Drawing search system, drawing search method, and program

Abstract

When searching based on the similarity of multiple indicators, it is possible to check the similarity of each indicator. [Solution] The drawing search system includes a search unit that searches for similar drawing information, which is existing drawing information similar to drawing information, from among multiple existing drawing information registered in an existing drawing database based on a similarity calculated from a detailed similarity, which is the similarity between drawing information and similar drawing information for each of multiple indicators, and a similarity display unit that displays the detailed similarity used by the search unit to search for similar drawing information together with the similar drawing information.

Description

[Technical Field]

[0001] The present invention relates to a drawing search system, a drawing search method, and a program. [Background technology]

[0002] Conventionally, there are known technologies that, when a newly ordered drawing data is uploaded, search a drawing database for drawing data from past orders that are similar in shape to the newly ordered drawing (for example, Patent Document 1, Non-Patent Documents 1 and 2). These search technologies use AI to search for drawing data with similar shapes. Furthermore, by registering the ordering company name, size, material, construction method, and other information in the drawing database along with the drawings, these search technologies can search not only for similar shapes of drawings but also for size, material, and other information based on text. Furthermore, some of these search technologies have the function of searching the drawing database for estimates associated with the drawing data along with the drawing data. Furthermore, some of these search technologies display the similarity used in the search as a numerical value when displaying the search results. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-022330 [Non-patent literature]

[0004] [Non-Patent Document 1] [online], REVOX Co., Ltd., [Retrieved July 19, 2024], Internet〈URL:https: / / www.sellbot.jp〉 [Non-patent document 2] [online], Technoa Co., Ltd., [Retrieved July 19, 2024], Internet〈URL: https: / / www.techs-s.com / product / ai-ruiji〉 Summary of the Invention [Problem to be solved by the invention]

[0005] Conventional search technologies allow users to recognize the degree of similarity between drawings by displaying the degree of similarity as a numerical value. However, when a search is performed based on multiple indices, it is not clear which search indices are highly evaluated among the multiple search results. This can be time-consuming for users, as they are unsure which of the multiple search results they should prioritize. When a search is performed based on the similarity for each of a plurality of indices, it is required to be able to check the similarity for each indices.

[0006] The present invention has been made in consideration of the above points, and provides a drawing search system, drawing search method, and program that can check the similarity for each index when performing a search based on the similarity for each of multiple indexes. [Means for solving the problem]

[0007] (1) One aspect of the present invention is a drawing search system that includes a search unit that searches for similar drawing information, which is existing drawing information similar to drawing information, from among multiple existing drawing information registered in an existing drawing database based on a similarity calculated from a detailed similarity, which is the similarity between the drawing information and the similar drawing information for each of multiple indicators, and a similarity display unit that displays the detailed similarity used by the search unit to search for the similar drawing information together with the similar drawing information. (2) In one aspect of the present invention, in the drawing search system described in (1) above, the similarity display unit displays the detailed similarity used by the search unit to search for the similar drawing information, and the similarity together with the similar drawing information. (3) In one aspect of the present invention, in the drawing search system described in (1) above, the similarity display unit displays the detailed similarity used by the search unit to search for the similar drawing information as a value measured on a predetermined scale or in a predetermined display format. (4) One aspect of the present invention is a drawing search method having a search step for searching for similar drawing information, which is existing drawing information similar to drawing information, from among multiple existing drawing information registered in an existing drawing database, based on a similarity calculated from a detailed similarity, which is the similarity between the drawing information and the similar drawing information for each of multiple indicators, and a similarity display step for displaying the detailed similarity used in the search step to search for the similar drawing information together with the similar drawing information. (5) One aspect of the present invention is a program for causing a computer to execute a search step for searching for similar drawing information, which is existing drawing information similar to drawing information, from among multiple existing drawing information registered in an existing drawing database, based on a similarity calculated from a detailed similarity, which is the similarity between the drawing information and the similar drawing information for each of multiple indicators, and a similarity display step for displaying the detailed similarity used in the search step to search for the similar drawing information together with the similar drawing information. [Effects of the Invention]

[0008] According to the present invention, when a search is performed based on the similarity for each of a plurality of indices, the similarity for each indices can be confirmed. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram showing an example of an outline of the configuration of a drawing search system 1 according to an embodiment of the present invention. [Figure 2] 1 is a diagram illustrating an example of a functional configuration of a drawing search system 1 according to an embodiment of the present invention. [Figure 3] FIG. 2 is a diagram showing an example of drawing data Z1 according to an embodiment of the present invention. [Figure 4] FIG. 10 is a diagram showing an example of the flow of a search process according to an embodiment of the present invention. [Figure 5] FIG. 10 is a diagram showing an example of a search result screen according to an embodiment of the present invention. [Figure 6] FIG. 10 is a diagram showing an example of a priority setting screen according to the embodiment of the present invention. [Figure 7] FIG. 10 is a diagram showing an example of a priority setting screen according to the embodiment of the present invention. [Figure 8] FIG. 10 is a diagram showing an example of a search result screen according to an embodiment of the present invention. [Figure 9] FIG. 10 is a diagram showing an example of a search result screen according to an embodiment of the present invention. [Figure 10] FIG. 10 is a diagram showing an example of a flow of a quotation creation process according to an embodiment of the present invention. [Figure 11] FIG. 10 is a diagram showing an example of a search result screen according to an embodiment of the present invention. [Figure 12] 10 shows an example of an estimate display screen according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0010] (Embodiment) Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a diagram showing an example of the outline of the configuration of a drawing search system 1 according to this embodiment. As an example, the drawing search system 1 is a system for searching drawings that are similar to a newly ordered drawing from drawings that have been ordered in the past. As an example, the drawings are drawings of parts or products. A user uses the order prices of similar drawings obtained by the search as a reference when determining the estimate price for the newly ordered drawing. The user may be, for example, an employee of a parts or product manufacturing plant.

[0011] A user uses a terminal device 2 to send new drawing information N1 to the drawing search system 1. The drawing search system 1 searches for similar drawing information similar to the new drawing information N1 from among multiple existing drawing information registered in the existing drawing database A1 based on multiple search conditions. In the drawing search system 1, priorities are set for multiple search conditions. The user sets desired priorities for multiple search conditions according to the items (indicators) that the user wants to emphasize when searching for drawings.

[0012] An example of the drawing search system 1 is a cloud server. The drawing search system 1 may be realized as a single computer or as a virtual server.

[0013] The terminal device 2 is, for example, a personal computer (PC). Note that the terminal device 2 may also be a smartphone, a tablet terminal, or the like.

[0014] 2 is a diagram showing an example of the functional configuration of the drawing search system 1 according to this embodiment. The drawing search system 1 includes a drawing information receiving unit 10, a search unit 11, a drawing feature information generating unit 12, a priority setting unit 13, a similarity display unit 14, an estimate creating unit 15, a drawing data managing unit 16, an operation receiving unit 17, an output unit 18, and a storage unit 19.

[0015] Of these functional units, all but the storage unit 19 are realized by, for example, a CPU (Central Processing Unit) loading a program from a ROM (Read Only Memory) into a RAM (Random Access Memory) and executing processing in accordance with the program. The ROM and RAM are included in the storage unit 19. These functional units may be provided in a single computer or may be distributed across multiple servers.

[0016] The drawing information receiving unit 10 receives drawing information. The drawing information receiving unit 10 receives drawing information transmitted from the terminal device 2 to the drawing search system 1. In the following description, the drawing information received by the drawing information receiving unit 10 may be referred to as searched drawing information. In one example of this embodiment, the searched drawing information is new drawing information N1, which is drawing information about a new drawing.

[0017] The drawing information includes, for example, drawing data, drawing-related information, a drawing ID, and linking information. Note that one or more of the drawing-related information and the linking information may be omitted from the drawing information.

[0018] An example of drawing data is shown in Figure 3. The drawing data Z1 shown in Figure 3 includes shape data Z11, processing data Z12, processing data Z13, and character / numeric data Z14. The shape data Z11 indicates the outline of the part drawing. The processing data Z12 and processing data Z13 indicate drawing symbols. The character / numeric data Z14 indicates the text for the drawing number "0000-0000," the client company name "XX Co., Ltd.", and the drawing name "Bracket."

[0019] Drawing-related information is various information associated with a drawing. Drawing-related information is text information. Examples of drawing-related information include the drawing name, drawing number, supplier name, factory name, registration date / edit date, deadline, current status, material, surface treatment, plate thickness, dimensional information, price information, person in charge, and memo information. The deadline may be, for example, a quotation or a parts delivery deadline. The current status indicates the progress of the drawing, such as "not yet created," "in creation," or "completed." Dimensional information indicates specific dimensions such as maximum dimensions, depth, width, and height. Drawing-related information may also include information extracted from drawing data. For example, dimensional information is extracted from processing data Z12 and processing data Z13, as shown in Figure 3. The drawing name, drawing number, and supplier name are extracted from character-numeric data Z14, as shown in Figure 3.

[0020] The link information indicates the relationship between drawings, for example, indicating that a drawing was created based on another drawing.

[0021] As mentioned above, drawing information includes one or more of the following: information indicating the shape of the drawing, information indicating features other than the shape of the drawing, such as the name of the drawing, the name of the customer, dimensions, processing information, drawing number, etc. Therefore, drawing information includes one or more of information indicating the shape of the drawing, information indicating one or more of the name and material of the drawing, information indicating the dimensions of the drawing, and information indicating the processing content of the drawing.

[0022] Note that the user may add other types of information to the drawing information in addition to the above-mentioned information. Furthermore, the drawing information may be deleted from the drawing information. That is, the drawing information receiving unit 10 may be designed so that it can add and register drawing information for the drawing later, or delete already registered drawing information, not just when it first receives new drawing information N1.

[0023] The search unit 11 searches for similar drawing information from among multiple existing drawing information registered in the existing drawing database A1 based on multiple search conditions with set priorities. Each of the multiple search conditions is specified by multiple indicators. The existing drawing information is, for example, drawing information about drawings that have been ordered in the past. The similar drawing information is existing drawing information that is similar to the searched drawing information. The search unit 11 searches for similar drawing information based on similarity. The similarity is calculated from detailed similarity, which is the similarity between the drawing information and similar drawing information for each of the multiple indicators. In calculating the similarity, a weight is set for the detailed similarity according to the priority set for each of the multiple search conditions.

[0024] The drawing characteristic information generating unit 12 generates drawing characteristic information from the drawing information.

[0025] The drawing feature information includes shape information, processing information, and character / numeric information contained in the drawing information, as well as feature amount information generated by extracting feature portions from these pieces of information. The feature amount information will be described in detail later. Shape information is information that indicates the shape of a drawing. The shape of a drawing includes one or more of the external shape of the drawing or the internal shape (also referred to as "internal shape"). One example of shape information is image information (information in pixel units) that indicates the figure. Shape data Z11 shown in FIG. 3 is an example of shape information. Note that shape information may also be text information that indicates the figure.

[0026] The processing information is information that indicates the processing content, and is indicated by drawing symbols or the like included in the drawing. The text-numeric information is information indicating text or numbers written in the drawing.

[0027] The feature information indicates feature amounts extracted based on the shape information, processing information, and character-numeric information. The feature information extracted from the shape information includes, for example, information about the edges of the drawing. The feature information extracted from the processing information includes, for example, the type of manufacturing symbol, information about the positional relationship between the manufacturing symbols, and information about the positional relationship between the manufacturing symbol and the drawing.

[0028] The user may specify which type of information among the drawing information is to be used in the process of generating drawing feature information from the drawing information by the drawing feature information generating unit 12. For example, if the user specifies shape information and processing information among the drawing information, the drawing feature information generating unit 12 generates the drawing feature information from the shape information and processing information, and does not use character-numeric information in generating the drawing feature information.

[0029] The priority setting unit 13 sets priorities for a plurality of indices used in a drawing search, that is, a plurality of search conditions.

[0030] In one example of this embodiment, the multiple indices used in drawing search can be selected from up to two of the four indices: shape similarity, name similarity, dimension similarity, and processing content similarity. That is, the user selects up to two of the four presented search conditions as search conditions. Shape similarity means that the shapes (external or internal shapes) of the drawings are similar. Name similarity means that text information such as the drawing name or material is similar. Dimension similarity means that dimensional information is similar. Processing content similarity means that the type of processing content or the number of processing content is similar.

[0031] The types of selectable indices are not limited to the four indices described above. The multiple indices may include accuracy similarity and internal shape similarity. Accuracy similarity refers to a similarity in machining accuracy. Internal shape similarity refers to a similarity in the machined shape of the inside of a part, rather than the external shape. The number of selectable indices may be two, three, five or more.

[0032] The similarity display unit 14 displays "detailed similarity" and "overall similarity" for multiple indices selected as search conditions used by the search unit 11 for the search, along with similar drawing information. "Detailed similarity" is the similarity derived for each index (i.e., search condition). "Overall similarity" is the similarity derived by aggregating the detailed similarities derived for each index.

[0033] The estimate creation unit 15 refers to the drawing information of the top n (n is a predetermined natural number) similar drawing information with the highest similarity to the drawing information from the similar drawing information searched by the search unit 11, and creates an estimate for the searched drawing information based on the estimated amount and drawing feature information contained therein.

[0034] The drawing data management unit 16 manages the existing drawing database A1 and drawing information. For example, the drawing data management unit 16 registers new existing drawing information in the existing drawing database A1 based on a user operation. Alternatively, the drawing data management unit 16 deletes existing drawing information from the existing drawing database A1 based on a user operation. The drawing data management unit 16 adds a new type of information as drawing information based on a user operation. Furthermore, the drawing data management unit 16 deletes a specified type of information from the drawing information based on a user operation. The drawing data management unit 16 may also manage the types of drawing information used by the drawing feature information generation unit 12 in the process of generating drawing feature information.

[0035] The operation reception unit 17 receives various operations from the user. The operations are performed via the terminal device 2. The terminal device 2 includes, for example, a mouse, a keyboard, or a touch panel.

[0036] The output unit 18 outputs various types of information to the terminal device 2. The information output by the output unit 18 is displayed, for example, on a display unit provided in the terminal device 2. The display unit includes, for example, a liquid crystal display or an organic electroluminescence (EL: Electro-Luminescence) display.

[0037] The storage unit 19 stores various types of information. For example, the storage unit 19 stores an existing drawing database A1. The existing drawing database A1 stores existing drawing information and drawing feature information generated from the existing drawing information in association with each other. The storage unit 19 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device.

[0038] Next, a description will be given of a search process using the drawing search system 1. Fig. 4 is a diagram showing an example of the flow of the search process according to this embodiment.

[0039] Step S10: The drawing information receiving unit 10 receives the searched drawing information. The user inputs new drawing information N1 using the terminal device 2. The terminal device 2 transmits the input new drawing information N1 to the drawing search system 1. The drawing information receiving unit 10 receives the new drawing information N1 received from the terminal device 2 as searched drawing information.

[0040] Here, the drawing characteristic information generating unit 12 generates drawing characteristic information from the drawing information, and stores the generated drawing characteristic information in the storage unit 19 in association with the retrieved drawing information.

[0041] Step S20: The search unit 11 displays a search screen on the terminal device 2. The search unit 11 causes the output unit 18 to transmit data showing the search screen to the terminal device 2. The terminal device 2 displays the search screen based on the data.

[0042] Fig. 5 is a diagram showing an example of a search screen according to this embodiment. A search screen B1 shown in Fig. 5 includes a search method designation button, a priority setting form, and a narrowed search condition selection form.

[0043] The search method specification buttons consist of a similarity search button C1 and a refined search button C2. The similarity search button C1 is a button for selecting a similarity search. The refined search button C2 is a button for selecting a refined search. When the similarity search is selected, the user sets the priority of multiple similarity search conditions at least for the similarity search. When the similarity search is selected, it is possible to further refine the search results detected by combining similarity searches. On the other hand, when the refined search is selected, only a refined search is performed, and no similarity search is performed. The following describes the case where similarity search is selected.

[0044] The similarity search conditions are search conditions for a search based on similarity to drawing information (similarity search). The similarity search conditions are an example of first search conditions. The refined search conditions are search conditions for a search based on specified information that specifies existing drawing information (refined search). The refined search conditions are an example of second search conditions. Drawing-related information is an example of specified information.

[0045] Furthermore, in a similarity search, at least one search condition for the similarity search is set. That is, when a similarity search is selected as a search method on the search screen B1 of FIG. 5, no similarity search condition is set and only a refined search condition is not set. Therefore, when a first search mode (for example, a similarity search) is selected, the search unit 11 searches for similar drawing information based on at least the first search condition (for example, a similarity search condition). On the other hand, when a second search mode (for example, a refined search) is selected, the search unit 11 searches for similar drawing information based only on the second search condition (for example, a refined search condition).

[0046] Step S30: The priority setting unit 13 sets priorities for the similarity judgment indices used in the similarity search. As described above, in this embodiment, the user is allowed to select up to two of the four indices as similarity judgment indices. The priority setting form D1 is a form for selecting the similarity judgment indices with the highest priority. The priority setting form D2 is a form for selecting the similarity judgment indices with the second highest priority.

[0047] The user sets priorities for multiple similarity search conditions (similarity determination indices) by performing predetermined operations on the priority setting form D1 and the priority setting form D2. The priority setting form D1 and the priority setting form D2 are, for example, pull-down menus. In the example shown in FIG. 5, a shape similarity index is selected as the search condition with the highest priority, and a processing content similarity index is selected as the search condition with the second highest priority. Note that only the similarity determination index with the highest priority may be selected, and the similarity determination index with the second highest priority may not be selected.

[0048] If the user wishes to further refine the search in combination with the similarity search, the user selects one or more refined search conditions by performing a predetermined operation on the refined search condition specification form E1. The refined search conditions are set to one or more predetermined items included in the drawing-related information of the drawing information. One or more refined search conditions are selected from the one or more items. Note that if the user selects "similarity search" as the search mode, it is not necessary to set conditions for the refined search.

[0049] The user makes the necessary settings for the selected search mode (first search mode or second search mode, i.e., similarity search or refined search in this embodiment). In the example shown in FIG. 5, the similarity search mode (first search mode) is selected as the search mode, so once the priorities for the similarity judgment indices to be used in the similarity search are set, the user can press the search button F1 included in the search screen B1 via the terminal device 2. The user inputs search conditions required for the refined search (second search mode) as necessary and presses the search button F1. When the search button F1 is pressed, the drawing search system 1 starts a search.

[0050] Step S40: The search unit 11 searches for similar drawing information from among the multiple existing drawing information registered in the existing drawing database A1 based on multiple similarity search conditions (similarity determination indexes) with priorities set. Here, the search unit 11 calculates the similarity based on the detailed similarity. As an example, the search unit 11 calculates the similarity based on equation (1).

[0051]

number

[0052] In equation (1), the variable Similarity indicates the degree of similarity. sim indicates the detailed similarity to the shape similarity. sha indicates the weight given to shape similarity. sim indicates the detailed similarity to the name similarity. name indicates the weight for name similarity. sim indicates the degree of detail similarity relative to the size similarity. size indicates the weight given to the dimensional similarity. sim indicates the detailed similarity for the similarity of the processed content. process indicates the weight for the similarity of the processed content.

[0053] The detailed similarity may be calculated using any known method for calculating similarity between features. For example, the detailed similarity is cosine similarity. In this case, the detailed similarity is calculated by extracting feature vectors from the features of each index and comparing the inner products between the feature vectors.

[0054] These weights (weights w sha , weight w name , weight w size , weight w process ) is set to a value according to the priority set for the search condition corresponding to that weight. A weight for a search condition with the highest priority is set to a predetermined value that is greater than the weight for a search condition with the second highest priority. No priority is set for search conditions that are not selected, and the weight for that search condition is set to 0.

[0055] In addition, the right side of formula (1) is multiplied by a factor corresponding to the refined search condition. For ease of understanding, formula (1) uses, as an example, the factor Corp, which corresponds to the refined search condition for the company name. sim wcorp (where "wcorp" is "w corp " means the same value as "). Power exponent w corp The value is set to 1 if "Company Name" is selected as a narrowed search condition, and 0 if "Company Name" is not selected. Corp sim When "Company name" is selected as the narrowed search condition, the value is set to 1 if the "Company name" matches between the searched drawing information and the similar drawing information, and to 0 if the "Company name" does not match between the searched drawing information and the similar drawing information. Although omitted in formula (1), factors for other refinement search conditions are also included in the factor Corp. sim wcorp Similarly, it is multiplied on the right side of equation (1).

[0056] In the present embodiment, an example in which first and second priority levels are set has been described, but the present invention is not limited to this. In addition to first and second priority levels, third and subsequent priority levels may also be set.

[0057] 5, the search screen B1 has been described as an example in which the drawing-related information is used as a narrowing search criterion and only the drawing characteristic information is used as a search criterion for the similarity search, but this is not limiting. The drawing-related information may include drawing-related information that is used as a search criterion for the similarity search. When the drawing-related information is used as a search criterion for the similarity search, detailed similarity is calculated not only for the drawing characteristic information but also for the drawing-related information.

[0058] For example, all of the drawing-related information used as search conditions among the drawing-related information may be used as search conditions for the similarity search. In this case, the refined search condition specification form E1 is not displayed on the search screen B1 shown in Figure 5, and all search conditions are selected using the priority setting form. Note that since the drawing-related information is text information, the similarity between texts is calculated as the detailed similarity for the drawing-related information.

[0059] 6 and 7 show other examples of the screen for setting the priority order. In the priority setting screen B2 shown in FIG. 6, the priority is set as a priority level using a slide bar. Weights are set for multiple search conditions according to the position of the set slide bar. In the search screen B1 shown in FIG. 5, when the priority is set using the priority setting form D1 and the priority setting form D2, a predetermined weight according to the set priority is set for the search conditions for similarity search. On the other hand, in the example of the priority setting screen B2 shown in FIG. 6, the weight is changed according to the set priority level. The user can set the priority level for each search condition while visually checking the priority level using the slide bar.

[0060] In the priority setting screen B3 shown in FIG. 7, the priority is set using a form for inputting a numerical value as the priority level. Weights are set for multiple search conditions according to the input numerical value. In the example of the priority setting screen B3 shown in FIG. 7, the weights are also changed according to the set priority level. The user can set the priority level for each search condition while checking the priority level as a numerical value. When the priority level is input as text (indicating a numerical value), it is preferable that the highest value of the numerical value be displayed. For example, if the highest value is "10," the message displayed is "Please rate the priority level on a scale of 10 (higher priorities have higher numerical values)."

[0061] In this embodiment, even if the weights corresponding to the respective search conditions are equal to each other, it is described that the same priority is set to each of the search conditions.

[0062] Note that the priority setting screen B2 shown in Fig. 6 and the priority setting screen B3 shown in Fig. 7 may be displayed in place of the priority setting form D1 and the priority setting form D2, respectively, in the area where the priority setting form D1 and the priority setting form D2 shown in Fig. 5 are displayed. Alternatively, a button displaying "Set priority" or the like may be provided in the area, and when the button is pressed, the priority setting screen B2 or the priority setting screen B3 may be displayed as a pop-up.

[0063] Step S50: The search unit 11 displays the search results on the terminal device 2. The search unit 11 causes the output unit 18 to transmit the search results to the terminal device 2. The terminal device 2 displays a search result screen showing the search results based on the search results. Here, the similarity display unit 14 displays the similarity in the search results. With this, the drawing search system 1 ends the search process.

[0064] 8 is a diagram showing an example of a search result screen according to this embodiment. On the search result screen B4 shown in FIG. 8, searched drawing information G1 and search result H1 are displayed.

[0065] The searched drawing information G1 is the drawing information used in the search. For example, the searched drawing information G1 displays the drawing data (e.g., thumbnail) showing the drawing, the drawing ID, the drawing name among the drawing-related information, and the client company name.

[0066] The search results H1 display the top n (n is a predetermined natural number) of similar drawing information retrieved by the search unit 11, starting from the most similar. However, if search conditions for a refined search are set in addition to the similarity search, the number of search results displayed decreases because the n most similar results are further refined. For example, in FIG. 8, the search result H1 displays at the top that "4 results" were found. The search results H1 display, as an example, the similarity, drawing data (e.g., thumbnails), drawing ID, and drawing-related information such as the drawing name, client company name, and registration date of the drawing. Note that the search results may display drawing-related information other than that shown in FIG. 8.

[0067] When the similarity search mode is specified as the search mode, the "similarity" is displayed numerically for similar drawings displayed as search results. By specifically displaying the similarity, the user can specifically grasp the degree of similarity between the searched drawing and the similar drawing. Furthermore, when a search is performed by similarity, the search results are generally displayed in descending order of similarity. However, when the similarity is the same, the results are often displayed in descending order of similarity, for example, in descending order of the data registration date. However, if the similarity is not displayed numerically, even if drawings actually have the same similarity, the user may mistakenly believe that the drawing listed closer to the top has the highest similarity, which may result in an incorrect judgment. In this embodiment, the similarity is displayed numerically, allowing the user to accurately judge the similarity. Furthermore, when multiple similarity judgment indices are set, the similarity is displayed for each specified indices, allowing the user to examine the search results in more detail.

[0068] When the refined search mode is selected as the search mode, the search result H1 shown in Fig. 8 may not display numerical values ​​based on percentages of similarity. Also, although not shown, when a similar drawing displayed as a search result is selected, the screen transitions to a detailed display screen related to the drawing. The detailed display screen displays various drawing information (drawing-related information, etc.) related to the similar drawing that is not displayed in the search result H1.

[0069] Fig. 9 shows another example of a search result screen. When the similarity search mode is selected, the search result screen B5 shown in Fig. 9 is displayed as the search result display screen instead of the search result screen B4 in Fig. 8. The search result screen B5 shown in Fig. 9 displays searched drawing information G2, detailed search results K1, and detailed search results K2. The display format of the search results may be different between the search result screen B4 and the search result screen B5 depending on the search mode selected on the search screen B1. Specifically, when the similarity search mode is selected, the search result screen B5 is displayed, and when the refined search mode is selected, the search result screen B4 is displayed.

[0070] The detailed search result K1 differs from the search result H1 in that a detailed similarity K12 is displayed. The detailed similarity K12 is also displayed as a percentage together with the similarity K11. The detailed search result K1 shows the detailed similarity K12 for each of shape similarity and name similarity, which were selected as search conditions for the similarity search. The detailed search result K2 displays a similarity K21 and a detailed similarity K22, just like the detailed search result K1.

[0071] As shown on the search result screen B5, the similarity display unit 14 displays the detailed similarity and similarity used by the search unit 11 to search for similar drawing information together with the similar drawing information. Here, the similarity display unit 14 displays the similarity and detailed similarity used by the search unit 11 to search for similar drawing information using values ​​measured on a predetermined scale. The predetermined scale is, for example, a percentage or a predetermined score. Note that, on the search result screens shown in FIGS. 8 and 9, the similar drawing information displayed as a search result may also display the order amount (price information) when an order is placed for the production of parts corresponding to the drawing. Here, the order amount is significantly affected by the number of orders. Therefore, for easier comparison, the order amount may be divided by the number of orders, and the order amount per part may be displayed together with the order amount.

[0072] Here, the variables Similarity and Shape in Equation (1) sim , variable Name sim , variable Size sim , and the variable Process sim are values ​​used in the process of calculating the similarity (such as the distance between feature vectors), and are not necessarily values ​​that are easily understandable to the user, such as percentages. Note that the similarity display unit 14 may directly display each variable in formula (1) as the similarity and detailed similarity.

[0073] By displaying the detailed similarity on the search result screen B5, the user can search for similar drawing information more flexibly than when the detailed similarity is not displayed. The user can check to what extent each of the indicators for which they have set a high priority is reflected in the search results.

[0074] The detailed search result may display drawing-related information other than the drawing-related information included in the detailed search result K1 (or the detailed search result K2). The detailed search result may display, for example, one or more of the name of the responsible department, the name of the person in charge, the surface treatment, the delivery deadline, etc., together with the drawing-related information included in the detailed search result K1. The detailed search result may also display a link to an estimate.

[0075] The display mode of the detailed similarity is not limited to that shown in FIG. 9. The detailed similarity may be displayed using various graphs such as a rating chart, a doughnut chart, or a radar chart. The rating chart is a table consisting of rows and columns in which the detailed similarity is stored for each similar drawing information in the search results, and indicates a table in which the level of similarity is divided into multiple ranks and expressed, for example, by the number of stars (for example, the highest rating rank is "5 stars"). In this way, the similarity display unit 14 may display the detailed similarity used by the search unit 11 to search for similar drawing information using a predetermined display mode.

[0076] Next, the creation of a quotation using the drawing search system 1 will be described. Fig. 10 is a diagram showing an example of the flow of the quotation creation process according to this embodiment. Note that the processes from step S110 to step S150 are the same as the processes from step S10 to step S50 in Fig. 4, and therefore their explanation will be omitted. However, unlike step S50 in Fig. 4, the search result screen displayed in step S150 is a search result screen B6 shown in Fig. 11, instead of the search result screen B4 shown in Fig. 5.

[0077] This is a diagram showing an example of the search result screen shown in Figure 11. The search result screen B6 shown in Figure 11 displays searched drawing information G1, search result H1, estimate button L1, and search again button L2. The estimate button L1 is a button for creating an estimate. The search again button L2 is a button for executing a search again. Furthermore, instead of the search result H1, the search result H2 shown in Figure 9 may be displayed.

[0078] Step S160: The estimate preparation unit 15 determines whether the estimate reference drawing information is included in the search results based on an operation from the user. When the estimate creation button L1 is pressed with one or more of the search results H1 selected as estimate reference drawing information, the estimate preparation unit 15 determines that the estimate reference drawing information is included in the search results. On the search result screen B6 shown in FIG. 11, as an example, the search result H11 is selected as the estimate reference drawing information by selecting the select button L3. On the other hand, when the re-search button L2 is pressed with none of the search results H1 selected as estimate reference drawing information, the estimate preparation unit 15 determines that the estimate reference drawing information is not included in the search results.

[0079] If the estimate preparation unit 15 determines that the estimate reference drawing information is included in the search results (step S160; YES), it executes the process of step S170. On the other hand, if the estimate preparation unit 15 determines that the estimate reference drawing information is not included in the search results (step S160; NO), the search unit 11 executes the process of step S120 again. In other words, if the estimate reference drawing information is not included in the search results, the search screen is displayed on the terminal device 2, and the search is executed again.

[0080] Step S170: The estimate preparation unit 15 reads the price information included in the similar drawing information corresponding to the search result H11 selected as the estimate reference drawing information. In other words, past estimated amounts are read as a reference for the estimate. The estimate preparation unit 15 creates an estimate based on the read price information. When one search result H1 is selected as the estimate reference drawing information in step S160, the estimate preparation unit 15, for example, converts the amount indicated by the read price information into an amount that takes into account the number of orders required to create the current estimate, and creates an estimate. Specifically, the estimate preparation unit 15 divides the amount indicated by the read price information by the number of orders to calculate the order price per part, and creates an estimate using the amount obtained by multiplying the order price per part by the number of orders required to create the current estimate. The estimate preparation unit 15 may prepare an estimate by using the amount indicated by the read amount information as the estimated amount as is.

[0081] In step S160, if the estimate preparation unit 15 determines that the estimate reference drawing information is not included in the search results (if the search again button L2 is pressed when none of the search results H1 have been selected as estimate reference drawing information), an estimate may be prepared from the search results H1 based on predetermined conditions instead of performing the search again. For example, the estimate preparation unit 15 may prepare an estimate using the top 10 similarity results from the search results H1 as estimate reference drawing information. In another example, if a narrowing down condition is also set, the estimate preparation unit 15 may prepare an estimate using all of the search results H1 as the narrowed down search results, or the top to a predetermined rank of the search results H1 as estimate reference drawing information.

[0082] The estimate preparation unit 15 causes the terminal device 2 to display an estimate display screen, which is a screen for displaying the created estimate. FIG. 12 shows an example of the estimate display screen. The estimate display screen B7 shown in FIG. 12 displays the drawing ID, drawing name, client company name, estimated amount, and confirmation button M2. The estimated amount is displayed using amount input form M1. The estimate preparation unit 15 may display the estimate display screen in the format of an estimate.

[0083] In addition, when multiple search results H1 are selected as estimate reference drawing information in step S160, the estimate creation unit 15 reads the price information included in the similar drawing information corresponding to each of the multiple search results H1. As an example, the estimate creation unit 15 creates an estimate by using the average value of the amounts indicated by the multiple price information as the estimated amount. As another example, the estimate creation unit 15 may create an estimate by using a weighted average of the amounts indicated by the multiple price information, calculated according to the similarity to each of the multiple search results H1. In addition, the estimate creation unit 15 may use AI to create an estimate. In addition, when priority items are set for similarity search, an estimate may be created using an estimated amount according to the priority setting.

[0084] At least one of the estimate and price information corresponding to the search result H11 selected as the estimate reference drawing information in step S160 may be loaded, and the estimate and price information may be displayed alongside the corresponding search result H11 on the search result screen B6. The user can consider the estimated price while comparing each of the search results H11, in which past estimates and price information are displayed alongside each other, with the drawings included in the search drawing information G1.

[0085] Step S180: The estimate preparation unit 15 determines whether or not the estimated amount needs to be revised based on an operation from the user. When an operation is performed on the amount input form M1, the estimate preparation unit 15 determines that the estimated amount needs to be revised. When the estimate preparation unit 15 determines that the estimated amount needs to be revised (step S180; YES), it executes the process of step S190.

[0086] On the other hand, when the confirmation button M2 is pressed, the estimate preparation unit 15 determines that no correction to the estimated amount is necessary. When the estimate preparation unit 15 determines that no correction to the estimated amount is necessary (step S180; NO), the output unit 18 outputs the estimate file to the terminal device 2. Note that the estimate file may be stored in the memory unit 19 instead of being output to the terminal device 2 immediately when the confirmation button M2 is pressed. The estimate file stored in the memory unit 19 is downloaded from the terminal device 2 at a time desired by the user.

[0087] Step S190: The estimate preparation unit 15 corrects the estimated amount. The user inputs the corrected estimated amount into the amount input form M1. Thereafter, the estimate preparation unit 15 executes the process of step S180 again. With this, the drawing search system 1 ends the estimate creation process.

[0088] In the present embodiment, an example has been described in which, when priorities are set for multiple search conditions on the search screen B1 (FIG. 5), predetermined weights are set for the multiple search conditions according to the priorities, and similarity is calculated. The predetermined weights may be updated based on feedback from the user. As described above, in the process of step S160 (FIG. 10), whether or not the search results include the information is determined based on the user's operation. The predetermined weights may be updated based on machine learning using the user's judgment as to whether or not the search results include the estimate reference drawing information.

[0089] In this embodiment, an example has been described in which the searched drawing information is new drawing information N1 (drawing information about a new drawing), but this is not limited to this. The searched drawing information may be drawing information about a drawing in the middle of creation. The drawing in the middle of creation may be a handwritten drawing. The drawing in the middle of creation only needs to include shape information about the drawing in the middle of creation, and may not include at least one of processing information and character-numeric information. Furthermore, the drawing in the middle of creation may be text that explains the drawing. If the drawing in the middle of creation is text that explains the drawing, the searched drawing information only needs to include character-numeric information about the text. If the searched drawing information includes only character-numeric information, the search by the search unit 11 is performed based on the text.

[0090] In this embodiment, an example in which multiple similarity search conditions are selected from various forms displayed on the search screen B1 (FIG. 5) has been described, but the present invention is not limited to this. Drawing data included in the drawing information may be displayed as an image, and similarity search conditions may be specified by performing a predetermined operation on each part of the image. The predetermined operation may be, for example, a double-click or a tap.

[0091] For example, in an image displaying drawing data, when a predetermined operation is performed in the area where the drawing is displayed, an index of similar shape is selected as a search criterion. Similarly, in an image displaying drawing data, when a predetermined operation is performed in the area where a technical symbol is displayed, an index of similar processing content is selected as a search criterion. In an image displaying drawing data, when a predetermined operation is performed in the area where a name is displayed, an index of similar name is selected as a search criterion. In an image displaying drawing data, when a predetermined operation is performed in the area where dimensions are displayed, an index of similar dimension is selected as a search criterion. When each search criterion is selected, a menu for setting the priority for that search criterion is displayed, and the priority is set from the menu.

[0092] In addition, one or more of the drawing feature information generation unit 12, priority setting unit 13, similarity display unit 14, estimate creation unit 15, drawing data management unit 16, operation reception unit 17, and output unit 18 may be omitted from the configuration of the drawing search system 1.

[0093] As described above, the drawing search system 1 according to this embodiment includes the drawing information receiving unit 10 and the search unit 11. The drawing information receiving unit 10 receives drawing information (in this embodiment, new drawing information N1, which is searched drawing information). The search unit 11 searches for similar drawing information, which is existing drawing information that is similar to the drawing information, from among multiple existing drawing information registered in the existing drawing database A1 based on multiple search conditions (in this embodiment, multiple similarity search conditions).

[0094] With this configuration, the drawing search system 1 according to this embodiment can search for similar drawing information based on a plurality of search conditions, thereby improving the accuracy of drawing searches based on a plurality of search conditions.

[0095] In the drawing search system 1 according to this embodiment, the accuracy of the drawing search is improved, thereby improving the order acceptance rate and productivity as follows.

[0096] Regarding the improvement of the order rate, first, by referencing similar past performance, it is possible to understand the logic of the quote at the time and points to note, allowing even inexperienced users to create highly accurate quotes (improved accuracy by de-personalizing quotes). Second, there are cases where manufacturing cannot be completed in-house and some processing is outsourced to a partner company. In such cases, when a request for a quote is sent to the partner company, the quote will not be completed until a response is received. However, by referencing which partner company was previously requested to perform similar processing and for how much, there is no need to request a further quote from the partner company (improved speed of quote creation).

[0097] Regarding productivity improvement, firstly, by referring to points to note during manufacturing in similar past projects, mistakes during manufacturing can be reduced, and unnecessary costs can be reduced by preventing re-production (quality improvement at the manufacturing site). Secondly, when an orderer approaches a project, the company can check information from similar past projects, such as whether the project can be handled in-house and how many man-hours it will take (sales efficiency improvement).

[0098] As described above, the drawing search system 1 according to this embodiment includes the search unit 11 and the similarity display unit 14. The search unit 11 searches for similar drawing information, which is existing drawing information similar to the drawing information, from among multiple existing drawing information registered in the existing drawing database A1 based on the similarity calculated from the detailed similarity, which is the similarity between the drawing information and the similar drawing information for each of multiple indicators. The similarity display unit 14 displays the detailed similarity used by the search unit 11 to search for similar drawing information together with the similar drawing information.

[0099] With this configuration, the drawing search system 1 according to this embodiment can display the detailed similarity, so that when a search is performed based on the similarity for each of a plurality of indices, the similarity (detailed similarity) for each indices can be confirmed.

[0100] Being able to check detailed similarities leads to more efficient search behavior. By clearly showing the similarity for each index, not just the overall similarity, users can check whether the similarity of the index they consider important is high or not. In addition, they can more reliably search for drawings similar to the drawing they are looking for.

[0101] Note that a portion of the drawing search system 1 in the above-described embodiment may be implemented by a computer. In this case, a program for implementing this control function may be recorded on a computer-readable recording medium, and the program may be loaded and executed by a computer system. Note that the term "computer system" refers to a computer system built into the drawing search system 1, including hardware such as an operating system (OS) and peripheral devices. Furthermore, the term "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, read-only memories (ROMs), and compact disc-read-only memories (CD-ROMs), as well as storage devices built into the computer system, such as hard disks. Furthermore, the term "computer-readable recording medium" may also include media that dynamically store programs for a short period of time, such as communication lines used when transmitting programs via networks such as the Internet or telephone lines, or media that store programs for a fixed period of time, such as volatile memory within a computer system serving as a server or client. The program may be designed to implement a portion of the aforementioned functions, or may be capable of implementing the aforementioned functions in combination with a program already stored in the computer system. Furthermore, part or all of the drawing search system 1 in the above-described embodiment may be realized as an integrated circuit such as an LSI (Large Scale Integration). Each functional block of the drawing search system 1 may be individually implemented as a processor, or part or all of them may be integrated into a processor. Furthermore, the integrated circuit implementation method is not limited to LSI, and may be implemented using a dedicated circuit or a general-purpose processor. Furthermore, if an integrated circuit implementation technology that can replace LSI emerges due to advances in semiconductor technology, an integrated circuit based on that technology may be used.

[0102] One embodiment of the present invention has been described in detail above with reference to the drawings, but the specific configuration is not limited to that described above, and various design changes and the like are possible within the scope that does not deviate from the gist of the present invention. [Explanation of symbols]

[0103] 1...Drawing search system, 11...Search unit, 14...Similarity display unit, A1...Existing drawing database

Claims

1. a search unit that searches for similar drawing information, which is existing drawing information similar to drawing information, from among a plurality of existing drawing information registered in an existing drawing database, based on a similarity calculated from a detailed similarity, which is a similarity between the drawing information and the similar drawing information for each of a plurality of indexes; a similarity display unit that displays the detailed similarity used by the search unit to search for the similar drawing information together with the similar drawing information; A drawing search system comprising:

2. the similarity display unit displays the detailed similarity used by the search unit to search for the similar drawing information and the similarity together with the similar drawing information. The drawing search system according to claim 1 .

3. the similarity display unit displays the detailed similarity used by the search unit to search for the similar drawing information as a value measured on a predetermined scale or in a predetermined display mode. The drawing search system according to claim 1 .

4. a search step of searching for similar drawing information, which is existing drawing information similar to drawing information, from among a plurality of existing drawing information registered in an existing drawing database, based on a similarity calculated from a detailed similarity, which is a similarity between the drawing information and the similar drawing information for each of a plurality of indexes; a similarity display step of displaying the detailed similarity used in the search step for searching the similar drawing information together with the similar drawing information; A drawing search method having the following.

5. On the computer, a search step of searching for similar drawing information, which is existing drawing information similar to drawing information, from among a plurality of existing drawing information registered in an existing drawing database, based on a similarity calculated from a detailed similarity, which is a similarity between the drawing information and the similar drawing information for each of a plurality of indexes; a similarity display step of displaying the detailed similarity used in the search step for searching the similar drawing information together with the similar drawing information; A program to execute.

Images