthermore, a third related group is shown, indicating that the editing elements of the closet—"shelves," "hanger pipes," "drawers," and "baskets"—are being edited in that order consecutively. These editing elements are assigned element IDs "60," "61," "62," and "63," with the number of editable items being 8, 7, 2, and 4 respectively.

[0104] Furthermore, a fourth related group is shown, indicating that the kitchen editing elements "flooring," "wall covering," "countertop," and "sink" are edited in that order consecutively. These editing elements are assigned element IDs "64," "65," "66," and "67," with the number of editable items being 7, 4, 3, and 8 respectively.

[0105] [Operation of this embodiment] Next, the operation of this embodiment will be described. [Training data generation process] In the training data generation process, training data is generated based on the editing history data in Figure 10, similar to the training data generation in Figure 5.

[0106] Figure 11 shows the structure of the training data. As shown in Figure 11, each row of the training data contains the created or edited element ID 410, editing order information 412, and evaluation value 414.

[0107] The second row of Figure 11 shows two editing elements that were edited consecutively after the editing element in the first row of Figure 10, along with their editing order and evaluation value. The third row of Figure 11 shows two editing elements that were edited consecutively after the editing element in the second row of Figure 10, along with their editing order and evaluation value. The fourth row of Figure 11 shows two editing elements that were edited consecutively after the editing element in the third row of Figure 10, along with their editing order and evaluation value. Furthermore, the sixth row of Figure 11 shows two editing elements that were edited consecutively after the editing element in the fifth row of Figure 10, along with their editing order and evaluation value. The seventh row of Figure 11 shows two editing elements that were edited consecutively after the editing element in the sixth row of Figure 10, along with their editing order and evaluation value. The eighth row of Figure 11 shows two editing elements that were edited consecutively after the editing element in the seventh row of Figure 10, along with their editing order and evaluation value.

[0108] Furthermore, the 10th row of Figure 11 shows the three editing elements that were edited consecutively after the editing element in the 1st row of Figure 10, along with their editing order and evaluation value. The 11th row of Figure 11 shows the three editing elements that were edited consecutively after the editing element in the 2nd row of Figure 10, along with their editing order and evaluation value. Additionally, the 13th row of Figure 11 shows the three editing elements that were edited consecutively after the editing element in the 5th row of Figure 10, along with their editing order and evaluation value. The 14th row of Figure 11 shows the three editing elements that were edited consecutively after the editing element in the 6th row of Figure 10, along with their editing order and evaluation value.

[0109] Furthermore, row 16 of Figure 11 shows the four editing elements that were edited consecutively after the editing element in row 1 of Figure 10, along with their editing order and evaluation values, and row 18 of Figure 11 shows the four editing elements that were edited consecutively after the editing element in row 5 of Figure 10, along with their editing order and evaluation values.

[0110] [Trained Model Generation Process] In the trained model generation process, a first trained model is generated by performing training based on the training data in Figure 5, following steps S200 to S204. The first trained model is the same as the trained model in the first embodiment described above.

[0111] In the trained model generation process, following steps S200 to S204, a second trained model is generated by training based on the training data in Figure 11, similar to the generation of the first trained model. The second trained model is trained to maximize the evaluation value 420 based on the created or edited element ID 416, editing order information 418, and evaluation value 420.

[0112] Then, the process moves to step S206, where the first trained model and the second trained model generated in step S204 are stored in the storage device 42, and the series of processes is completed.

[0113] [Editing Order Information Estimation Process] Figure 12 is a flowchart showing the editing order information estimation process.

[0114] The editing order information estimation process is performed in response to requests from designers or other users. When executed on the CPU 30, it first proceeds to step S310, as shown in Figure 12.

[0115] In step S310, indicator information is obtained regarding the first indicator "editing time" or the second indicator "number of edited items," which indicate the value of the evaluation value. The process then proceeds to step S312, where the first trained model is selected if the indicator related to the obtained indicator information is the first indicator, and the second trained model is selected if the indicator related to the obtained indicator information is the second indicator.

[0116] Next, the process moves to step S314, where the IDs of created or edited elements are obtained from the CAD data currently being edited, and then the process moves to step S316.

[0117] In step S316, the edit order information is estimated from the created or edited element IDs obtained in step S314, using the first trained model and the second trained model of the storage device 42 that were selected in step S312 (hereinafter referred to as the "selected trained model"). The estimation method is the same as the process in step S302 in the first embodiment described above.

[0118] Next, the process moves to step S318, where the editing order information estimated by the selected trained model is displayed on the display device 44, and the series of processes is completed.

[0119] [When considering piano delivery] Next, we will explain the procedures when considering piano delivery.

[0120] If a designer wants to install a piano with a width of 120 mm in the cross-section drawing of Figure 9 using CAD software, they need to shorten the width of the toilet to prevent interference between the piano and the toilet wall when bringing the piano in. However, changing the width of the toilet necessitates editing other editing elements of the toilet. Therefore, the designer requests the estimation of the editing elements that need to be edited for the toilet after changing the width, and the order in which they should be edited. If the designer wants to obtain the editing elements and their editing order that will shorten the editing time, they select "editing time" as the indicator, and after steps S310 to S312, the first trained model is selected. Then, after steps S314 to S318, the created or edited element IDs are obtained from the CAD data currently being edited, and the editing order information is estimated and displayed from the obtained created or edited element IDs by the first trained model.

[0121] In contrast, if the designer wants to obtain editing elements with fewer editing items and their editing order, they select "number of editing items" as the indicator, and after steps S310 to S312, the second trained model is selected. Then, after steps S314 to S318, the IDs of created or edited elements are obtained from the CAD data currently being edited, and the second trained model estimates and displays the editing order information from the obtained created or edited element IDs.

[0122] [Effects of this embodiment] Next, the effects of this embodiment will be described. In this embodiment, index information relating to the first index or the second index is acquired, either the first trained model or the second trained model is selected based on the acquired index information, the created or edited element IDs are acquired, and the edit order information is estimated from the acquired created or edited element IDs using the selected trained model.

[0123] This allows us to obtain editing order information regarding multiple editing elements with high evaluation values ​​based on the first or second indicator, and their editing order.

[0124] [Third Embodiment] Next, a third embodiment of the present invention will be described. Figures 13 to 16 show this embodiment. Figures 3, 5 and 9 will also be referenced.

[0125] This embodiment differs from the first embodiment in that it utilizes a large language model. Below, only the differences from the first embodiment will be described, and the overlapping parts will be omitted.

[0126] [Configuration of this embodiment] First, the configuration of this embodiment will be described. Figure 13 is a block diagram showing the configuration of the network system according to this embodiment.

[0127] As shown in Figure 13, the Internet 199 is connected to a drawing creation support device 100 and a generation AI server 120 that generates response information using an AI (Artificial Intelligence) model in response to requests, enabling communication between the two.

[0128] [Generating AI Server 120] Next, the configuration of the generating AI server 120 will be described. Similar to the drawing creation support device 100, the generating AI server 120 has a hardware configuration similar to that of a general computer, with a CPU, ROM, RAM, and I / F connected by a bus, and is configured as, for example, a cloud server.

[0129] Figure 14 is a functional block diagram of the generation AI server 120. As shown in Figure 14, the generation AI server 120 is configured to include a plurality of AI models 50, an AI model control unit 52 that controls the AI ​​models 50, and a knowledge base 54 that registers data that the AI ​​models 50 refer to for inference.

[0130] AI Model 50 is an AI model trained on a large dataset and is a highly versatile model capable of performing various tasks. For example, AI Model 50 can employ a large-scale language model. A large-scale language model is a deep learning model that pre-trains a language model, which models human speech based on its probability of occurrence, on a vast amount of data. When a prompt is input, the large-scale language model statistically infers the probability of generating the next word from the sentence contained in the input prompt and outputs the inference result. For example, publicly known techniques described on the internet sites "https: / / chatgpt-lab.com / n / n418d3aa56f0b" and "https: / / agirobots.com / chatgpt-mechanism-and-problem / " can be used as large-scale language models. More specifically, for example, Titan Text G1 - Express, Titan Text G1 - Lite, Titan Image Generator G1, Titan Embeddings G1 - Text, Titan Embeddings Text V2, Titan Multimodal Embeddings G1, Claude, Claude Instant, Claude 3 Sonnet, Claude 3 Haiku, Claude 3 Opus, Jurassic-2 Mid, Jurassic-2 Ultra, Command, Command Light, Command R, Command R+, Embed English, Embed Multilingual, Llama 2 Chat 13B, Llama 2 Chat 70B, Llama 2 13B, Llama 2 70B, Llama 3 8b Instruct, Llama 3 70b Instruct, Mistral 7B Instruct, Mixtral 8X7B Instruct, Mistral Large, and Stable Diffusion XL can be adopted.

[0131] The AI ​​model control unit 52 selects one of the multiple AI models 50 to be used for inference in response to a selection request from the request processing unit 58. When a reference request is input from the request processing unit 58, the AI ​​model control unit 52 causes the selected AI model 50 (hereinafter referred to as the "selected AI model") to access data in the knowledge base 54 according to the input reference request. When a prompt is input from the request processing unit 58, the AI ​​model control unit 52 inputs the prompt to the selected AI model. Finally, when an execution request is input from the request processing unit 58, the AI ​​model control unit 52 causes the selected AI model to perform inference according to the input execution request, obtains the inference result from the selected AI model, and outputs the obtained inference result to the request processing unit 58.

[0132] The knowledge base 54 can register training data. The information registered in the knowledge base 54 is in a data format that the AI ​​model 50 can access (for example, vector data).

[0133] The generation AI server 120 is further configured to include a request receiving unit 56 that receives requests, a request processing unit 58 that processes the requests received by the request receiving unit 56, and a response information transmission unit 60 that transmits response information for the requests received by the request receiving unit 56 to the drawing creation support device 100.

[0134] The request receiving unit 56 receives a request from the drawing creation support device 100 for the generation of response information and outputs the received request to the request processing unit 58. The request includes (1) an element ID that has been created or edited, (2) a generation request to generate editing sequence information relating to a plurality of editing elements to be edited consecutively after a created or edited editing element and their editing order, (3) a selection request to select an AI model 50, and (4) a reference request to refer to the learning data of the knowledge base 54. (3) and (4) are not mandatory but are included as additional elements.

[0135] If the request received by the request receiving unit 56 includes a selection request or a reference request, the request processing unit 58 outputs the selection request or reference request to the AI ​​model control unit 52. It also generates a prompt to instruct the AI ​​model 50 based on the request received by the request receiving unit 56. For example, the prompt may request the AI ​​model 50 to generate editing sequence information relating to multiple editing elements to be edited consecutively after a created or edited editing element, based on the created or edited element ID, and the editing sequence information having an evaluation value of or greater than a predetermined value. The generated prompt and execution request are then output to the AI ​​model control unit 52, and if the AI ​​model control unit 52 inputs an inference result in response to the execution request, the input inference result is output to the response information transmission unit 60.

[0136] The response information transmission unit 60 transmits the response information, including the inference result input from the request processing unit 58, to the drawing creation support device 100.

[0137] The generation AI server 120 is further configured to include a request receiving unit 62 that receives requests and a learning data registration unit 64 that registers learning data in the knowledge base 54.

[0138] The request receiving unit 62 receives a request from the drawing creation support device 100 to register learning data, and outputs the received request to the learning data registration unit 64. The request includes (1) learning data.

[0139] The learning data registration unit 64 stores the learning data included in the request received by the request receiving unit 62 in storage (not shown) and converts it into a data format that the AI ​​model 50 can refer to (for example, vector data). Vector data can be generated by a technique (embedding) that converts data including characters, images, and sounds into numerical vectors. The converted learning data is then registered in the knowledge base 54. The AI ​​model control unit 52 makes the selected AI model refer to the learning data in response to a referencing request from the request processing unit 58.

[0140] [Operation of this embodiment] Next, the operation of this embodiment will be described. [Learning data registration process] Figure 15 is a flowchart of the learning data registration process.

[0141] The learning data registration process is performed in response to requests from designers and other users. When it is executed on the CPU 30, it first proceeds to step S400, as shown in Figure 15.

[0142] In step S400, the learning data shown in Figure 5 is acquired from the storage device 42, and the process proceeds to step S402.

[0143] In step S402, a request is sent to the generating AI server 120 requesting the registration of training data. The request includes (1) the training data acquired in step S400.

[0144] When the process in step S402 is completed, the series of processes ends. [Editing Order Information Acquisition Process] Figure 16 is a flowchart of the editing order information acquisition process.

[0145] The editing order information acquisition process is performed in response to requests from designers or other users. When executed on the CPU 30, it first proceeds to step S500, as shown in Figure 16.

[0146] In step S500, the element IDs of created or edited elements are obtained from the CAD data currently being edited, and the process proceeds to step S502.

[0147] In step S502, a request is sent to the generation AI server 120 requesting the generation of answer information. The request includes (1) the created or edited element ID obtained in step S500, (2) a generation request to generate editing sequence information relating to a plurality of editing elements to be edited consecutively after the created or edited editing element and their editing order, (3) a selection request to select a predetermined AI model 50, and (4) a reference request to refer to the learning data in the knowledge base 54.

[0148] Here, a generation request can be included that generates editing order information regarding a larger number of editing elements and their editing order with higher accuracy. In the example in Figure 3, priority is given to generating editing order information regarding 3 editing elements and their editing order rather than 2, and 4 editing elements rather than 3. This allows the designer to perform editing while visualizing larger editing units.

[0149] Furthermore, instead of prioritizing the longest possible sequence, the generation request can include generating editing sequence information regarding related groups of editing elements and their editing order. For example, if the AI ​​model 50 can infer that A, B, C, D, E, F, G, and H are the editing elements to be edited consecutively from the current editing state, and A-D and E-H are related groups, the request can be made to generate editing sequence information for A-D rather than editing sequence information for A-H. The determination of whether or not a group is related can be made, for example, by identifying two or more editing elements and their editing order from A-H that appear more than a predetermined number of times in the training data. This allows the designer to perform editing while visualizing related groups.

[0150] Next, the process moves to step S504, where response information is received from the generation AI server 120. Then, the process moves to step S506, where the editing order information included in the received response information is displayed on the display device 44, and the series of processes ends.

[0151] [When considering piano delivery] Next, we will explain the procedures when considering piano delivery.

[0152] If a designer wants to install a piano with a width of 120 mm in the cross-section drawing of Figure 9 using CAD software, they need to shorten the width of the toilet to prevent interference between the piano and the toilet wall when bringing the piano in. However, changing the width of the toilet necessitates editing other editing elements of the toilet. Therefore, when the designer requests the inference of the editing elements to be edited for the toilet and their editing order after changing the width of the toilet, the system proceeds through steps S500 to S506, obtaining created or edited element IDs from the currently edited CAD data, and then inferring and displaying editing order information from the obtained created or edited element IDs. In the inference process, the selected AI model references the learning data of the knowledge base 54.

[0153] [Effects of this embodiment] Next, the effects of this embodiment will be described. In this embodiment, a request is input to the AI ​​model 50 that includes a request to generate editing order information relating to a plurality of editing elements that have been edited consecutively after the created or edited element ID and the editing order thereof, and the information output from the AI ​​model 50 in response to the request is obtained.

[0154] This allows you to obtain editing order information regarding multiple editing elements that are edited consecutively after an already created or edited editing element, and the order in which they are edited. Therefore, you can understand the sequence of editing of multiple editing elements that are edited consecutively and their editing order.

[0155] Furthermore, in this embodiment, learning data including created or edited element IDs, a plurality of edited elements that are edited consecutively after that edited element, editing order information relating to the editing order, and evaluation values ​​relating to the editing of the plurality of edited elements is registered in the knowledge base 54, and in response to a request, information output from the AI ​​model 50 is obtained by referring to the learning data in the knowledge base 54.

[0156] This allows us to obtain editing order information regarding multiple editing elements with high evaluation values ​​and their editing order.

[0157] Furthermore, in this embodiment, a request is input to the AI ​​model 50 that includes a request to identify multiple editing elements and their editing order that appear more than a predetermined number of times as related groups, and to generate editing order information relating to the multiple editing elements and their editing order as identified related groups.

[0158] This allows editing to be performed while visualizing related groups. In this embodiment, step S402 corresponds to the registration means of invention 2 or 3, step S502 corresponds to the input means of invention 1 or the identification means of invention 8, step S504 corresponds to the acquisition means of invention 1 or 2, and the CAD data corresponds to the design information of invention 8 or 10. Furthermore, the created or edited element IDs correspond to the element information of inventions 1 to 3, and the learning data corresponds to the reference information of invention 2 or 3.

[0159] [Fourth Embodiment] Next, a fourth embodiment of the present invention will be described. Figure 17 is a diagram showing this embodiment. Figures 5, 9, 11 and 15 will also be referenced.

[0160] This embodiment differs from the second and third embodiments in that it causes the AI ​​model 50 to refer to first training data including evaluation values ​​based on a first indicator or second training data including evaluation values ​​based on a second indicator. Below, only the parts that differ from the second and third embodiments will be described, and the overlapping parts will be omitted from the explanation.

[0161] [Operation of this Embodiment] Next, the operation of this embodiment will be described. [Learning Data Registration Process] The learning data registration process is a process that is executed in response to a request from a designer or other user. When it is executed on the CPU 30, it first proceeds to step S400, as shown in Figure 15.

[0162] In step S400, the learning data shown in Figure 5 and Figure 11 are acquired from the storage device 42, and the process proceeds to step S402.

[0163] In step S402, a request is sent to the generating AI server 120 requesting the registration of training data. The request includes (1) the training data shown in Figure 5 acquired in step S400 as the first training data, and (2) the training data shown in Figure 11 acquired in step S400 as the second training data.

[0164] When the process in step S402 is completed, the series of processes ends. [Editing order information acquisition process] Figure 17 is a flowchart of the editing order information acquisition process.

[0165] The editing order information acquisition process is performed in response to requests from designers or other users. When executed on the CPU 30, it first proceeds to step S510, as shown in Figure 17.

[0166] In step S510, indicator information related to the first indicator "editing time" or the second indicator "number of edited items," which indicate the value of the evaluation value, is obtained. The process then moves to step S512, where the IDs of created or edited elements are obtained from the CAD data currently being edited, and the process moves to step S514.

[0167] In step S514, a request is sent to the generation AI server 120 requesting the generation of response information. The request includes (1) the created or edited element ID obtained in step S512, (2) a generation request to generate editing sequence information relating to a plurality of editing elements to be edited consecutively after the created or edited editing element and their editing order, (3) a selection request to select a predetermined AI model 50, and (4) a reference request to refer to the first learning data of the knowledge base 54 if the index related to the index information obtained in step S510 is the first index, or a reference request to refer to the second learning data of the knowledge base 54 if the index related to the index information obtained in step S510 is the second index.

[0168] Next, the process moves to step S516, where response information is received from the generation AI server 120, and then to step S518, where the editing order information included in the received response information is displayed on the display device 44, thus ending the series of processes.

[0169] [When considering piano delivery] Next, we will explain the procedures when considering piano delivery.

[0170] If a designer wants to install a piano with a width of 120 mm in the cross-section drawing of Figure 9 using CAD software, they need to shorten the width of the toilet to prevent interference between the piano and the toilet wall when bringing the piano in. However, changing the width of the toilet necessitates editing other editing elements of the toilet. Therefore, the designer requests an estimation of the editing elements that need to be edited for the toilet after changing the width, and the order in which they should be edited. If the designer wants to obtain the editing elements and their editing order that will shorten the editing time, they select "editing time" as the indicator. Then, after steps S510 to S518, the created or edited element IDs are obtained from the CAD data currently being edited, and the editing order information is inferred and displayed from the obtained created or edited element IDs. In the inference, the first training data of the knowledge base 54 is referenced by the selected AI model.

[0171] In contrast, if the designer wants to obtain editing elements with fewer editing items and their editing order, they can select "number of editing items" as the indicator. This will trigger steps S510 to S518, during which the IDs of created or edited elements will be obtained from the CAD data currently being edited. The editing order information will then be inferred and displayed from the obtained IDs of created or edited elements. In the inference process, the selected AI model will refer to the second training data in the knowledge base 54.

[0172] [Effects of this embodiment] Next, the effects of this embodiment will be described. In this embodiment, first learning data including evaluation values ​​based on the first indicator and second learning data including evaluation values ​​based on the second indicator are registered in the knowledge base 54, indicator information related to the first indicator or the second indicator is obtained, and a request is input to the AI ​​model 50 that includes a request to refer to either the first learning data or the second learning data in the knowledge base 54 based on the obtained indicator information.

[0173] This allows us to obtain editing order information regarding multiple editing elements with high evaluation values ​​based on the first or second indicator, and their editing order.

[0174] In this embodiment, step S510 corresponds to the index information acquisition means of Invention 4, step S514 corresponds to the input means of Invention 1 or 4, or the identification means of Invention 8, step S516 corresponds to the acquisition means of Invention 1 or 2, and the first learning data corresponds to the first reference information of Invention 4. The second learning data corresponds to the second reference information of Invention 4.

[0175] [Modifications] In the first to fourth embodiments and their modifications described above, reinforcement learning was adopted as the learning method, but the method is not limited to this, and supervised learning, semi-supervised learning, unsupervised learning, deep learning, or any other learning method can be adopted.

[0176] Furthermore, in the first to fourth embodiments and their variations described above, the training data is composed of created or edited element IDs, editing sequence information relating to a plurality of editing elements edited consecutively after the edited element and their editing order, and evaluation values ​​relating to the editing of the plurality of editing elements. However, it is not limited to this, and can be composed without including evaluation values. In this case, the training data is composed of created or edited element IDs, and editing sequence information relating to a plurality of editing elements edited consecutively after the edited element and their editing order.

[0177] Furthermore, in the first and second embodiments and their variations described above, the trained model was one that had been trained based on created or edited element IDs, editing order information, and evaluation values. However, it is not limited to this, and a model trained based on created or edited element IDs and editing order information can also be used.

[0178] Furthermore, in the second embodiment and its modified form, the first trained model and the second trained model can be configured as a single trained model.

[0179] Furthermore, in the third and fourth embodiments and their variations, the AI ​​model 50 can be configured as the trained model in the first and second embodiments and their variations.

[0180] Furthermore, in the first to fourth embodiments and their variations described above, multiple editing elements and their editing order were identified if their occurrence count in the training data was greater than or equal to a predetermined number. However, the invention is not limited to this, and it is also possible to identify elements whose occurrence count in the editing history data is greater than or equal to a predetermined number.

[0181] Furthermore, while the first to fourth embodiments and their variations described above handle a maximum of four editing elements and their editing order, the system is not limited to this, and can handle five or more editing elements and their editing order. There is no need to limit the number of editing elements; any number of editing elements is sufficient.

[0182] Furthermore, in the first to fourth embodiments and their variations described above, training data was generated by obtaining all combinations of 2 to 4 editing elements from the editing history data for multiple editing elements and their editing order. However, the model is not limited to this, and training data can be generated by obtaining from the editing history data only those editing elements and their editing order that appear more than a predetermined number of times in the editing history data. In this case, a trained model that has been trained on multiple editing history data may be used to estimate multiple editing elements and their editing order that appear frequently. This makes it possible to learn or infer multiple editing elements and their editing order that appear frequently in the editing history data.

[0183] Furthermore, in the first to fourth embodiments and their modifications described above, the editing history data is configured as separate data from the CAD data, but it is not limited to this and can be included in the CAD data and configured as an integrated entity.

[0184] Furthermore, in the third and fourth embodiments and their modifications described above, the AI ​​model 50 was made to refer to the learning data in the knowledge base 54. However, the AI ​​model 50 is not limited to this, and the learning data to be referenced in the knowledge base 54 can be included in the request. This allows the AI ​​model to be applied even in configurations that do not have a knowledge base 54. Specifically, for example, the following configuration can be adopted.

[0185] [Invention A1] An input means for inputting a request to the AI ​​model that includes element information relating to an element that has been created or edited in the design information, and a request for generating editing sequence information relating to a plurality of elements that will be edited consecutively after the element and the order in which they will be edited; and an acquisition means for acquiring information output from the AI ​​model in response to the request, wherein the request includes reference information relating to an element that has been created or edited, and editing sequence information relating to a plurality of elements that will be edited consecutively after the element and the order in which they will be edited.

[0186] This allows you to obtain editing order information regarding multiple elements that will be edited consecutively after an already created or edited element, and the order in which they will be edited. Therefore, you can understand the sequence of multiple elements that will be edited consecutively and the order in which they will be edited.

[0187] [Invention A2] In Invention A1, the request includes element information relating to an element that has been created or edited, editing sequence information relating to a plurality of elements to be edited consecutively after the element and the order in which they are edited, and reference information including evaluation values ​​relating to the editing of the plurality of elements.

[0188] This allows us to obtain editing order information regarding multiple elements with high evaluation values ​​and their editing order.

[0189] Embodiments of Inventions A1 and A2 will be described as variations of the third embodiment described above. In step S502, a request for the generation of response information is sent to the generation AI server 120. The request includes (1) an element ID that has been created or edited obtained in step S500, (2) a generation request to generate editing order information relating to a plurality of editing elements that will be edited consecutively after the created or edited editing element and the editing order thereof, (3) a selection request to select a predetermined AI model 50, and (4) learning data obtained in step S400.

[0190] [Invention A3] Invention A2 is further provided with an indicator information acquisition means for acquiring indicator information relating to a first indicator that shows an indicator of the value of the evaluation value or a second indicator different from the first indicator, wherein the input means inputs the request to the AI ​​model, which includes either first reference information including the element information, the editing sequence information and the evaluation value based on the first indicator, or second reference information including the element information, the editing sequence information and the evaluation value based on the second indicator, based on the indicator information acquired by the indicator information acquisition means.

[0191] This allows us to obtain editing order information regarding multiple elements with high evaluation values ​​based on the first or second indicator, and their editing order.

[0192] An embodiment of Invention A3 will be described as a modification of the fourth embodiment described above. In step S514, a request for the generation of response information is sent to the generation AI server 120. The request includes (1) an element ID that has been created or edited obtained in step S512, (2) a generation request to generate editing sequence information relating to a plurality of editing elements that will be edited consecutively after the created or edited editing element and the editing order thereof, (3) a selection request to select a predetermined AI model 50, and (4) first learning data obtained in step S400 if the index related to the index information obtained in step S510 is the first index, or second learning data obtained in step S400 if the index related to the index information obtained in step S510 is the second index.

[0193] [Invention A4] Inventions A1 to A3 are further provided with a means for identifying related groups of elements and their editing order that appear more than or equal to a predetermined number of times in the design information, wherein the request includes a request to generate the editing order information relating to the multiple elements and their editing order as related groups identified by the identification means.

[0194] [Invention A5] In Inventions A1 to A3, the plurality of elements are elements whose setting or modification requires human judgment, and whose setting or modification affects other elements.

[0195] [Invention A6] In Inventions A1 to A3, the design information is design information for designing buildings.

[0196] Furthermore, in the third and fourth embodiments and their modifications described above, the AI ​​model 50 was made to refer to information in the knowledge base 54. However, the AI ​​model 50 is not limited to this, and information to be referenced in the knowledge base 54 (for example, the training data in Figure 5 or Figure 11) can be obtained by web search or the like, and the AI ​​model 50 can be made to refer to the search results and perform inference. This configuration can be realized, for example, by RAG (Retrieval Augmented Generation).

[0197] Furthermore, while the first to fourth embodiments and their variations described above used a trained model or AI model 50, the system is not limited to these, and for example, the following configuration can be adopted.

[0198] [Invention B1] The invention comprises an element information acquisition means for acquiring element information relating to created or edited elements in design information, and a search means for retrieving the editing sequence information corresponding to the element information acquired by the element information acquisition means from a storage means that stores editing sequence information relating to a plurality of elements to be edited consecutively after a created or edited element and their editing order, in association with the element information relating to the created or edited element.

[0199] This allows you to obtain editing order information regarding multiple elements that will be edited consecutively after an already created or edited element, and the order in which they will be edited. Therefore, you can understand the sequence of multiple elements that will be edited consecutively and the order in which they will be edited.

[0200] [Invention B2] In Invention B1, the storage means stores editing sequence information relating to a plurality of elements to be edited consecutively after a created or edited element and the editing order thereof, in association with element information relating to the created or edited element and evaluation values ​​relating to the editing of the plurality of elements, and the search means searches for the editing sequence information corresponding to the element information acquired by the element information acquisition means, wherein the evaluation value is equal to or greater than a predetermined value.

[0201] This allows us to obtain editing order information regarding multiple elements with high evaluation values ​​and their editing order.

[0202] Embodiments of Inventions B1 and B2 will be described as modifications of the first embodiment described above. The storage device 42 stores an editing sequence information table having a data structure similar to the learning data in Figure 5. In step S302, editing sequence information corresponding to the created or edited element IDs obtained in step S300, with an evaluation value of a predetermined value or higher, is searched from the editing sequence information table.

[0203] [Invention B3] In Invention B2, the storage means includes a first storage means for storing the editing sequence information in association with the element information and the evaluation value based on a first index indicating an index of the value of the evaluation value, and a second storage means for storing the editing sequence information in association with the element information and the evaluation value based on a second index different from the first index, and comprises an index information acquisition means for acquiring index information relating to the first index or the second index, and a storage means selection means for selecting either the first storage means or the second storage means based on the index information acquired by the index information acquisition means, and the search means searches for the editing sequence information from the storage means selected by the storage means selection means.

[0204] This allows us to obtain editing order information regarding multiple elements with high evaluation values ​​based on the first or second indicator, and their editing order.

[0205] An embodiment of Invention B3 will be described as a modification of the second embodiment described above. The storage device 42 stores a first editing sequence information table having a data structure similar to the learning data in Figure 5, and a second editing sequence information table having a data structure similar to the learning data in Figure 11. In step S312, if the index related to the index information acquired in step S310 is the first index, the first editing sequence information table is selected, and if the index related to the acquired index information is the second index, the second editing sequence information table is selected. In step S316, editing sequence information corresponding to the created or edited element ID acquired in step S314, with an evaluation value of a predetermined value or higher, is searched from the editing sequence information table selected in step S312.

[0206] [Invention B4] In Invention B2, the storage means stores the element information, the editing sequence information, the evaluation value based on a first indicator that indicates an indicator of the value of the evaluation value, and indicator information relating to the first indicator in association with each other, and also stores the element information, the editing sequence information, the evaluation value based on a second indicator different from the first indicator, and indicator information relating to the second indicator in association with each other, and includes indicator information acquisition means for acquiring indicator information relating to the first indicator or the second indicator, and the search means searches for the editing sequence information corresponding to the element information acquired by the element information acquisition means and the indicator information acquired by the indicator information acquisition means, where the evaluation value is equal to or greater than a predetermined value.

[0207] This allows us to obtain editing order information regarding multiple elements with high evaluation values ​​based on the first or second indicator, and their editing order.

[0208] An embodiment of Invention B4 will be described as a modification of the second embodiment described above. The storage device 42 stores an editing sequence information table in which, for each row, (1) created or edited element ID 410, editing sequence information 412, evaluation value 414, and index information relating to the first index, or (2) created or edited element ID 416, editing sequence information 418, evaluation value 420, and index information relating to the second index are registered. In step S316, editing sequence information corresponding to the created or edited element ID obtained in step S314 and the index information obtained in step S310, with an evaluation value of a predetermined value or higher, is searched from the editing sequence information table.

[0209] [Invention B5] Inventions B1 to B4 are provided with a selection means for identifying related groups of elements and their editing order that appear more than a predetermined number of times in the design information, and the search means searches the editing order information relating to the multiple elements and their editing order as related groups identified by the selection means.

[0210] This allows editing to be performed while visualizing related groups. [Invention B6] In Inventions B1 to B4, the plurality of elements are elements whose setting or modification requires human judgment, and whose setting or modification affects other elements.

[0211] [Invention B7] In Inventions B1 to B4, the design information is design information for designing buildings.

[0212] Furthermore, in inventions B1 to B7 and their modifications, storing editing sequence information in association with element information, etc., includes, for example, (1) storing editing sequence information and element information, etc., in the same record, or directly associating them, and (2) storing them through one or more pieces of information in between, such as providing a table for storing editing sequence information and intermediate information in association, and a table for storing element information, etc., and intermediate information in association. In other words, any data structure can be adopted as long as it is possible to trace the editing sequence information from the element information, etc. Note that the editing sequence information only needs to be stored in the storage means in association with the element information, etc., and it is not necessarily required that the element information, etc. be stored in the storage means.

[0213] Furthermore, in inventions B1 to B7 and their modifications, the storage means stores editing sequence information by any means and at any time. The editing sequence information may be stored in advance, or it may not be stored in advance, but rather stored by external input or the like when the drawing creation support device 100 is in operation.

[0214] Furthermore, in the third and fourth embodiments and their variations described above, the prompt is set to request the AI ​​model 50 to generate editing sequence information relating to a plurality of editing elements to be edited consecutively after a created or edited editing element, and the editing order thereof, based on the created or edited element ID, provided that the evaluation value is above a predetermined value. However, the prompt is not limited to this, and the prompt can be set to request the AI ​​model 50 to generate editing sequence information relating to a plurality of editing elements to be edited consecutively after a created or edited editing element, and the editing order thereof.

[0215] Furthermore, in the third and fourth embodiments and their modifications described above, the training data shown in Figure 5 or Figure 11 was registered in the knowledge base 54. However, the knowledge base 54 can also be used to register (1) editing history data, (2) reference information including element information relating to created or edited elements, and editing order information relating to a plurality of elements to be edited consecutively after the element and their editing order, or (3) reference information including element information relating to created or edited elements, editing order information relating to a plurality of elements to be edited consecutively after the element and their editing order, and evaluation values ​​relating to the editing of the plurality of elements. The reference information in (2) or (3) may include editing order information estimated using the trained model in the first and second embodiments and their modifications described above.

[0216] Furthermore, in the third and fourth embodiments and their variations described above, vector data was registered in the knowledge base 54, but the invention is not limited to this, and data in any format can be registered.

[0217] Furthermore, in the first to fourth embodiments and their modifications described above, a configuration was adopted that includes one of the following: estimation processing using a trained model, acquisition processing from the AI ​​model 50, and search processing using a table. However, the system is not limited to these configurations, and a configuration including multiple of these processes can be adopted. In this case, it is possible to control which process is prioritized. For example, a configuration can be adopted that prioritizes the process with the lowest current load among multiple processes, a configuration that prioritizes the process whose processing result has been adopted by the designer to a high degree (meaning the number of times, percentage, or other degree of adoption), or a configuration that prioritizes the process which is used by the designer to a high degree (meaning the number of times, percentage, or other degree of use).

[0218] Furthermore, in the first to fourth embodiments and their variations described above, multiple editing elements and their editing order were learned or inferred. However, the "multiple editing elements" to be learned or inferred can be editing elements A whose setting or modification requires human judgment, and whose setting or modification affects other editing elements B. This allows for obtaining editing order information regarding multiple editing elements and their editing order, taking into account the relationship between editing elements A and B. For editing elements that do not require human judgment, editing can be automated using, for example, the technology described in Japanese Patent Publication No. 7341580. Therefore, by targeting editing elements that are difficult to automate, the editing work can be made more efficient.

[0219] Furthermore, while the first and second embodiments and their modifications described above are implemented as a single device, the system 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.

[0220] Furthermore, in the third and fourth embodiments and their modifications described above, the generating AI server 120 integrates the functions of the AI ​​model 50, AI model control unit 52, knowledge base 54, request receiving unit 56, request processing unit 58, response information transmission unit 60, request receiving unit 62, and learning data registration unit 64. However, it is not limited to this configuration, and some functions can be configured on separate servers or the like.

[0221] Furthermore, while the third and fourth embodiments and their modifications described above were implemented as a network system, the system is not limited to this and can also be implemented as a single device or application.

[0222] Furthermore, while the third and fourth embodiments and their modifications described above describe the case where the invention is applied to a network system consisting of the Internet 199, the invention is not limited to this, and may also be applied to a so-called intranet that communicates using the same method as the Internet 199. Of course, it is not limited to networks that communicate using the same method as the Internet 199, but can be applied to any network using any communication method.

[0223] Furthermore, in the first to fourth embodiments and their modifications described above, the drawing creation support device 100 is configured to utilize the storage device 42, but it is not limited to this, and can also be configured to utilize an external storage device such as a database server.

[0224] Furthermore, in the first to fourth embodiments and their modifications described above, the process shown in the flowcharts of Figures 4, 6, 8, 12, 15, 16, and 17 was described in the case of executing a program pre-stored in ROM 32. However, the invention is not limited to this, and the program showing these procedures may be read into RAM 34 from a storage medium in which the program is stored and then executed.

[0225] Here, "storage medium" refers to any storage medium that can be read by a computer, regardless of whether it is an electronic, magnetic, or optical reading medium, including semiconductor storage mediums such as RAM and ROM, magnetic storage mediums such as FD and HD, optical reading mediums such as CD, CDV, LD, and DVD, and magnetic / optical reading mediums such as MO.

[0226] Furthermore, the first to fourth embodiments and their variations described above are mutually applicable. Moreover, the present invention is applicable to other cases without departing from the spirit of the invention, not limited to the first to fourth embodiments and their variations. For example, the present invention can be applied to a wide range of design work, such as automobile design, mechanical design, and circuit design.

[0227] 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, 120... Generation AI server, 50... AI model, 52... AI model control unit, 54... Knowledge base, 56... Request receiving unit, 58... Request processing unit, 60... Answer information transmission unit, 62... Request receiving unit, 64... Learning data registration unit, 199... Internet, 400, 404... Element information, 402... Editing time, 406... Number of editing items, 410, 416... Element ID, 412, 418... Editing order information, 414, 420... Evaluation value

Claims

1. A design support system comprising: an input means for inputting a request to an AI model that includes element information relating to created or edited elements in design information, and a request for generating editing sequence information relating to a plurality of elements to be edited consecutively after said element and the editing order thereof; and an acquisition means for acquiring information output from the AI ​​model in response to said request.

2. The design support system according to claim 1, comprising registration means for registering reference information, including element information relating to created or edited elements, and editing order information relating to a plurality of elements to be edited consecutively after said element and the editing order thereof, in a knowledge base that can be referenced by the AI ​​model, wherein the acquisition means acquires information output from the AI ​​model by referring to the reference information in the knowledge base in response to the request.

3. The design support system according to claim 2, wherein the registration means registers in the knowledge base element information relating to an element that has been created or edited, editing sequence information relating to a plurality of elements to be edited consecutively after the element and the editing order thereof, and reference information including evaluation values ​​relating to the editing of the plurality of elements.

4. The design support system according to claim 3, wherein the registration means registers in the knowledge base first reference information including the element information, the editing sequence information and the evaluation value based on a first index indicating an index of the value of the evaluation value, and second reference information including the element information, the editing sequence information and the evaluation value based on a second index different from the first index; the system comprises index information acquisition means for acquiring index information relating to the first index or the second index; and the input means inputs into the AI ​​model a request including a request to refer to either the first reference information or the second reference information in the knowledge base based on the index information acquisition means.

5. The design support system according to claim 1, wherein the request includes element information relating to an element that has been created or edited, and reference information relating to a plurality of elements to be edited consecutively after the element and the order in which they are edited.

6. The design support system according to claim 5, wherein the request includes element information relating to an element that has been created or edited, editing sequence information relating to a plurality of elements to be edited consecutively after the element and the editing order thereof, and reference information including evaluation values ​​relating to the editing of the plurality of elements.

7. The design support system according to claim 6, comprising indicator information acquisition means for acquiring indicator information relating to a first indicator that indicates an indicator of the value of the evaluation value or a second indicator different from the first indicator, wherein the input means inputs the request to the AI ​​model, based on the indicator information acquired by the indicator information acquisition means, which includes either first reference information including the element information, the editing sequence information and the evaluation value based on the first indicator, or second reference information including the element information, the editing sequence information and the evaluation value based on the second indicator.

8. A design support system according to any one of claims 1 to 7, comprising a means for identifying related groups of elements and their editing order that appear more than or equal to a predetermined number of times in the design information, wherein the request includes a request to generate editing order information relating to the plurality of elements and their editing order as related groups identified by the identifying means.

9. A design support system according to any one of claims 1 to 7, wherein the plurality of elements are elements whose setting or modification requires human judgment and whose setting or modification affects other elements.

10. A design support system according to any one of claims 1 to 7, characterized in that the design information is design information for designing buildings.

Images