JP2019075062A - Design support apparatus and design support method - Google Patents

Abstract

To reduce human burden by re-using existing design data to enable the entire design operation to be efficient.SOLUTION: A design support apparatus 100 according to the present invention has a storage device 101 for storing database for existing design database relating to architectural structures, and an arithmetic apparatus 104 for extracting, from design information relating to an architectural structure of a design object, information on existence or absence of predetermined facilities and information on arranged positions of the predetermined facilities in the architectural structure, identifying existing design data having similarities regarding the existence and arranged position information based on machine learning algorithm learning determination in such similarities, and outputting the existing design data to a predetermined apparatus.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a design support apparatus and a design support method, and more particularly, to a technology for reducing the human load by reusing existing design data, and enabling the efficiency of the entire design work. .

  Various design tasks in the construction industry are difficult to achieve efficiency through automation and the like because of the uniqueness of each building to be designed. Therefore, there is also a problem that it is difficult to maintain and establish the business system only by continuing the appropriate human capital investment. On the other hand, with the recent development of information processing technology in both software and hardware, a movement to search for more efficient design work has also begun.

  The following techniques have been proposed as conventional techniques for improving the efficiency of such design work. That is, a part name rule making process that configures a part name as an identification code of a predetermined part including similar parts by a predetermined name and a predetermined specification item, and drawing of the predetermined part based on the part name The drawing creation rule making step of predetermining the procedure for enabling the processing in a predetermined CAD system, the predetermined name of the part constituting the data of the part name, and the data of the predetermined specification item And a part name data input process for inputting to a CAD system, wherein the predetermined CAD system automatically creates a drawing of a predetermined part corresponding to the part name based on the data of the input part name. A CAD drawing automatic creation method (see Patent Document 1) or the like is proposed.

  In addition, CAD data reading means for reading elevation data and plan view data of a building created by a CAD (Computer Aided Design) device, and information based on the elevation data, wherein the outline of the building is At the intersection point with the ground surface, at least the information on the height of each layer when the building was divided into each layer was divided at the position of the vertex on the outline on which the coordinate in the horizontal axis differs, and the building was divided into each layer. Layer division data generation means for generating layer division data, outer shape data generation means for generating outer shape data holding information of each outer shape of the building shown in a plan view corresponding to each of the layers, and The three-dimensionalization data in which the external shape shown is the external shape of the uniform plan view in the layer shown in the plan view is all data using the layer division data and the external shape data. Three-dimensional data generation means for generating the layer and displaying the three-dimensional data on the screen according to the reference point, and receiving the installation position of the lightning rod in the three-dimensional data and the height of the lightning rod Accepts the input position of the lightning rod data generation means for combining the data to the three-dimensional data of the building, the installation position of the upper conductor arranged as lightning protection equipment on the roof of the building indicated by the three-dimensional data, and the height of the upper conductor. Receiving input of the installation position of the mesh conductor arranged as lightning protection equipment on the wall surface of the building indicated by the three-dimensional data and the three-dimensional data of the above-mentioned three-dimensional data of the three-dimensional data A three-dimensional data generation system for a building, comprising mesh conductor generation means for combining three-dimensional data of the mesh conductor with three-dimensional data of the building; It has also been proposed (see Patent Document 2), etc..

JP-A-8-249371 JP 2008-102843 A

  For example, a case where automatic design is subsequently performed according to a predetermined rule based on a building or floor plan of a building, or a figure shown by the floor plan or the like is vectorized, and existing design data is based on the similarity In any case, the uniqueness of each building mentioned above is assumed, and in any case the existing design data can not be reused efficiently, but the amount of computation becomes enormous and the processing diverges There is also a risk of Therefore, there is a strong possibility that the task can be performed only at a processing speed which is not practical.

Also, even if the intention of automatic design is deliberately set, if the design accuracy of the product is not good, it will eventually require additional confirmation and correction work by a knowledgeable design person in charge. In such a case, not only the efficiency of design work can not be achieved, but a situation where the load on personnel is higher than usual also occurs.
Therefore, an object of the present invention is to provide a technology capable of reducing the human load by reusing existing design data and enabling the efficiency of the entire design work.

  The design support apparatus of the present invention for solving the above problems includes the presence or absence and the arrangement position of a predetermined facility in the building based on a storage device holding a database of existing design data on the building and design information on the building to be designed. A computing device that extracts each piece of information, specifies existing design data having similarity with respect to each piece of information by a machine learning algorithm that performs learning regarding determination of the similarity, and outputs the existing design data to a predetermined device; It is characterized by including.

  Further, according to the design support method of the present invention, the information processing apparatus including the storage device holding the database of the existing design data on the building is based on the design information on the building to be designed. Extracting each piece of information of position, specifying existing design data having similarity with respect to each piece of information by a machine learning algorithm that performs learning regarding determination of the similarity, and outputting the existing design data to a predetermined device It features.

  According to the present invention, it is possible to reuse existing design data to reduce human load and to make the entire design work more efficient.

It is a network block diagram containing the design support device of this embodiment. It is a figure showing the example of hardware constitutions of the design support device in this embodiment. It is a figure which shows the hardware structural example of the person-in-charge terminal in this embodiment. It is a figure which shows the structural example of design data DB in this embodiment. It is a figure which shows the example of a flow of the design support method in this embodiment. It is an explanatory view showing the example 1 of an output in this embodiment. It is an explanatory view showing the example 2 of an output in this embodiment. It is an explanatory view showing the example 3 of an output in this embodiment.

――― Network configuration ―――

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a network configuration diagram including a design support apparatus 100 of the present embodiment. The design support apparatus 100 shown in FIG. 1 is a computer apparatus that aims to reuse existing design data to reduce human load and to make the entire design work more efficient.

  Specifically, such a design support apparatus 100 can be assumed to be a server apparatus operated by various builders and designers who undertake design work from the builders. In any case, it is an apparatus provided with a track record of having performed design work for many buildings, and a person who can use a large amount of design data corresponding to it, that is, existing design data, manages and operates.

  The design support apparatus 100 of the present embodiment is communicably connected to the person-in-charge terminal 200 operated by the person in charge of design work via the network 10. The person-in-charge terminal 200 displays existing design data output as a processing result by the design support apparatus 100 according to the present embodiment. On the other hand, the person in charge judges the existing design data displayed on the person in charge terminal 200 with regard to the building to be designed next, that is, the existing design data judged to be divertable and suitable. Perform design work using. That is, the person in charge in this case appropriately utilizes the presented existing design data as a basis for designing without making a design from scratch, and greatly improves the work efficiency.

  In addition, regarding each of the existing design data displayed on the person-in-charge terminal 200, the result of determination as to whether the person in charge can be diverted as described above is returned from the person-in-charge terminal 200 to the design support device 100. The result of this determination is fed back to the machine learning algorithm in the design support apparatus 100.

  The machine learning algorithm 110 (described later) reflects the determination result fed back in this way in similarity determination (determination of existing design data having similarity with respect to the presence or absence of a predetermined facility in the building to be designed) and the subsequent determination. Let Specifically, learning is carried out to strengthen the affirmation about the relationship between the existing design data that the person in charge has determined as “reusable”, the presence or absence of the predetermined facility in the building to be designed at that time, and the arrangement position. On the other hand, learning is carried out to strengthen the negative about the relationship between the existing design data determined by the person in charge as described above to be "not divertable" and the presence / absence of the predetermined facility and the arrangement position in the building to be designed.

  As a process of reflecting such feedback contents in the algorithm, that is, a machine learning method, an existing one such as deep learning by a convolutional neural network may be appropriately used.

The machine learning algorithm 110 according to the present embodiment performs machine learning about the relationship between the feature value such as the presence or absence of the predetermined facility in the building and the arrangement position thereof and the success or failure of the feedback value, that is, the similarity. It will be done.
――― Hardware configuration ―――

  Subsequently, hardware configurations of the design support apparatus 100 and the person-in-charge terminal 200 will be described. First, the hardware configuration of the design support apparatus 100 is as follows.

  FIG. 2 is a diagram showing an example of the hardware configuration of the design support apparatus 100 in the present embodiment. The design support apparatus 100 reads out to the memory 103 a storage device 101 configured by an appropriate non-volatile storage element such as a hard disk drive, a memory 103 configured by a volatile storage element such as a RAM, and the program 102 stored in the storage device 101. And a central processing unit 104 such as a CPU that performs various determinations, calculations, and control processing as well as general control of the apparatus itself, and a communication unit 105 that is connected to the network 10 and is responsible for communication processing with the person-in-charge terminal 200. Prepare.

In the storage device 101, at least a program 102 for mounting a function necessary as the design support apparatus 100 of the present embodiment and a design data DB 125 for storing existing design data are stored. Details of the data configuration of the design data DB 125 will be described later. The program 102 also includes a machine learning algorithm 110.
The hardware configuration of the person-in-charge terminal 200 is as follows. FIG. 3 is a diagram showing an example of the hardware configuration of the person-in-charge terminal 200 in the present embodiment.

The person-in-charge terminal 200 reads out to the memory 203 the storage device 201 configured by an appropriate non-volatile storage element such as a hard disk drive, the memory 203 configured by a volatile storage element such as a RAM, and the program 202 stored in the storage device 201. An operation device that performs overall control of the device itself and performs various determinations, calculations, and control processing, as well as an operation device 204 such as a CPU, and an input device that receives key inputs and the like of persons in charge of design work (for example, CAD operators) 205, a display device 206 such as a display for displaying processing data, and a communication device 207 that is connected to the network 10 and performs communication processing with the above-described design support apparatus 100. It is assumed that the CAD program 210 is included in the above-mentioned program 202. The person in charge launches the CAD program 210 to carry out a design job of creating a design drawing.
――― Data structure example ―――

  Next, an example of a data structure in a database used by the design support apparatus 100 of the present embodiment will be described. FIG. 4 is a diagram showing a configuration example of the design data DB 125 held by the design support apparatus 100 of the present embodiment.

  The design data DB 125 is, for example, a database storing design data that the designer has designed in the past with respect to various buildings, that is, existing design data. Naturally, the existing design data in the design data DB 125 is increased each time the daily design work is performed.

  The specific data structure of the design data DB 125 uses the design data ID uniquely identifying the existing design data as a key, the design date, information of the building to be designed, a design data file that is an entity of the existing design data, etc. It is a set of records in which each value is associated.

  The existing design data is not particularly limited as long as it is design drawing data and design document data based thereon. As a specific example, so-called structural drawing, design drawing, equipment drawing, housing drawing, etc. can be assumed. However, in any case, it is assumed that the information on the piping structure in the building to be designed is included. As this information, each information of the presence or absence of the header for water supply / drainage in the said building and its arrangement position is assumed.

Further, in the present embodiment, as an example, design data created by the designer in the past is assumed as the above-mentioned existing design data, but design data provided by another company is stored in the design data DB 125 as existing design data. It is also possible to assume a form to be used, and the source of design data is not particularly limited.
――― Example flow ―― ―

  Hereinafter, an operation example of the design support apparatus 100 in the present embodiment will be described based on the drawings. FIG. 5 is a diagram showing an example of the flow of the design support method in the present embodiment. The various operations described below are realized by each program that the design support apparatus 100 reads and executes in its memory. The program is composed of code for performing various operations described below.

  As an example of the situation assumed here, it is assumed that a designer is commissioned by a construction company for design work on a piping structure of a certain building. In addition, on the occasion of this consignment, it is assumed that necessary information such as a floor plan, a housing diagram, and an equipment diagram is given as design information on the piping structure to be designed. The design support apparatus 100 holds such design information in the storage device 101.

  In this case, it is assumed that the person in charge of the designer mentioned above operates the person in charge terminal 200 to designate the above-mentioned design information with respect to the piping structure of the building to be designed this time (see FIG. 6). The person-in-charge terminal 200 that has received this designation notifies the design support device 100 that the designation has been made (s100).

  On the other hand, the design support apparatus 100 reads out the design information that is the target of the designation from the storage device 101, and extracts each piece of information on the presence / absence of the predetermined facility and the arrangement position in the piping structure from the design information (s101). The information extracted here is each information of the presence / absence of the header for water supply and drainage and the arrangement position in the piping structure to be designed.

  Subsequently, the design support device 100 performs design data of the existing design data having similarity with respect to each information of presence / absence and arrangement position of the header extracted in s101 described above according to the machine learning algorithm 110 performing learning regarding determination of the similarity. It specifies in DB125 (s102).

  In this specification, the design support apparatus 100 specifies a plurality of existing design data having similarities with respect to the presence or absence of the header and the arrangement position but different in the pipe route among the design contents of the pipe structure.

The presence or absence of the header in the building and the placement position tend to indicate the layout of the floor and the configuration of the foundation in the building. Therefore, there is a high possibility that the buildings having the similar presence or absence and the arrangement position of the header in the piping structure are the buildings having the similar layout and foundation configuration. Therefore, if some of the design contents of the piping route performed in the past regarding such similar buildings are specified, it becomes easy to search the pattern of the piping route which is easy to divert from the design object of this time among them.
Next, the design support device 100 transmits the plurality of pieces of existing design data identified in s102 to the person-in-charge terminal 200 via the network 10 (s103).

  On the other hand, the person-in-charge terminal 200 described above receives a plurality of existing design data from the design support apparatus 100, and causes the display device 206 to display this (see FIG. 7) (s104).

  In addition, the person-in-charge terminal 200 uses the input device 205 for the result of the determination of the suitability of the design contents of the piping route for the piping structure in the building to be designed for each of the plurality of existing design data displayed in s104. Acquire from the person in charge (see FIG. 8) (s105).

The person in charge in this case refers to, for example, the relationship with the building foundation with respect to each of a plurality of existing design data being displayed on the person-in-charge terminal 200 while referring to floor plans and the like of the building to be designed this time. The above-mentioned conformity judgment is performed in terms of whether or not it is a piping route.
Also, the person-in-charge terminal 200 in this case transmits the result of the compatibility determination acquired in s105 to the design support apparatus 100 via the network 10 (s106).

  On the other hand, the design support apparatus 100 receives the result of the compatibility determination transmitted by the person-in-charge terminal 200 as described above, feeds it back to the machine learning algorithm 110 (s107), and ends the process.

  The machine learning algorithm 110 in this case, as already described, determines the feedback result in the same manner as the similarity determination (the existing design data having similarity with respect to the presence or absence of the header and the arrangement position in the piping structure to be designed) Judgment) shall be reflected. Specifically, learning is carried out to strengthen the affirmation about the relationship between the existing design data judged as “compatible” by the person in charge described above and the presence or absence of the header and the arrangement position in the piping structure to be designed at that time. On the other hand, learning is carried out to strengthen the negative about the relationship between the existing design data judged by the person in charge as described above to be "not suitable" and the presence / absence and the arrangement position of the header in the piping structure to be designed.

  Thereafter, the accuracy of similarity determination in the machine learning algorithm 110 will continue to improve each time the above-described design work is performed. Therefore, the probability that the existing design data distributed by the design support apparatus 100 to the person-in-charge terminal 200 is also preferable to the person in charge is increased.

As mentioned above, although the best mode etc. for carrying out the present invention were concretely explained, the present invention is not limited to this, and can be variously changed in the range which does not deviate from the gist.
According to the present embodiment, it is possible to reuse existing design data to reduce human load and to make the entire design work more efficient.

10 Network 100 Design Support Device 101 Storage Device 102 Program 103 Memory 104 Arithmetic Device 105 Communication Device 110 Machine Learning Algorithm 125 Design Data DB
200 person in charge terminal 201 memory device 202 program 203 memory 204 arithmetic device 205 input device 206 display device 207 communication device 210 CAD program

Claims (8)

A storage device holding a database of existing design data on a building;
From the information for design about the building to be designed, each information of the presence / absence of the predetermined equipment in the building and the arrangement position is extracted, and the existing design data having similarity with respect to each information is learned about the judgment of the similarity An arithmetic unit that specifies the machine learning algorithm to be performed and outputs the existing design data to a predetermined unit;
A design support apparatus characterized by including. The arithmetic device is
In specifying the existing design data having the similarity, a plurality of existing design data having similarity with respect to each of the pieces of information but differing with respect to predetermined items in the design content is specified, and the plurality of existing design data is used as the predetermined device To output
The design support apparatus according to claim 1, characterized in that: The arithmetic device is
Regarding each of a plurality of existing design data outputted to the predetermined device, the result of the judgment of the conformity of the design contents of the predetermined item with respect to the building to be designed is obtained from the predetermined user via the predetermined interface It further executes a process of feeding back the result of the determination to the machine learning algorithm.
The design support apparatus according to claim 2, characterized in that: The storage device is
In the database, existing design data of a piping structure installed in a building is stored as the existing design data regarding the building,
The arithmetic device is
From the design information on the building to be designed, the respective information of the presence / absence and the arrangement position of the header in the piping structure of the building is extracted, and the existing design data of the piping structure having similarities regarding the respective information It is specified by a machine learning algorithm that performs learning regarding determination of similarity, and the existing design data of the piping structure is output to a predetermined device.
The design support apparatus according to any one of claims 1 to 3, characterized in that: An information processing apparatus comprising a storage device holding a database of existing design data on a building,
From the information for design about the building to be designed, each information of the presence / absence of the predetermined equipment in the building and the arrangement position is extracted, and the existing design data having similarity with respect to each information is learned about the judgment of the similarity Identify by the machine learning algorithm to be performed, and output the existing design data to a predetermined device,
Design support method characterized by The information processing apparatus
In specifying the existing design data having the similarity, a plurality of existing design data having similarity with respect to each of the pieces of information but differing with respect to predetermined items in the design content is specified, and the plurality of existing design data is used as the predetermined device Output,
The design support method according to claim 5, characterized in that: The information processing apparatus
Regarding each of a plurality of existing design data outputted to the predetermined device, the result of the judgment of the conformity of the design contents of the predetermined item with respect to the building to be designed is obtained from the predetermined user via the predetermined interface Further executing a process of feeding back the result of the determination to the machine learning algorithm;
The design support method according to claim 6, characterized in that: The information processing apparatus
In the storage device, in the database, existing design data of a piping structure installed in the building is stored as the existing design data regarding the building,
From the design information on the building to be designed, the respective information of the presence / absence and the arrangement position of the header in the piping structure of the building is extracted, and the existing design data of the piping structure having similarities regarding the respective information Specify by a machine learning algorithm that performs learning related to similarity determination, and output existing design data of the piping structure to a predetermined device.
The design support method according to any one of claims 5 to 7, characterized in that:

Images