JP5530732B2 - Work instruction generation system, method and program - Google Patents

Description

  The present invention relates to a technique for creating a work instruction for a worker using a computer.

  For the purpose of sharing parts of each product, there is a technology that automatically searches for parts that have been designed and registered in the past and that have similar shape characteristics. In addition, there is a technology for automatically searching and deleting feature parts such as holes and fillets for the purpose of simplifying the model shape for analysis. In addition, for the purpose of reusing cutting conditions, there is a technology that searches for processed parts that are similar in shape characteristics that affect the cutting conditions of the processed parts, and reuses the machining specifications.

Japanese Patent Application No. 2003-202774 Japanese Patent Application No. 2005-202774 Japanese Patent Application No. 2003-202592

  If we provide a method for reusing the manufacturing process of parts and the welding assembly process of parts for a range of work instructions represented by can manufacturing processes including processing, assembly and welding of parts, It is possible to reduce the man-hours for creating work instructions for the production process of the new model. In addition, when registering a process of a plurality of similar parts in the same product, the process of registration can be reduced by enabling the process registration in a batch by the function of searching for similar parts.

  However, when reusing an existing process of a part having a similar shape, a simple search based on the shape appearance size as in Patent Documents 1 and 2 has a problem in accuracy as a conventional method for searching for a similar shape.

  An object of the present invention is to recycle a work instruction sheet with high accuracy with respect to the above problem.

In a form in which the computer executes the process flow generation of the production work, the assembly work, and the welding work by reusing the work instruction information of the existing piece similar to the search source piece,
A function of acquiring the shape and physical feature of a piece from CAD via a piece information acquisition unit;
It has a piece / process database that is a database that registers the shape and physical features of the piece and the manufacturing process of the piece.
A similar part / process search function that searches for existing processes of parts and similar parts whose shape and physical features approximate to the search source part,
A part / process registration function that registers a part and its manufacturing process and generates a process flow;
Has an assembly animation database to register assembly animation control codes and viewer models.
Causes the computer to perform the assembly process and welding process setup process registered by the process registration unit and the assembly flow generation function for generating a series of assembly work and welding process process flows in accordance with the assembly sequence indicated by the assembly animation control code. This is a work instruction generation program.

    According to the present invention, by making it possible to reuse an existing process of a part of an existing model whose shape and characteristics are similar to those of a target model for which a work instruction is created, the man-hour for creating the work instruction is reduced. .

It is a block diagram of the system concerning one Example of this invention. It is the flow of the process of a similar part search concerning one Example of this invention. It is a flow of processing of processing process registration concerning one example of the present invention. It is the flow of the process of the work registration concerning one Example of this invention. It is an example of the manufacturing process flow concerning one Example of this invention. It is an example of the work information of the manufacturing process flow concerning one Example of this invention. It is the flow of allocation of the similar process concerning one Example of this invention.

  FIG. 1 shows the configuration of a computer system that implements the present invention. The computer system includes a configuration of a processing unit 11, a storage unit 12, an input unit 2, an output unit 3, and a connection line 13 connecting them. The processing unit 11 is a processor such as a CPU, the storage unit 12 is an HDD or a semiconductor memory, the input unit 3 is a keyboard or a mouse, the output unit 3 is a display or a printer, and the connection line 13 is For example, a connection cord or a network.

  The processing unit 11 then executes a program stored in a storage device or the like, thereby performing CAD1, a piece information acquisition unit 4, a similar piece / process search unit 6, a process registration unit 5, and an assembly flow generation unit 7 It functions as the processing / assembly process composite unit 10. The storage device stores a piece / process DB 8 and an assembly animation DB 9.

  The CAD 1 loads information on a product model for which a process is designed and a work instruction is to be created. The piece information acquisition unit 4 acquires the feature quantities of the pieces constituting the product from the loaded product model information.

  The piece / process database 8 (hereinafter referred to as a piece / process DB) stores and manages the acquired feature quantities of the pieces and process information of the pieces.

  The input unit 2 accepts user support input, and the output unit 3 displays and prints the processing results of the computer system.

  The similar part process search unit 6 searches for a part similar to a part serving as a search key and its process.

  The process registration unit 5 generates a process flow of incidental processes such as the manufacturing process of the pieces and the attachment and setup of the restraining jig for the assembly work and the welding work, and registers them in the piece / process DB.

  The assembly animation database 9 (abbreviated as assembly animation DB in the figure) registers the viewer model of the assembly and its animation control code.

  The assembly flow generation unit 7 adds a process flow of incidental processes such as the manufacturing process of the pieces registered by the process registration unit 5 and the binding jig installation and setup required for welding work to the assembly operation of the assembly animation. An assembly flow is generated and registered in the piece / process DB.

  The processing / assembly process composite unit 10 generates a composite process by linking related processes among the processing processes and assembly processes registered in the part / process DB 8 and registers them in the part / process DB 8.

  Next, the processing method of the system of the present invention will be described.

(Process registration process)
A process registration process method performed by the process registration unit 5 will be described. In the process registration process, the process included in the manufacturing process, the work included in the process, and information and order related to them are registered. The information regarding the process / work includes text, reference diagrams, and moving images. When the work is actually performed, these texts, reference diagrams, and moving images are displayed according to the order of the work, and serve as work instructions.

  The process registration procedure differs depending on the process type. There are three types of processes: machining process, assembly process, and combined process / assembly process. In the present embodiment, the registration of the process steps will be described.

  FIG. 4A shows an example of the machining process flow of the process of registration in this process, and FIG. 4B shows the work order relationship of the work registered in the process. These can be displayed on the output unit 3 and can be corrected via the input unit 2.

  The machining process flow shown in FIG. 4A includes a process information 41 indicating the process name of one process and the contents of the process among a plurality of processes included in the machining process, the work shown in order, and each work. The work information 43 indicating the contents of is included.

  The process information 41 is stored and managed in the piece / process DB 8 together with the piece feature amount information. The process information is composed of the following information. (1) “Process name” that is the name of the process, (2) “Part number” of the target part, (3) “Process type” that distinguishes the type of the process from the machining process to the assembly process, ( 4) Work sequence 402, (5) “Reference drawing / moving image” for reference of the process.

  The work information 43 describing the contents of work constituting the process is composed of text, numerical values, and files including the following contents. (1) Work name, (2) “Part number” of the part to be worked, (3) “Work content” expressed in text and numerical values, (4) 3D or 2D drawing or still image “Reference diagram” which is a file, and (5) “Movie” which is a moving image file indicating work contents such as work instruction animation.

  FIG. 4B shows an example of the work order relationship 42. The work order is the time-series context of work. As data, an order relationship between “pre-work” that is a work performed immediately before a certain work and “post-work” that is a work performed immediately after the work is shown. "Pre-operation" and "post-operation" can take a plurality of operations. In the example of the process of FIG. 4A, the data on the work order relation is registered as shown in FIG.

  The processing flow defines the contents to be displayed on a monitor installed beside the worker when the actual processing is performed. For example, when the worker performs the work a, the work information of the work a is displayed on the monitor. In addition, when the operation proceeds to the operation b, the operator operates so that the operation information of the operation b is displayed, and the reference diagram registered in the operation information of the operation b is also displayed.

  Next, the process flow registration process in the case of only the machining process will be described with reference to FIG.

  In step 301 in FIG. 3A, a process ID for a new process is allocated. In step 302, the process type is registered in accordance with the input, but in this process flow, the process type is registered as “processing”. Here, when assigning the pieces, the process proceeds to step 303, and selection of a piece model (search source piece) to be a search base is received from the user by the input unit 2. Step 304 performs a similar part search process 29 for searching for similar parts from the parts stored in the part / process DB 8 using the search source part as a search key part. In step 305, one or a plurality of similar pieces that are search results of the similar piece search process are displayed on the output unit 3 as candidate pieces to be registered in the process (registration target candidate pieces). The similar part model is highlighted (emphasized or marked). In step 306, the modification of the user's piece made as necessary is accepted, and the piece to be registered is selected. In step 307, the selected piece is set as the target piece of the machining process, and its piece ID is registered. Steps 303 to 306 are omitted when the piece allocation is not performed.

  In step 308, the process name of the process information 41 is registered based on the information input by the user. In step 309, if there is a request for registration of a reference diagram or a moving image of the corresponding process, a registration process to the process information 41 is performed.

  After the process for the process is performed, a registration process is performed for each work unit included in the process.

  In step 310, the work registration 311 shown in FIG. In step 312, a new work ID is assigned to the target work. In step 313, the work name input by the user is registered in the work information 43. In the case of performing the part assignment, the process proceeds to step 314, and the part ID registered in the process is inherited from the part registered in the process in the registration process for the process, and the work is also registered. In step 315, the modification of the target piece ID of the work is accepted from the user, and the piece ID unrelated to the work is deleted. In step 316, if there is a pre-work or post-work of the corresponding work, registration is accepted as the work ID of those work as the pre-work ID or post-work ID.

  In step 317, the work content is registered as text, and in step 318, a reference diagram or a moving image file that serves as a reference for the work content is registered. The registered text, reference map, and moving image are the text, reference map, and moving image that are associated with the fragment ID registered in the work.

  When the user's work registration for the work is completed, the user registers the next work. When the registration of the work included in the process is completed, the registration of the other process is performed, and when all the process registration is completed, the registration of the machining process is ended.

  Here, the similar part search performed in steps 303 to 306 will be described. In the similar part search, a part similar to the search source part is searched from the part / process DB.

(Partition information registration process)
A description will be given of a piece information registration processing method for registering piece information in the piece / process DB, which is performed as data creation in the piece / step DB used in the search process. The piece information is performed on a piece of a product model for which an existing instruction is created in advance. Load the information of the product model to be registered in the process onto CAD1, and the part information acquisition unit 2 acquires the feature quantities of the parts that make up the product model from the model information on CAD1 and corresponds to the part ID Attach it to the piece / process database 8. Here, the feature amount of the piece includes the following information. (1) Dimensions of the smallest rectangular parallelepiped containing the pieces, (2) Mass of the pieces, (3) Volume of the pieces, (4) Surface area of the pieces, (5) Moment of inertia of the pieces, (6 ) Inertial spindle of part, (7) Material.

  The feature amount is expressed by the following feature vector.

  (1) Dimension of the smallest cuboid that contains a piece: A three-dimensional vector in which the three dimensional values (height, length, and depth) of the smallest cuboid that contain a piece are arranged in ascending order of d1 <d2 <d3. One feature vector is represented by (d1, d2, d3).

  (2) Mass of the piece, (3) Volume of the piece, (4) Surface area of the piece: A three-dimensional vector (m, v, s) represents one feature vector.

  (5) Main moment of inertia of a piece: A feature vector is represented by a three-dimensional vector (p1, p2, p3) in which three scalars of the main moment of inertia are arranged in descending order of p1> p2> p3.

  (6) Inertial principal axis of a piece: Three three-dimensional vectors (e11, e12, e13), (e21, e22, e23), (e31, e32, e33) representing the principal principal axis are arranged in one row vector 9 One feature vector is represented by a dimension vector (e11, e12, e13, e21, e22, e23, e31, e32, e33).

  (7) Material: A feature vector is represented by an m-dimensional vector (c1, c2,..., Cm) in which m component amounts c1, c2,.

  Next, the similar part search processing method of the similar part search unit 6 will be described. The piece feature amount information is registered in the piece / process DB 8. When one search key part (abbreviated as key part) is given, a part similar to the key part is searched and a registered existing process including the similar part is assigned. Is the “similar part / process search unit 6”. The feature amount information of the key pieces and similar pieces is stored in advance in the piece / process DB by the above-described piece information registration process.

数３のｄは，キー部片と検索対象部片の特徴ベクトルの差分ベクトルとキー部片の特徴ベクトルの内積をキー部片の特徴ベクトルの大きさの２乗で除し絶対値をとったスカラーである。ｄ＝０のとき，a=xとなり，特徴ベクトルは同一である。dが大きくなるほど、ｘはaとより大きく異なることを示している。以後、このdを類似度と呼ぶ。許容される類似度ｄの上限値をdmaxと設定する。このとき，
（数４） ｄ≦ｄｍａｘ
をみたす特徴ベクトルｘをもつ部片を，特徴ベクトルaを持つキー部片と類似と判断する。 First, a method for searching for similar parts will be described in detail. A k-dimensional vector whose parameter is the numerical feature value of the registered information piece in the piece information registration process is called a feature vector. Here, the feature vector of the key piece is a, and that of the search target piece is x. At this time, d is evaluated based on x and a.
(Equation 1) a = (p ₁ , p ₂ , p ₃ ,... P _k )
(Equation 2) x = (q ₁ , q ₂ , q ₃ ,... Q _k )

D in Equation 3 is an absolute value obtained by dividing the inner product of the difference vector of the feature vector of the key piece and the search target piece and the feature vector of the key piece by the square of the size of the feature vector of the key piece. It is a scalar. When d = 0, a = x and the feature vectors are the same. It shows that as d increases, x differs greatly from a. Hereinafter, this d is referred to as similarity. The upper limit value of the allowable similarity d is set as dmax. At this time,
(Equation 4) d ≦ dmax
It is determined that the piece having the feature vector x satisfying the condition is similar to the key piece having the feature vector a.

（数８） d_av＝(d₁+d₂+・・・+d_n)/n
（数９） d_av≦dmax Here, when there are n feature vectors for one piece, the feature pieces of the key piece and the search target piece are a _i for the key piece and x _i for the search target piece. Let the similarity corresponding to the similarity d of Equation 7 be d _i . The average value d _av shown in the following equation 8 is used as the overall similarity when there are n similarities. The upper limit value of the allowable similarity d _av is suppressed by dmax.
(Equation 5) a _i = (p _i1 , p _i2 , p _i3 ,... P _ik ) i = (1,2... N)
(Expression 6) x _i = (q _i1 , q _i2 , q _i3 ,... Q _ik ) i = (1, 2... N)

( _Equation 8) d _av = (d ₁ + d ₂ +... + D _n ) / n
(Equation 9) d _av ≦ dmax

(Similar part search processing)
A flow of the similar part search processing method of the similar part search unit 6 will be described. FIG. 2 shows a flowchart of the similar part search process performed by the similar part / process search unit 6. In step 21, the input unit 2 accepts a feature vector similarity upper limit value dmax setting in Expression 2 from the user. In step 22, when an input for specifying a search target model is accepted, a setting for limiting the search range of the pieces to the pieces belonging to the specified model is performed. In step 23, the input unit 2 accepts specification of a search key piece from the user.

  Next, loop processing from step 24 to step 25 is performed for all the pieces of the search target model. In step 26, the feature amount and feature vector of the key piece are calculated from the CAD, and their similarity d is calculated from the feature vector of the key piece and the feature vector of the search target model. In step 27, if d calculated in step 26 is equal to or less than the upper limit value dmax, the ID is registered with the piece as a similar piece. If d exceeds the upper limit dmax, the part is not registered as a similar part.

  In step 25, if all the pieces have been processed, the process proceeds to the next process, and if any pieces remain, the loop is returned.

In step 28, if the key part of the search source is set to request registration as a similar part, the ID of the key part is also registered in the similar part and the procedure is terminated.
(Similar process search process)
Next, a process for assigning similar processes performed by the similar part / process search unit 6 will be described. FIG. 5 shows an allocation process flow of similar processes. At the stage of starting the similar process search process, a process flow relating to a new product model is created by the process registration process.

  First, in step 1101, the feature vector similarity upper limit value dmax in the equation 4 is set.

  Next, in step 1102, a new product model that is created in the process registration process and that is to be registered by the user is set as a search source model, a search source key part of the search source model is received, and a similar part search procedure is performed. Call. A similar part ID as a search result is received as a return value from the similar part search procedure. In step 1103, when a search target model is specified by the user, the search target fragment is limited to the input model, and the following procedure is performed with all the pieces of the search target model as the corresponding fragment. Do.

In step 1105, the similarity d _av is calculated as the similarity of the feature vector of the corresponding piece with respect to the feature vector of the search source piece. If d _av exceeds the upper limit dmax, change to another search target model and repeat the process.

In step 1106, if d _av is less than or equal to the upper limit value dmax, all existing processes or existing operations including the ID of the corresponding piece are displayed in the process flow of the search target model. In step 1107, the process or operation selected by the user in the display process is copied, and the process ID of the process flow of the new product model (inspection model) is assigned. In step 1108, the part ID of the previous search source model similar part is registered as the part ID of the new process. In step 1109, when the user requests editing of the process, the editing of the process content is accepted and the update content is registered.

  By the above similar process search process, the process flow of the existing model having a piece similar to the new product model can be reused as the process flow of the new product model.

  In the above, the process of Example 1 was demonstrated. In the present embodiment, it is possible to search for similar parts for the parts to be processed, and to reuse the work instruction information (text, reference diagram, video) associated with the similar parts. This makes it easy to create work instructions for new product models.

  DESCRIPTION OF SYMBOLS 1 ... CAD, 2 ... Input part, 3 ... Output part, 4 ... Piece information acquisition part, 5 ... Process registration part, 6 ... Similar piece / process search part, 7... Assembly flow generation unit, 8... Piece / process database (part piece / process DB), 9... Assembly animation database (assembly animation DB), 10.

Claims (4)

In a work instruction generation system that generates a manufacturing process flow using existing work instruction information of a piece similar to a piece of a product model,
A part and a process database that stores the manufacturing process using the part together with work instruction information;
A piece information acquisition unit for acquiring information on the first piece of the product model for creating a work instruction from the CAD of the product model;
A similar part / process search unit that searches for a second part having a feature quantity similar to that of the first part from the parts stored in the part / process database;
A process registration unit that generates a manufacturing process flow of the product model,
The feature amount is represented by a multidimensional feature amount vector of a plurality of elements,
The difference between the feature quantity vector of the first piece and the feature quantity vector of the second piece is projected onto the feature quantity vector of the first piece, and the feature quantity vectors of the first piece are It is characterized by judging similarity based on the value divided by the inner product of
The feature quantity includes at least one of a mass of the piece, a volume of the piece, a surface area of the piece, an inertial moment of the piece, and a material of the piece .   The similar part / process search unit uses a process flow for the second part as a process flow in which a first part similar to the second part is used. The work instruction generation system according to 1.   The work instruction generation system according to claim 2, wherein the process flow includes work instruction information.   4. The work instruction generation system according to claim 3, wherein the work instruction information includes any one of a text, a figure, a moving image, and an animation.

Images