Work Support System

Abstract

To provide a work support system that helps facilitate the use of BIM, which is beneficial in the field of architectural technology. [Solution] The work support system 1 includes a knowledge model 10 and a processing device 3 that outputs output factors corresponding to input factors by searching the knowledge model 10. The work support system 1 is configured such that the knowledge model 10 is constructed using factors that are related to the field of construction technology and used in BIM, and legal information that represents laws and regulations related to the field of construction technology, and the processing device 3 inputs request information requesting support from the knowledge model 10 regarding the construction-related information and legal information into the knowledge model 10 as input factors, and presents output factors corresponding to the input factors to a worker using an information processing terminal 6.

Description

[Technical Field]

[0001] The present invention relates to a work support system. [Background technology]

[0002] A design support system disclosed in Patent Document 1, for example, has been known for some time. The conventional design support system is a system that supports design in BIM (Building Information Modeling). A design device constituting the conventional design support system supports design by using templates that hold various pieces of architectural information as attributes for three-dimensional models of floors, balconies, etc., and arranges the three-dimensional models in a virtual space. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-174427 Summary of the Invention [Problem to be solved by the invention]

[0004] Sharing information using BIM is extremely beneficial for those working in the field of building technology, including architectural design, construction, and maintenance, as it ensures the accuracy of the necessary information, which differs for each construction process, and the speed with which information can be obtained. However, in order to use BIM in this way, which is beneficial to those working in the field, there are times when specialized skills are required. This has been an obstacle to the widespread adoption of BIM, and it is essential and desired to provide support so that even those working in the field who lack specialized skills can easily use BIM.

[0005] An object of the present invention is to provide a work support system that helps facilitate the use of BIM, which is beneficial in the field of architectural technology. [Means for solving the problem]

[0006] The work support system of the present invention comprises a knowledge model including one or more reference factors from among a plurality of factors, connection factors that connect to the reference factors from among the factors, and connection conditions for connecting the reference factors and the connection factors, and a computer device that inputs input factors that are reference factors to the knowledge model and outputs output factors that are connection factors that correspond to the input factors by searching the knowledge model using the connection conditions, wherein the knowledge model is related to at least the field of architectural technology including architectural education, architectural design, architectural construction, and architectural maintenance, and is constructed using factors that include architectural-related information used in BIM (Building Information Modeling) and legal information that represents laws and regulations related to the field of architectural technology, and the computer device The correlation between multiple factors that construct the knowledge model is determined by a trained model generated by machine learning, and the trained model is used as a connection condition. inputting request information requesting support from the knowledge model for construction-related information and legal information as input factors into the knowledge model in response to an instruction from a worker engaged in the field of construction technology; Searching the knowledge model using connection conditions The system is configured to present to the practitioner output factors corresponding to the input factors. [Effects of the Invention]

[0007] According to the work support system of the present invention, even if a worker in the field of architectural technology does not have sufficient specialized skills to use BIM, the knowledge model can be used to provide appropriate support according to the worker's requests. As a result, the worker can obtain useful information from BIM while receiving support and proceed with the work. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating a schematic configuration of a work support system. [Figure 2] FIG. 2 is a diagram for explaining the configuration of a knowledge model. [Figure 3] FIG. 1 is a diagram for explaining the interconnection of factors in a knowledge model. [Figure 4] FIG. 10 is a diagram for explaining information linking in a knowledge model. [Figure 5] 5 is a diagram for explaining the association of information relating to the connection conditions in FIG. 4. FIG. [Figure 6] 10 is a data table showing an example of factors. [Figure 7] FIG. 10 is a data table showing an example of connection conditions, illustrating rank / rank connection. [Figure 8] FIG. 10 is a data table showing an example of connection conditions, illustrating rank / range connections. [Figure 9] FIG. 2 is a block diagram showing the configuration of an arithmetic and control device. [Figure 10] FIG. 2 is a block diagram showing a configuration of an information processing terminal. [Figure 11] FIG. 2 is a functional block diagram showing functions of the information processing terminal. [Figure 12] FIG. 1 is a diagram for explaining work support by a work support system. DETAILED DESCRIPTION OF THE INVENTION

[0009] 1. Overview of the work support system An overview of a work support system according to an embodiment will be described below with reference to FIG. 1. The work support system 1 according to this embodiment is applied to the field of architectural technology, which involves designing and constructing buildings. The field of architectural technology includes architectural education, architectural design, architectural construction, architectural maintenance, and the like. The work support system 1 according to this embodiment is a system that utilizes a knowledge model 10 that describes knowledge about architectural-related information, such as the techniques, know-how, insights, and experience accumulated by practitioners in the field of architectural technology, including architectural engineers and architectural construction engineers, as well as legal information that represents laws and regulations related to architectural design and construction.

[0010] 1, the work support system 1 includes a storage device 2 and an arithmetic processing device 3. In the work support system 1, the storage device 2, the arithmetic processing device 3 which is a computer device, a plurality of information processing terminals 6, and a knowledge model 10 stored in the storage device 2 are connected via a network N which includes the Internet or the like. In other words, in the work support system 1, the storage device 2 (knowledge model 10), the arithmetic processing device 3, and a plurality of information processing terminals 6 which are external devices are configured to be able to communicate with each other.

[0011] The storage device 2 stores (contains) a knowledge model 10, which will be described in detail later, and various information related to the knowledge model 10. The arithmetic processing device 3 constructs the knowledge model 10 and performs arithmetic processing using the knowledge model 10. For this purpose, an input device 4 and a display device 5 are connected to the arithmetic processing device 3. Examples of the multiple information processing terminals 6 include computer devices that execute BIM tools, which will be described later, used for architectural design, construction, and maintenance, and computer devices that perform production management and inventory management of building materials, etc.

[0012] In the work support system 1 of this embodiment, the arithmetic processing device 3 executes various arithmetic processing using the knowledge model 10 stored in the storage device 2, thereby providing architectural design support, architectural construction support, and architectural maintenance support using the information processing terminal 6. The knowledge model 10 used in the work support system 1 will be described below. Note that in the work support system 1, the information processing terminal 6 can also directly execute various information processes using the knowledge model 10.

[0013] 2. Details of the structure of Knowledge Model 10 The knowledge model 10 describes and stores knowledge related to the above-mentioned building-related information in the field of building technology in a predetermined format. That is, as shown in Figures 2 and 3, the knowledge model 10 uses building-related information expressed as characters, figures, images, videos, etc. as factors 11, and expresses the relationships between the factors 11 by interconnecting the factors 11 in a network form, and the relationships between the factors 11 are described in a predetermined format.

[0014] As shown in Figure 2, the knowledge model 10 is constructed by using factors 11, such as various data (text data, image data, video data) including characters, figures, images, videos, etc. that represent technical terms, technical definitions, knowledge, experience, and know-how based on knowledge and experience, which are keywords for architecture-related information, and by interconnecting the factors 11 in a network configuration. In this way, the knowledge model 10 represents the relationships between architecture-related information, thereby storing architecture-related knowledge (including tacit knowledge) as explicit knowledge. The architecture-related information, i.e., knowledge, stored in the knowledge model 10 is expressed, for example, in a form that can be easily interpreted systematically within a general-purpose framework.

[0015] Here, the configuration of the knowledge model 10 will be explained. In the knowledge model 10, as specifically shown in FIG. 3, the factor 11 to be processed among the multiple factors 11 is called a reference factor 111. Each factor 11, including the reference factor 111, is linked to factor information 12, which is its own internal information. A factor 11 directly connected to the reference factor 111 is called a connection factor 112. Connection information 13 is linked between each factor 11 (for example, between the reference factor 111 and the connection factor 112). The connection information 13 indicates the connection relationship between two interconnected factors 11 (the reference factor 111 and the connection factor 112). The connection information 13 is linked to a connection condition 14.

[0016] In particular, a factor 11 that has a hierarchical relationship with a reference factor 111 is called a higher-level linking factor 113 (or a lower-level linking factor 113) in the knowledge model 10, as shown in Fig. 3. Linking information 15 indicating the hierarchical relationship between the reference factor 111 and the linking factor 113 is linked to the reference factor 111 and the linking factor 113. Furthermore, linking conditions 16 are linked to the linking factor 113.

[0017] In the knowledge model 10, as shown by the dashed line in Fig. 2, the basic pattern is that two adjacent factors 11 (for example, a reference factor 111 and a connection factor 112) are interconnected by connection information 13. In the knowledge model 10, as shown in Fig. 2, a network form of factors 11 is formed by repeating the basic pattern for combinations required to express a certain technical field in the field of architectural technology. Therefore, in the knowledge model 10, architectural-related information, which is knowledge, is made explicit knowledge by representing multiple factors 11 in a network form.

[0018] Next, the content of each piece of information in the knowledge model 10 and the linkage of the information will be explained using Figure 4. The factor 11 is identified by a factor code C. The factor 11 can usually hold a factor name Fn, a factor notation type Ft, and a factor information code Cf. The factor 11 is linked to the factor information 12 by the factor information code Cf. Here, the factor notation type Ft indicates the representation format of the data handled in the factor 11, and examples include character notation representing character data and symbol data, and numeric notation representing numeric data.

[0019] Examples of the numerical notation include actual value notation, range notation, and rank notation. Actual value notation is a notation that expresses data handled by factor 11 using actual values ​​(actual values) given to factor 11. Range notation is a notation that expresses data handled by factor 11 using a representative value of a numerical range corresponding to a numerical range separated as a range. Rank notation is a notation that expresses data handled by factor 11 using a rank value corresponding to a set ranking.

[0020] Examples of character notation include name notation, symbol notation, etc. Name notation is notation that uses names to represent data handled by the factor 11. Symbol notation is notation that uses symbols to represent data handled by the factor 11.

[0021] The factor information 12 is identified by a factor information code Cf. The factor information 12 can usually hold a real value notation code Cj, a range notation code Ch, a rank notation code Cr, a name classification code Cn, and a symbol classification code Cs.

[0022] 4, a factor 11 is linked to connection information 13, which is information about the connection when interconnecting with a technically related factor 11, via a factor code C. The connection information 13 is identified by a connection information code Ck. The connection information 13 can usually hold a factor code Ca of the reference factor 111, a factor code Cb of the connection factor 112, and a connection condition code Cd. The connection information 13 is linked to a connection condition 14, which indicates the connection relationship between the reference factor 111 and the connection factor 112, via the connection condition code Cd.

[0023] Furthermore, the factor 11 is linked to binding information 15, which is information about binding with the upper factor 113 (or lower factor 113) via the factor code C. The binding information 15 is identified by the binding information code Cv. The binding information 15 can usually hold the factor code Ca of the reference factor 111, the factor code Cc of the upper factor or lower factor to which it is bound, the binding direction Dr indicating whether the binding destination is the upper factor or lower factor, and the binding condition code Cm. The binding information 15 is linked to a binding condition 16 indicating the binding relationship with the factor 11 to which it is bound via the binding condition code Cm.

[0024] Here, the connection condition 14 and the linking condition 16 will be explained. The connection condition 14 is for transmitting information between factors 11 that are technically interrelated in expressing the technical system related to architecture. Specifically, the connection condition 14 functions as a condition for determining the factor value of the connected factor 11 (i.e., the connecting factor 112) in the connection relationship from the factor value of the reference factor 11 (i.e., the reference factor 111). The connection condition 14 does not define a close hierarchical relationship like the linking condition 16, which will be described later, but rather expresses the technical interrelationship by mixing it with the connection condition 14 of the adjacent factor 11. Therefore, the factor value of the reference factor 111 is not directly reflected in the connecting factor 112, but is generally reflected in the connecting factor 112 in a limited manner, for example, by weighting according to a regression equation or contribution rate.

[0025] On the other hand, the combining condition 16 is for transmitting information between factors 11 having a hierarchical relationship (parent-child relationship or master-slave relationship). Specifically, the combining condition 16 functions as a condition for determining the factor value of a factor 11 (i.e., combining factor 113) that is combined by a hierarchical relationship from the factor value of a reference factor 11 (i.e., reference factor 111). The combining condition 16 combines the reference factor 111 and combining factor 113, which are in a close (e.g., physically) hierarchical relationship. Therefore, the factor value of the reference factor 111 is almost directly reflected in the combining factor 113 by the combining condition 16. Examples of hierarchical relationships include a combination of a subordinate concept factor and a superordinate concept factor, a combination of a component factor and a component-subordinate factor, and a combination of a characteristic expression factor and a specified expression factor.

[0026] Fig. 5 is a diagram showing the linking of information related to the connection condition 14. Fig. 5 shows an example of a more detailed lower-level connection condition 14 linked to the connection condition 14. The connection condition 14 is defined according to the basic pattern, i.e., the factor representation type Ft (see Fig. 4) of the data handled in two adjacent and interconnected factors 11.

[0027] As shown in Figures 4 and 5, the connection condition 14 can usually hold a factor relation code Cfr as its own internal information, and as necessary, a factor contribution Fk and a conjugate connection condition code Cw. The connection condition 14 is further linked to lower-level connection conditions 14 by a connection condition code Cd. The factor relation code Cfr is a code that represents the relationship between the reference factor 111 and the connection factor 112, and corresponds to a phenomenon, result, requirement, goal, etc. The factor contribution Fk is, for example, a contribution that represents the strength of the interrelationship between the reference factor 111 and the connection factor 112, and represents the weighting between the factors 11 in the interconnection. Here, the factor contribution Fk can also be set to include "0."

[0028] The conjugate connection condition code Cw is a condition when two or more divisors 11 form a pair and connect to one divisor 11. The divisor code C associated with two or more divisors 11 connected according to the conjugate connection condition code Cw is specifically called a "conjugate divisor code."

[0029] 5, examples of the lower-level connection conditions 14 include a logical expression connection condition 141, an arithmetic expression connection condition 142, a range connection condition 143, a rank connection condition 144, a real value connection condition 145, a symbol connection condition 146, and a name connection condition 147. These lower-level connection conditions 14 further hold address codes including conjugate connection condition codes Cw that link to lower levels and describe specific information about the connection conditions 14. By tracing these codes, specific information about the connection conditions 14 can be obtained.

[0030] As shown in Fig. 5, the address code can be exemplified by the logical expression code Cl or the constraint condition code Cx held in the logical expression connection condition 141 linked to the conjugate connection condition code Cw. Also, the address code can be exemplified by the registered approximation formula code Cp held in the arithmetic expression connection condition 142 linked to the conjugate connection condition code Cw. Furthermore, when there are multiple lower-level connection conditions 14, as shown in Fig. 5, it is also possible to set a priority P indicating which of the logical expression connection condition 141 and the arithmetic expression connection condition 142 should be used first.

[0031] The registered approximation formula code Cp is linked to, for example, a function generated between the factors 11. Here, the function generated between the factors 11, in other words, the trained model described below, is stored outside the knowledge model 10, for example, in a search program for searching the knowledge model 10 in the storage device 2. As a result, when searching for a trained model, the trained model can be called from outside the knowledge model 10 by tracing the conjugate connection condition code Cw assigned between the corresponding pair of factors 11 and referring to the registered approximation formula code Cp. Note that the registered approximation formula code Cp can also be linked to a logical formula or the like as long as it is a mathematical formula. Furthermore, the registered approximation formula code Cp is not limited to being linked to the conjugate connection condition code Cw, and can also be linked to, for example, a connection condition code Cd related to a one-to-one connection of the factors 11.

[0032] Examples of the address code include a range / actual value connection code Chj, a range / range connection code Chh, a range / rank connection code Chr, a range / name connection code Chn, and a range / symbol connection code Chs, which are held in the range connection condition 143. Examples of the address code include a rank / actual value connection code Crj, a rank / range connection code Crh, a rank / rank connection code Crr, a rank / name connection code Crn, and a rank / symbol connection code Crs, which are held in the rank connection condition 144.

[0033] Although not shown in the figure, the join conditions 16 are also linked to lower join conditions 16 in a similar manner. Examples of lower join conditions 16 include logical expression join conditions, calculation expression join conditions, range join conditions, rank join conditions, real value join conditions, symbol join conditions, and name join conditions. As with the connection conditions 14, with the join conditions 16, information on the specific join conditions 16 can be obtained by tracing the address codes that lead to the lower levels in which specific information on the join conditions 16 is written.

[0034] 2-1.Specific example of data table for Knowledge Model 10 3 and 4, the knowledge model 10 is configured by linking factors 11, factor information 12, connection information 13, connection conditions 14, bond information 15, and bond conditions 16 to one another. Furthermore, as shown in FIG. 5, the connection conditions 14 are linked to lower-level connection conditions 141-147. Furthermore, the knowledge model 10 has, for example, each of the factors 11, factor information 12, connection information 13, and connection conditions 14 as a data table. Similarly, the knowledge model 10 also has each of the lower-level connection conditions 141-147 as a data table. Each element shown in FIGS. 4 and 5 may have an independent data table, or some elements may have a common data table.

[0035] For example, a portion of a data table related to factor 11 is configured as shown in Fig. 6. That is, the data table related to factor 11 has a factor code C, a factor name Fn, a factor notation type Ft, and a factor information code Cf. A factor information code Cf is assigned to each factor name Fn. Furthermore, the factor information code Cf in the data table is set to a numerical value that matches the factor information code Cf in a data table (not shown) related to factor information 12. In this way, by matching the codes in each data table, the factors 11 are linked to each other.

[0036] Also, a portion of the data table relating to the connection condition 14 is configured as shown in Figures 7 and 8. However, Figures 7 and 8 show a data table in which the connection condition 14 and the lower rank connection condition 144 are combined and only a portion of the data table is excerpted. The connection condition 14 shown in Figure 7 is for connecting the "reference factor 111" and the "connection factor 112," and is a rank / rank connection. In other words, a correspondence relationship between the rank value of the "reference factor 111" and the rank value of the "connection factor 112" is defined. Although not shown, each defined rank value is assigned a connection condition code Cd.

[0037] 8 is for connecting the "reference factor 111" and the "connection factor 112," and is a rank / range connection. In other words, it defines the correspondence between the rank value of the "reference factor 111" and the range value of the "connection factor 112." The "range nominal value" shown in FIG. 8 is a representative value of a specific range defined by a lower and upper measurement limit values ​​that define the range value. Although not shown, a connection condition code Cd is assigned to each of the defined rank values / range values.

[0038] The connection condition code Cd in the data table relating to the connection condition 14 is set to a numerical value that matches the connection condition code Cd in the data table (not shown) relating to the connection information 13. By matching the codes in each data table in this way, the connection information 13 and the connection condition 14, that is, the reference factor 111 and the connection factor 112, are linked to each other.

[0039] As described above, the knowledge model 10 represents various types of building-related information, including know-how based on past experience and knowledge, as factors 11, and the relationships between the building-related information, i.e., knowledge, are expressed by interconnecting the factors 11 in a network. The knowledge model 10 then describes the relationships between the building-related information in a predetermined format, thereby storing knowledge (including tacit knowledge and know-how) as explicit knowledge. Therefore, according to the knowledge model 10, the factors 11 of building-related information in the field of building technology can be interconnected in a network via the connection information 13, connection conditions 14, etc., and knowledge (including know-how) can be stored as digital information as stylized explicit knowledge in a form that simulates the human brain.

[0040] 3. Details of the configuration of the processing unit 3 Next, the configuration of the arithmetic processing device 3 will be described with reference to Fig. 9. The arithmetic processing device 3 performs machine learning using a plurality of algorithms, using the reference factors 111 and the connection factors 112 connected to the reference factors 111 that form the knowledge model 10 stored in the storage device 2 as a training data set. As a result, the arithmetic processing device 3 generates a plurality of trained models for the correlation between the factors 11.

[0041] When generating a trained model, the arithmetic processing device 3 of this embodiment determines, as optimal, a trained model among the multiple trained models generated, for example, one in which the factor values ​​of the connection factors 112 calculated using each trained model have the smallest error relative to the target factor value of the connection factors 112. Then, the arithmetic processing device 3 stores the trained model determined to be optimal as the connection condition 14 in the knowledge model 10 (more specifically, in the storage device 2). Furthermore, the arithmetic processing device 3 uses the connection condition 14 stored in the knowledge model 10 to set a relationship between new reference factors 111 and connection factors 112, thereby constructing the knowledge model 10. For this reason, as shown in FIG. 9 , the arithmetic processing device 3 includes a learning processing device 31 capable of executing a learning phase, a knowledge model construction device 32 capable of executing an inference phase, and an information output device 33 that uses the knowledge model 10 to output connection factors 112 (i.e., answers) to reference factors 111 (i.e., questions or requirements) input by a practitioner in the field of construction technology.

[0042] The learning processing device 31 includes a training dataset acquisition unit 311, a training dataset storage unit 312, a trained model generation unit 313, and a judgment unit 314. The training dataset acquisition unit 311 is capable of constructing the knowledge model 10, and acquires, for example, all factors 11 that are construction-related information stored in the storage device 2 as a training dataset. Specifically, when executing the learning phase, the training dataset acquisition unit 311 acquires, as a training dataset, a basic pattern of the knowledge model 10, i.e., a plurality of reference factors 111 in which two adjacent factors 11 are interconnected by connection information 13, and one connection factor 112.

[0043] 9, the training dataset acquisition unit 311 can also acquire factors 11 as a training dataset from each of a plurality of information processing terminals 6 including a server installed in an architectural firm, etc. The training dataset acquired by the training dataset acquisition unit 311, i.e., the plurality of factors 11, is stored in the training dataset storage unit 312.

[0044] The trained model generation unit 313 performs machine learning using the factors 11 stored in the training dataset storage unit 312, more specifically, a plurality of reference factors 111 and one connection factor 112. In this embodiment, the trained model generation unit 313 performs machine learning using a plurality of different algorithms, such as a support vector machine (SVM), a random forest, and deep learning.

[0045] Specifically, the trained model generation unit 313 uses a plurality of different algorithms to extract, for example, the relationship between the reference factors 111 and the connection factors 112 stored in the training dataset storage unit 312, thereby stylizing the relationship. In addition, the trained model generation unit 313 uses a plurality of different algorithms to generate information on the interrelationship between adjacent reference factors 111 and connection factors 112.

[0046] Furthermore, the trained model generation unit 313 uses a plurality of different algorithms to calculate the factor contribution Fk based on the relationship between the reference factor 111 and the connection factor 112 and information on the interrelationship between the reference factor 111 and the connection factor 112. As a result, the trained model generation unit 313 generates a plurality of trained models regarding the connection conditions 14 between the reference factor 111 and the connection factor 112 according to each algorithm.

[0047] Here, as described above, information on the interrelationship between the reference factors 111 and the connection factors 112 and the factor contribution Fk are generated in principle by the trained model generation unit 313. However, instead of or in addition to the trained model generation unit 313 generating them, it is also possible for an expert who is knowledgeable about, for example, architectural design, architectural construction, architectural maintenance, etc. to manually set or adjust the information on the interrelationship between the reference factors 111 and the connection factors 112 and the factor contribution Fk in advance based on their past experience.

[0048] In this embodiment, a case is illustrated in which a plurality of trained model generation units 313 each generate a plurality of trained models corresponding to a different algorithm, as shown in Fig. 9. However, it is also possible for one trained model generation unit 313 to execute a plurality of algorithms and generate trained models corresponding to the respective algorithms.

[0049] The determination unit 314 determines the optimal trained model, i.e., the connection condition 14, in the knowledge model 10 from among the multiple trained models generated by the trained model generation unit 313. For example, when the reference factor 111 and the connection factor 112 are connected according to the generated trained model, the determination unit 314 of this embodiment determines that the trained model in which the factor value of the connected connection factor 112 has the smallest error with respect to the target factor value (actual value that is the actual factor value) of the connection factor 112 is optimal.

[0050] Specifically, for example, when the factor values ​​of the reference factor 111 and the connection factor 112 are both expressed in rank notation (see FIG. 7), the determination unit 314 calculates the rank value of the connection factor 112 using each generated trained model. Then, the determination unit 314 determines the error of the calculated rank value with respect to the actual rank value (actual value, which is the target factor value) of the connection factor 112. As a result, the determination unit 314 identifies the trained model that has been determined to have the smallest error as the optimal trained model for the knowledge model construction device 32 to set the connection condition 14, i.e., the connection condition 14.

[0051] After generating the optimal trained model, i.e., the connection condition 14, when new building-related information, i.e., new unknown factors 11, are added to the knowledge model 10, the learning processing device 31 again generates multiple trained models and identifies the optimal trained model. Specifically, the trained model generation unit 313 of the learning processing device 31 generates multiple trained models as described above, including the newly added factors 11 (reference factors 111 and connection factors 112) in addition to the factors 11 (reference factors 111 and connection factors 112) used in the previous machine learning. Then, the determination unit 314 identifies the optimal trained model, i.e., the connection condition 14, from among the multiple trained models, as described above. This maintains the knowledge model 10 in an optimal state, and as a result, accurate results can be provided when a search is performed.

[0052] The knowledge model construction device 32 mainly includes a trained model storage unit 321, a factor acquisition unit 322, a connection condition setting unit 323, and a knowledge model storage processing unit 324. The trained model storage unit 321 is provided, for example, outside the knowledge model 10, and stores the trained model generated by the trained model generation unit 313 to set the connection condition 14, in association with the conjugate connection condition code Cw and the registered approximate formula code Cp. The factor acquisition unit 322 acquires newly input factors 11 to construct the knowledge model 10.

[0053] The connection condition setting unit 323 sets a connection condition 14 for the newly input factor 11 based on the factor 11 acquired by the factor acquisition unit 322 and the trained model stored in the trained model storage unit 321. The knowledge model storage processing unit 324 stores the newly input factor 11, the connection condition 14, and connection information 13 based on the connection condition 14 (i.e., the factor code Ca of the reference factor 111 and the factor code Cb of the connection factor 112, etc.) in the knowledge model 10.

[0054] When a storage device 2 is provided in each of the information processing terminals 6, the knowledge model storage processing unit 324 supplies (distributes) the connection information 13 and the connection conditions 14 set by the connection condition setting unit 323 to each of the storage devices 2 via the network N. This allows, for example, an architectural designer who is a worker using the information processing terminal 6 or an architectural designer with little design experience to always use the latest knowledge model 10.

[0055] The information output device 33 mainly includes a factor input unit 331 and a factor output unit 332. The factor input unit 331 inputs questions and requirements input via the input device 4 as reference factors 111 (hereinafter, may be referred to as "input factors 111") to the knowledge model 10 constructed by the knowledge model construction device 32. Note that the factor input unit 331 can input questions and requirements input by an employee to the information processing terminal 6, which is an external device, as input factors 111 to the knowledge model 10, for example.

[0056] The factor output unit 332 outputs connection factors 112 (hereinafter, may be referred to as "output factors 112") that are output from the knowledge model 10 as having a relationship with the input factors 111 input by the factor input unit 331. Here, the factor output unit 332 can display the output factors 112 (e.g., answers) on the display device 5. Note that, for example, when an employee transmits input factors 111 (e.g., questions or requests) from the information processing terminal 6, the factor output unit 332 can output output factors 112 (e.g., answers) having a relationship to the information processing terminal 6 via the network N. In this case, the information processing terminal 6 can display the acquired output factors 112 (e.g., answers) on the display 66.

[0057] 4. Details of the configuration of the information processing terminal 6

[0058] Next, we will explain the configuration of the information processing terminal 6. In this embodiment, the information processing terminal 6 is a computer device used by architectural designers, building contractors, and the like, who are professionals in the field of architectural technology, and the following explanation will mainly exemplify a case where it is operated by an architectural designer.

[0059] 10 is a block diagram showing an example of the hardware configuration of the information processing terminal 6. The information processing terminal 6 is a computer device (for example, a personal computer) and mainly includes a CPU 61, a memory 62, an input device 63, a communication interface 64, a recording medium interface 65, and a display 66. The components 61-66 are connected to each other via a bus 67.

[0060] The CPU 61 performs overall control of the information processing terminal 6. Examples of the memory 62 include a ROM and a RAM. The ROM stores various programs that can be executed by the CPU 61. The RAM is used as a work area for the CPU 61. The programs stored in the memory 62 are loaded into the CPU 61, causing the CPU 61 to execute the coded processes.

[0061] Here, the various programs include CAD programs (CAD tools) that create two-dimensional data of buildings, CAD programs (CAD tools) that create three-dimensional data of buildings, etc. Programs that create three-dimensional data of buildings include BIM programs (BIM tools) that generate three-dimensional models of buildings.

[0062] An architectural designer or an operator (hereinafter, sometimes referred to as "operator, etc.") who is a worker using the information processing terminal 6 can generate various BIM data by using the BIM tool executed by the CPU 61. The BIM data is data including BIM elements such as building shapes, spatial relationships, the quantity and characteristics of building components, and properties of building elements.

[0063] The input device 63 includes a touch panel, a keyboard, a mouse, a microphone, etc. The communication interface 64 is connected to a network N including the Internet, etc., via a communication line. As a result, the information processing terminal 6 is connected to the arithmetic processing device 3 and the knowledge model 10 provided in the task support system 1 via the network N. The communication interface 64 also serves as an interface between the network N and the inside of the information processing terminal 6, and in particular controls the input and output of various data processed using the knowledge model 10 in the task support system 1. Note that examples of the communication interface 64 include a modem and a LAN (Local Area Network) adapter.

[0064] The recording medium interface 65 controls the reading and writing of data from and to recording media (not shown), such as magnetic disks, optical disks, and memory cards, under the control of the CPU 61. The display 66 is an output device that displays images such as characters, drawings, and three-dimensional models. Note that the information processing terminal 6 can also include, as output devices, a speaker for audio output and a printer for paper output (both not shown) in addition to the display 66.

[0065] Here, we will explain the characteristics of CAD tools that generate 2D and 3D data, and BIM tools that generate BIM data. Generally, CAD tools and BIM tools differ in their respective characteristics as follows. Specifically, CAD tools allow for easy and flexible drawing by simply drawing and erasing lines using an information processing terminal 6. Therefore, in the field of architectural technology, there are a relatively large number of operators who understand and can operate CAD tools. While manual creation is a time-consuming process in the design, construction, and maintenance of buildings, CAD tools allow for flexible response to individual requests and changes. Currently, individual building-related work is carried out by printing and sharing the created drawings on paper or other media, and this is another area where the use of CAD tools is considered advantageous.

[0066] On the other hand, CAD tools typically create and output drawings for each element of a building (e.g., a detailed drawing of each part of an overall drawing). Therefore, at a construction site, for example, if a part of a drawing is changed during construction, the change must be reflected separately in the overall drawing and other related detailed drawings. Therefore, when using CAD tools, it may be necessary to reconcile changes to some drawings with other drawings. In such cases, workers at a construction site, for example, must share information about the changes (change information) after drawings reflecting the changes are created and output. Furthermore, when using CAD tools, information necessary for construction (construction information) is written on drawings and output, so workers share the construction information through the output drawings. Thus, when using CAD tools, change information and construction information are shared through drawings output on media such as paper, which makes updating and checking this information cumbersome, potentially resulting in a slower process of information sharing.

[0067] Unlike CAD tools, BIM tools can automatically generate BIM data online. For example, after generating a 3D model of a building using a template, a BIM tool can automatically generate drawings for each BIM element from the generated 3D model using an information processing terminal 6 connected to a network N and operated by a worker. This allows for automatic generation of drawings for each BIM element, even if changes are required during construction at a construction site. For example, changes made to the 3D model using the information processing terminal 6 are automatically reflected in each drawing, ensuring consistency between drawings. Therefore, when a BIM tool is used, a worker can quickly and easily check change information and building information using the information processing terminal 6.

[0068] On the other hand, when BIM tools are operated using an information processing terminal 6, the operator is required to acquire specialized skills such as operation methods and knowledge. For this reason, currently, there are few operators and other personnel in the architectural technology field who can understand and operate BIM tools, making it difficult to quickly respond to various requests, including the creation and modification of templates, from architectural designers who are unfamiliar with operating BIM tools.

[0069] Therefore, the inventors of the present application have conceived of building a work support system 1 that can provide BIM design support so that even operators who have not sufficiently acquired specialized skills or who have little experience in designing, constructing, or maintaining buildings can obtain various types of building information that is useful when designing, constructing, or maintaining a building using a BIM tool. Below, by taking the example of performing a building design, we will specifically explain how to perform a building design using a BIM tool on an information processing terminal 6 after receiving design support from the work support system 1. Note that the specific operation of the BIM tool on the information processing terminal 6 is realized by a well-known BIM tool (BIM application), so detailed explanation of the operation will be omitted.

[0070] 4-1. Functional configuration of information processing terminal 6 11 , the information processing terminal 6 of this embodiment mainly includes an acquisition unit 611, a generation unit 612, a reception unit 613, a response unit 614, and a confirmation unit 615. The acquisition unit 611, the generation unit 612, the reception unit 613, the response unit 614, and the confirmation unit 615 are realized by the CPU 61. That is, the CPU 61 executes a predetermined program installed in advance, thereby realizing the functions of the acquisition unit 611, the generation unit 612, the reception unit 613, the response unit 614, and the confirmation unit 615.

[0071] The acquisition unit 611 acquires, for example, three-dimensional data of a building created using a CAD tool. Instead of acquiring three-dimensional data of a building, the acquisition unit 611 can acquire, for example, two-dimensional data (floor plan) and floor height information representing the heights of the building's floors. In this case, the information processing terminal 6 can generate three-dimensional data from the two-dimensional data and floor height information using a dedicated application. Furthermore, as will be described later, for example, when an operator or the like directly generates a three-dimensional model of a building using a provided template including BIM elements representing the building elements that make up the building, the acquisition unit 611 can be omitted.

[0072] The generation unit 612 generates BIM data representing a three-dimensional model including building elements, i.e., BIM elements, based on the three-dimensional data acquired by the acquisition unit 611. Examples of BIM elements include three-dimensional data for walls, floors, columns, ceilings, interior materials, finishing materials, and fixtures. Examples of BIM data include an architectural design model generated in the design of a building's architectural design, a structural design model generated in the design of a building's structure, and an equipment design model generated in the design of a building's equipment. The generation unit 612 can generate each of the architectural design models, structural design models, and equipment design models shown above simultaneously or separately.

[0073] The receiving unit 613 receives various inputs such as text, figures, and voices by an operator or the like using the input device 63. Specifically, the receiving unit 613 receives input of questions, requests, etc. (hereinafter, may be referred to as "questions, etc.") by the operator or the like to the answering unit 614, and input of confirmation items by the operator or the like to the confirmation unit 615. Here, the questions, etc. and confirmation items are request information that requests support from the knowledge model 10.

[0074] The answering unit 614 presents answers to questions, etc., from an operator, etc., received by the receiving unit 613. Specifically, the answering unit 614 transmits information (e.g., text data) representing the questions, etc., from an operator, etc., to the work support system 1 via the communication interface 64. Here, examples of the questions, etc., include questions requesting answers on how to generate or operate BIM elements (BIM data), requests to generate templates used when generating BIM elements (BIM data), various information linked to BIM elements in the templates, and requests to present proposals for the design and construction of buildings. Then, the answering unit 614 acquires information (e.g., text data or graphic data) representing answers to the questions, etc., generated by the work support system 1, that is, generated using the knowledge model 10, via the communication interface 64, and displays the information on the display 66 as answers to the questions, etc.

[0075] The confirmation unit 615 presents the confirmation results for the confirmation items by the operator or the like received by the reception unit 613. Specifically, the confirmation unit 615 transmits information (e.g., text data) representing the confirmation items by the operator or the like to the work support system 1 via the communication interface 64. Examples of the confirmation items include confirmation of whether a designed or constructed building complies with building regulations, and confirmation of whether the building specifications (e.g., various building areas) of the designed building comply with pre-set building specifications. The confirmation unit 615 then acquires information (e.g., text data or graphic data) representing the confirmation results for the confirmation items generated by the work support system 1, that is, generated using the knowledge model 10, via the communication interface 64, and displays the confirmation results for the confirmation items (determination results DI, described later) on the display 66.

[0076] 5. Examples of assistance modes in the work assistance system 1 Next, a specific example of a support mode in which the processing device 3 and the information processing terminal 6 support a worker using the knowledge model 10 in the work support system 1 will be described. Note that the following description will exemplify a case in which an operator or the like generates an architectural design model on the information processing terminal 6.

[0077] 5-1. First example When generating BIM data for an architectural design model, for example, a room in a building, an operator or the like generates the data using various templates T displayed in a template area 661 on the display 66, as shown in FIG. 12 . That is, the operator or the like uses the input device 63 to select a template T displayed in the template area 661. The generation unit 612 then uses the selected template T as a BIM element to generate BIM data based on the two-dimensional data or three-dimensional data displayed in the drawing area 662. Alternatively, the generation unit 612 can generate BIM data by dragging and dropping the selected template T onto the two-dimensional data or three-dimensional data displayed in the drawing area 662. Furthermore, the operator or the like can use the input device 63 to create a new drawing using the selected template T in the drawing area 662.

[0078] In this way, when generating an architectural design model, for example, an operator can input a request to the answering unit 614 to display BIM elements, i.e., template T, representing interior walls, interior fittings, and fixtures, for example, by text input or voice input to the receiving unit 613 using the input device 63 in the help area 663 on the display 66. In this case, in order to generate BIM data for the room more accurately, in other words, to obtain a more appropriate template T as a BIM element, the operator can specify, for example, the type of building (e.g., a house, an apartment, an office, a hospital, a store, etc.) and the architectural specifications of the interior walls, interior fittings, columns, and fixtures (e.g., wallpaper material and color, column material, door and window material, etc.). Then, in the information processing terminal 6, the answering unit 614 transmits a request for BIM elements, i.e., template T, including the items (type of building and architectural specifications) input by the operator as described above, to the arithmetic processing device 3 via the network N as information such as a question.

[0079] In the work support system 1, the factor input unit 331 of the arithmetic processing device 3 inputs keywords such as "interior wall," "apartment," "interior," and "wall material" via the network N as input factors 111 for the knowledge model 10 stored in the storage device 2. As a result, the factor output unit 332 of the arithmetic processing device 3 outputs a "template T" as the output factor 112 from the knowledge model 10. Here, the template T, which is the output factor 112 output for the input factor 111, may be, for example, a template T previously generated by a design office or the like, or a template T newly generated using newly developed building materials or functions.

[0080] Specifically, the arithmetic processing device 3 starts from the acquired input factors 111 and searches the knowledge model 10 while tracing the connection information 13 and connection conditions 14. This allows the factor output unit 332 to output multiple candidate templates T, which are composed of image data or drawing data capable of generating BIM elements (buildings and three-dimensional elements) with different "wall covering materials and colors" for an "interior wall" for an "apartment building" requested by an operator, for example. The candidate templates T can be linked to building information that may include, for example, the manufacturer and product number (catalog number) of the wallpaper material used, the quantity (size and area), material cost, inventory status, and current delivery date. The arithmetic processing device 3, more specifically, the factor output unit 332, then outputs the output candidate templates T to the information processing terminal 6 via the network N.

[0081] In the information processing terminal 6, the response unit 614 acquires the candidate templates T output via the communication interface 64, and displays them in the template area 661 of the display 66, as shown in Fig. 12. This allows the operator or the like to generate an interior wall as a BIM element using an appropriate template T from the displayed candidate templates T, and to generate BIM data of the room (three-dimensional data of the building) using the generated BIM element of the interior wall.

[0082] In this case, the generated BIM data is composed of multiple BIM elements, and each BIM element is linked to building information. As a result, in processes later than the design process (construction and maintenance processes), workers working in each process can display the architectural design model and equipment design model online using information processing terminals 6 connected to network N, and display and confirm the building information linked to the "interior wall BIM elements" that form the "room BIM data." In this case, by checking the building information linked to the BIM elements (templates T) by the knowledge model 10 on the information processing terminals 6, workers can also confirm, for example, the mechanical strength of the cross material, the manufacturer's inventory status, cost information, etc.

[0083] 5-2.Second example Furthermore, the operator or the like can request the answering unit 614, for example, calculation results of the floor area for each room or for all rooms in the structural design model. That is, the operator or the like inputs, for example, an area calculation of the floor area of ​​a floor represented by "BIM data of the room" as a request item in the help area 663 or check area 664 of the display 66 to the receiving unit 613 using the input device 63. Here, the answering unit 614 transmits, for example, dimensional information of the wall center (including the room shape) based on the "BIM data of the room" to the arithmetic processing device 3 via the network N as a question or the like.

[0084] In the work support system 1, the factor input unit 331 of the arithmetic processing device 3 inputs the keyword "floor area calculation" together with the "wall center dimension information" acquired as the input factor 111 for the knowledge model 10 stored in the storage device 2. As a result, the factor output unit 332 of the arithmetic processing device 3 outputs, as the output factor 112 from the knowledge model 10, the "floor area M" calculated for the room shape identified based on the wall center dimension information.

[0085] Specifically, in the arithmetic processing device 3, the knowledge model 10 is searched while tracing the connection information 13 and the connection conditions 14, and the factor output unit 332 outputs the floor area M as the calculation result K, for example, according to the area calculation formula used when performing an area calculation for a similar room shape in response to a past calculation request. Then, the arithmetic processing device 3, more specifically the factor output unit 332, outputs the calculation result K to the information processing terminal 6 via the network N.

[0086] In the information processing terminal 6, the answering unit 614 acquires the calculation result K output via the communication interface 64, and displays the floor area M in the check area 664 of the display 66, as shown in Fig. 12. This allows the operator or the like to very easily confirm the floor area M as the calculation result K without having to perform the area calculation themselves.

[0087] 5-3. Third example Furthermore, the operator or the like can check, for example, whether the "room BIM data" in the structural design model conforms to pre-set building specifications or whether it complies with laws and regulations, using the confirmation unit 615. That is, the operator or the like inputs, for example, a request to the reception unit 613 using the input device 63, in the check area 664 of the display 66, to determine whether the floor area of ​​the "room BIM data" conforms to the floor area of ​​the floor pre-set in the building specifications. Here, the confirmation unit 615 transmits, for example, dimensional information of the wall center (including the room shape) as a question or the like to the arithmetic processing device 3 via the network N based on the "room BIM data."

[0088] In the work support system 1, the factor input unit 331 of the arithmetic processing device 3 outputs the "floor area M" calculated as described above. Here, the arithmetic processing device 3 may use the knowledge model 10 to calculate and output the room perimeter, room volume, etc. in addition to the output floor area. Then, the arithmetic processing device 3, more specifically the factor output unit 332, outputs the calculation result K to the information processing terminal 6 via the network N.

[0089] In the information processing terminal 6, the confirmation unit 615 acquires the calculation result K output via the communication interface 64, and displays the floor area M in a check area 664 on the display 66, as shown in Fig. 12. The confirmation unit 615 also displays the set floor area Mo, which is set in advance according to the building specifications, in the check area 664. Note that the building specifications can be stored in the information processing terminal 6 in advance by, for example, an operator or the like when designing a building.

[0090] The confirmation unit 615 can then compare the floor area M with the set floor area Mo and present a judgment as to whether or not the floor area M matches the set floor area Mo. This allows the operator or the like to extremely easily confirm the floor area M as the calculation result K without having to perform the area calculation themselves, and also to easily confirm whether or not the "room BIM data" generated (designed) for the set floor area Mo matches the building specifications.

[0091] 5-4.Fourth example The operator or the like can also use the confirmation unit 615 to confirm, for example, whether the building coverage ratio, which indicates the ratio of the building area of ​​a building to the site area, and the floor area ratio, which indicates the ratio of the total floor area to the site area, in the architectural design model or the structural design model, satisfy regulations (e.g., the Building Standards Act). Specifically, the operator or the like inputs a request, for example, whether the building coverage ratio and floor area ratio satisfy the Building Standards Act, into the reception unit 613 using the input device 63 in the check area 664. Here, the confirmation unit 615 transmits, as questions or the like, the dimensional information of the site (including the site shape), the exterior dimensions (including the exterior shape) of the architectural design model, and the dimensional information of the wall center lines (including the room shapes) based on the "BIM data of all rooms" of the building to the arithmetic processing device 3 via the network N.

[0092] In the work support system 1, the factor input unit 331 of the arithmetic processing device 3 inputs the acquired "site dimension information" as the input factor 111 for the knowledge model 10 stored in the storage device 2, and also inputs the keyword "site area calculation." As a result, the arithmetic processing device 3 searches the knowledge model 10 while tracing the connection information 13 and connection conditions 14, and the factor output unit 332 outputs the "site area" calculated from the knowledge model 10 based on the site dimension information as the output factor 112.

[0093] Furthermore, the factor input unit 331 of the arithmetic processing device 3 inputs the "exterior dimensions of the architectural design model" and the keyword "area calculation of building area" as input factors 111 to be input to the knowledge model 10. As a result, the arithmetic processing device 3 searches the knowledge model 10 while tracing the connection information 13 and connection conditions 14, and the factor output unit 332 outputs the "building area" calculated based on the exterior dimensions from the knowledge model 10 as the output factor 112.

[0094] Furthermore, the factor input unit 331 of the arithmetic processing device 3 inputs the "wall center dimension information" and the keyword "area calculation of total floor area" as the input factor 111 to be input to the knowledge model 10. As a result, the arithmetic processing device 3 searches the knowledge model 10 while tracing the connection information 13 and connection conditions 14, and the factor output unit 332 outputs the "total floor area" calculated based on the wall center dimension information from the knowledge model 10 as the output factor 112.

[0095] Next, in the arithmetic processing device 3, the factor input unit 331 inputs the calculated "site area" and "building area" as input factors 111 to be input to the knowledge model 10, as well as the keyword "calculation of building coverage ratio." As a result, the arithmetic processing device 3 searches the knowledge model 10 while tracing the connection information 13 and connection conditions 14, and the factor output unit 332 outputs the "building coverage ratio B," calculated as the ratio of the building area to the site area, as the output factor 112 from the knowledge model 10, as the calculation result K. Then, the arithmetic processing device 3, more specifically the factor output unit 332, outputs the calculation result K to the information processing terminal 6 via the network N.

[0096] Furthermore, in the arithmetic processing device 3, the factor input unit 331 inputs the calculated "site area" and "total floor area" as input factors 111 to be input to the knowledge model 10, as well as the keyword "calculate floor area ratio." As a result, the arithmetic processing device 3 searches the knowledge model 10 while tracing the connection information 13 and connection conditions 14, and the factor output unit 332 outputs the "floor area ratio F," calculated as the ratio of the total floor area to the site area, as the output factor 112 from the knowledge model 10, as the calculation result K. Then, the arithmetic processing device 3, more specifically the factor output unit 332, outputs the calculation result K to the information processing terminal 6 via the network N.

[0097] Furthermore, in the arithmetic processing device 3, the factor input unit 331 inputs the calculated "building coverage ratio B" and "floor area ratio F" as input factors 111 to be input to the knowledge model 10, as well as the keyword "legal judgment." As a result, the arithmetic processing device 3 searches the knowledge model 10 while tracing the connection information 13 and connection conditions 14, and the factor output unit 332 outputs, as the output factor 112 from the knowledge model 10, a judgment result DI indicating whether or not the "building coverage ratio B" and "floor area ratio F" satisfy the standards set forth in the Building Standards Act, which is a law. Then, the arithmetic processing device 3, more specifically the factor output unit 332, also outputs the judgment result DI to the information processing terminal 6 via the network N.

[0098] In the information processing terminal 6, the confirmation unit 615 acquires the calculation result K and the judgment result DI output via the communication interface 64, and displays in the check area 664 whether or not the criteria are met, along with the building coverage ratio B and floor area ratio F. This allows the operator or the like to very easily confirm the building coverage ratio B and floor area ratio F as the calculation result K without having to perform the calculations themselves.

[0099] Furthermore, the operator or the like can easily check as a judgment result DI whether the generated (designed) building complies with laws and regulations (for example, the Building Standards Act, etc.). Note that, instead of the arithmetic processing device 3 (i.e., the knowledge model 10) determining whether the building coverage ratio B and floor area ratio F meet the legal standards, if the information processing terminal 6 has acquired legal information representing the laws and regulations in advance, the information processing terminal 6 may determine whether the acquired "building coverage ratio B" and "floor area ratio F" meet the standards.

[0100] 5-5.Fifth example Furthermore, an operator or the like can receive, for example, proposals of multiple design proposals that can be used as reference when designing a building on the information processing terminal 6. In this case, the operator or the like can input, by text input or voice input to the reception unit 613 using the input device 63 in the help area 663, to the answering unit 614, for example, information on the dimensions of the site on which the building is to be constructed, the condition of the surrounding area of ​​the site (presence and location of roads, etc.), and the desired exterior style of the building (Western or Japanese style, etc.), and request that a building proposal for a building that satisfies these conditions and relevant laws and regulations be proposed. Then, in the information processing terminal 6, the answering unit 614 transmits the questions and the like input by the operator or the like to the arithmetic processing device 3 via the network N.

[0101] In the work support system 1, the factor input unit 331 of the arithmetic processing device 3 inputs the keyword "proposal of architectural plan" together with the "site dimension information," "surrounding conditions," and "appearance of the building" acquired as input factors 111 for the knowledge model 10 stored in the storage device 2. As a result, the factor output unit 332 of the arithmetic processing device 3 outputs the "generated plan G" generated as an architectural plan from the knowledge model 10 as the output factor 112.

[0102] Here, the generation proposal G, which is the output factor 112 output for the input factor 111, is generated by, for example, varying the appearance or shape of the building or varying the layout of the building on the site based on the appearance of a building that was designed in the past or reflects the latest trends, past layout patterns of buildings on the site, the latest laws and regulations, etc. Furthermore, the knowledge model 10 can generate multiple patterns of generation proposal G, for example, as an architectural design model, to serve as a reference for operators, etc. when designing an architecture.

[0103] Specifically, the arithmetic processing device 3 searches the knowledge model 10 while tracing the connection information 13 and connection conditions 14 starting from the acquired input factors 111, and the factor output unit 332 outputs, as a generation proposal G, a design design model generated according to, for example, blueprints, three-dimensional models, or design models of buildings previously designed by the architectural firm to which the architectural designer belongs or of buildings that reflect current trends. Then, the arithmetic processing device 3, more specifically the factor output unit 332, outputs the generation proposal G to the information processing terminal 6 via the network N.

[0104] Furthermore, by searching the knowledge model 10, the arithmetic processing device 3 can output, as output factors 112, generation instruction information to be input to another generation execution program (generation execution application) that automatically generates images and figures based on given conditions, for example, to serve as a reference for an operator or the like when designing an architecture. That is, in the arithmetic processing device 3, the factor input unit 331 inputs, as input factors 111, a keyword such as "architectural plan proposal" along with "site dimension information," "surrounding environment information," and "building exterior." As a result, the arithmetic processing device 3 searches the knowledge model 10 starting from the input factors 111 while tracing the connection information 13 and connection conditions 14, and the factor output unit 332 outputs, as output factors 112, generation instruction information that includes, for example, characteristic items (exterior and room characteristics, etc.) of Japanese-style buildings and Western-style buildings that have been designed in the past according to the shape of the site and the surrounding environment.

[0105] Then, the arithmetic processing device 3 inputs the generated generation instruction information as a prompt to, for example, a generation AI program (generation AI application) on the cloud via the network N. As a result, the generation AI program (generation AI application) outputs a generated image that can become a design proposal. In this case, the generation AI program (generation AI application) can use a well-known program (application). As a result, the arithmetic processing device 3, more specifically, the factor output unit 332, can output the generated image generated by the generation program (generation application) to the information processing terminal 6 as a generation proposal G.

[0106] In the information processing terminal 6, the response unit 614 acquires the generation plan G output via the communication interface 64 and displays the generation plan G consisting of multiple architectural design models in the help area 663, as shown in Fig. 12. This allows operators, etc., especially architectural designers, to design buildings that comply with the latest laws and regulations while referring to past buildings and buildings that reflect the latest trends when designing their own buildings. As described above, operators, etc., can request the confirmation unit 615 to confirm the floor area M, building coverage ratio B, and floor area ratio F of the building they have actually designed.

[0107] 5-6. Sixth example Furthermore, when the information processing terminal 6 is used in building construction or building maintenance, an operator or the like can check, for example, the cost and delivery date of building materials based on the building information linked to the BIM elements. In this case, the operator or the like inputs, for example, the name and catalogue number of the building material represented by the building information to the answering unit 614 by text input or voice input to the receiving unit 613 using the input device 63 in the help area 663, and requests that the cost and delivery date of the corresponding building material be checked. Then, in the information processing terminal 6, the answering unit 614 transmits the questions and the like input by the operator or the like to the arithmetic processing device 3 via the network N.

[0108] In the work support system 1, the factor input unit 331 of the arithmetic processing device 3 inputs the keyword "confirm cost and delivery date" together with the acquired "name" and "catalog number" as input factors 111 for the knowledge model 10 stored in the storage device 2. As a result, the factor output unit 332 of the arithmetic processing device 3 outputs, as output factors 112, for example, address information of the server device of the manufacturer of the building material specified by the catalog number or address information of the server device of the store that manages inventory of the building material.

[0109] Specifically, the arithmetic processing device 3 searches the knowledge model 10 starting from the acquired input factor 111 while tracing the connection conditions 14, and the factor output unit 332 acquires and outputs address information, for example, of a server device of a manufacturer or store published on the Internet, which address information was registered when the server device was accessed in the past. The arithmetic processing device 3 then accesses the server device based on the output address information and acquires page information PA representing a homepage created to notify customers of costs and delivery times. The arithmetic processing device 3, more specifically, the factor output unit 332, then outputs the acquired page information PA to the information processing terminal 6 via the network N.

[0110] In the information processing terminal 6, the response unit 614 acquires the page information PA output via the communication interface 64, and displays, for example, the cost and delivery date based on the page information PA in the help area 663. This allows the operator or the like to confirm the cost and delivery date of the necessary building materials, for example, in a pre-meeting when carrying out building construction or building maintenance, and enables the operator or the like to accurately set a work schedule or create a work estimate with high accuracy.

[0111] As can be understood from the above explanation, even if a worker in the field of architectural technology does not have sufficient specialized skills to use BIM, appropriate support can be provided according to the worker's requests by using the knowledge model 10. As a result, the worker can obtain useful information from BIM and proceed with his / her work while appropriately receiving the support he / she desires on the information processing terminal 6.

[0112] 6. Variations In the work support system 1 of the above-described embodiment, the storage device 2, the arithmetic processing device 3, and the information processing terminal 6 are connected via a network N, and the arithmetic processing device 3 searches the knowledge model 10 stored in the storage device 2 to provide support information for supporting work on the information processing terminal 6. That is, in the above-described embodiment, the storage device 2, the arithmetic processing device 3, and the information processing terminal 6, which are connected via the network N, work in cooperation with each other to provide support information to an operator or the like who uses the information processing terminal 6.

[0113] However, for example, the information processing terminal 6 can be configured to directly search the knowledge model 10 stored in the storage device 2 and acquire the support information. That is, in this case, the information processing terminal 6 can acquire the support information directly without going through the arithmetic processing device 3. Also, for example, instead of storing the knowledge model 10 in the storage device 2, the information processing terminal 6 can be configured to store the knowledge model 10. In this case, for example, even when not connected to the network N, the information processing terminal 6 can search the knowledge model 10 and provide the support information to the operator or the like.

[0114] Also, for example, it is possible to provide the storage device 2 and the arithmetic processing device 3 integrally. In other words, in this case, the arithmetic processing device 3 can directly construct the knowledge model 10 and also directly search the knowledge model 10. Furthermore, for example, it is possible to provide the storage device 2, the arithmetic processing device 3 and the information processing terminal 6 integrally. In other words, in this case, it is possible to construct the knowledge model 10 using the information processing terminal 6 and also directly search the constructed knowledge model 10.

[0115] Furthermore, in the work assistance system 1 of the above-described embodiment, the knowledge model 10 is stored in the storage device 2 connected to the network N, and assistance information is provided by searching the stored knowledge model 10. However, for example, it is also possible to configure the system so that a distributor having the storage device 2 and the arithmetic processing device 3 constructs the knowledge model 10 for the purpose of providing assistance information to an operator who operates the information processing terminal 6, and distributes the constructed knowledge model 10 to the information processing terminal 6 via the network N.

[0116] In this case, for example, general building-related information, i.e., a knowledge model 10 expressing the relationship between knowledge, is provided to the information processing terminal 6 by distribution. Then, similar to the arithmetic processing device 3, the information processing terminal 6 can proceed with machine learning of the relationship between the reference factors 111 and the connection factors 112, i.e., the connection information 13 and the connection conditions 14, and continue to build the knowledge model 10. As a result, the knowledge model 10 continues to be built using building-related information (including know-how) obtained when an operator or the like performs building design, building construction, building maintenance, etc., and it becomes possible to provide more useful support information to the operator or the like. [Explanation of symbols]

[0117] 1...work support system, 2...storage device, 3...arithmetic processing device (computer device), 31...learning processing device, 311...training dataset acquisition unit, 312...training dataset storage unit, 313...trained model generation unit, 314...judgment unit, 32...knowledge model construction device, 321...trained model storage unit, 322...factor acquisition unit, 323...connection condition setting unit, 324...knowledge model storage processing unit, 33...information output device, 331...factor input unit, 332...factor output unit, 4...input device, 5...display device, 6...information processing terminal (computer device, external device), 61...CPU, 611...acquisition unit, 612...generation unit , 613...reception unit, 614...answer unit, 615...confirmation unit, 62...memory, 63...input device, 64...communication interface, 65...recording medium interface, 66...display, 661...template area, 662...drawing area, 663...help area, 664...check area, 67...bus, 10...knowledge model, 11...factor, 111...criteria factor (input factor), 112...connection factor (output factor), 12...factor information, 13...connection information, 14...connection condition, 15...connection information, 16...connection condition, T...template, G...generation plan, K...calculation result, DI...judgment result, PA...page information

Claims

1. a knowledge model including one or more reference factors among a plurality of factors, connection factors among the factors that connect to the reference factor, and connection conditions for connecting the reference factor and the connection factor; a computer device that inputs input factors, which are the reference factors, into the knowledge model, and searches the knowledge model using the connection conditions to output output factors, which are the connection factors corresponding to the input factors; The knowledge model: The information is constructed using factors that are related to at least the architectural technology field, including architectural education, architectural design, architectural construction, and architectural maintenance, and that are used in BIM (Building Information Modeling), and legal information that represents laws and regulations related to the architectural technology field, The computer device A trained model generated by machine learning regarding the correlation between the plurality of factors that construct the knowledge model is set as the connection condition, A work support system configured to input request information requesting support from the knowledge model for the construction-related information and the legal information as the input factors into the knowledge model in response to instructions from a worker engaged in the construction technology field, search the knowledge model using the connection conditions, and present the worker with the output factors corresponding to the input factors.

2. The computer device 2. The work support system according to claim 1, further comprising an answering unit that searches the knowledge model based on the input factors input by the worker and presents the output factors, which are answers to the input factors, to the worker.

3. The input factor is a template used by the worker in the BIM, and is a requirement requesting the provision of the template to which at least the building-related information is linked, The answering section: The work support system according to claim 2 , wherein the knowledge model is searched based on the requirements, and the template serving as the output factor is presented to the worker.

4. The answering section: The work support system according to claim 3 , wherein the knowledge model is searched to generate the template linked to the construction-related information and present the template to the worker.

5. The input factors are questions requiring answers input by the worker using the BIM; The answering section: The work support system according to claim 2 , wherein the knowledge model is searched based on the question, and the answer serving as the output factor is presented to the worker.

6. The input factors are requirements for providing a design plan of a building referenced by the practitioner using the BIM, the design plan being in a style intended by the practitioner and satisfying the laws and regulations; The answering section: The work support system according to claim 2 , wherein the knowledge model is searched based on the requirements, and the design proposals that become the output factors are presented to the worker.

7. The input factors are requirements requiring calculated areas related to buildings designed using the BIM; The answering section: The work support system according to claim 2, wherein the area serving as the output factor calculated according to a formula for calculating the area obtained by searching the knowledge model based on the requirements is presented to the worker.

8. The computer device 2. The work support system according to claim 1, further comprising a confirmation unit that searches the knowledge model based on the input factors input by the worker to confirm whether the output factors output based on the input factors satisfy a predetermined standard.

9. The input factors are confirmation items that require calculating an area related to a building designed using the BIM and confirming whether the calculated area satisfies the criteria, The confirmation unit:

9. The work support system according to claim 8, wherein the area calculated according to a formula for calculating the area obtained by searching the knowledge model based on the confirmation items is compared with the standard, and a judgment result that becomes the output factor is presented to the worker.

10. The work support system according to claim 8 , wherein the standard is determined by the law related to the construction technology field.

11. The computer device 2. The work support system according to claim 1, further comprising: a factor input unit that inputs the input factors from an external device to the knowledge model; and a factor output unit that outputs the output factors from the knowledge model to the external device.

12. The knowledge model: The work support system according to claim 1 , wherein the factors are the construction-related information and the legal information, and the factors are interconnected in a network form to represent the relationship between the factors.

Images