A lean management quick changeover system

Abstract

The invention relates to the technical field of enterprise management, and discloses a lean management rapid mold change system, comprising: a mold change process information storage module, which is used to store process information required for mold change; a mold change task release module, which is used to release mold change tasks; a mold change operation tracing module, which is used to count the time used by operators for mold change operation steps when performing mold change operations; a mold change operation time regularization module, which is used to standardize the time required for mold change operation steps to obtain reference time consumption of mold change operation steps; a personnel skill evaluation module, which compares the time used by operators for mold change operation steps when performing mold change operations with the reference time consumption of mold change operation steps to evaluate the operator's operation ability for the mold change operation; combining hardware to obtain relevant information about production lines and enterprises, and performing statistical analysis based on data to realize visualization of mold change processes and results, so as to facilitate targeted optimization of mold change operations.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise management, and more specifically, to a lean management quick-change system. Background Art

[0002] In the current intelligent manufacturing field, manufacturers are facing problems such as low production efficiency, high scrap rate, and frequent quality problems in the multi-variety and small-batch production mode. The production process requires constant adjustment of equipment parameters, replacement of tool fixtures, and compilation or calling of CNC programs, which will increase the complexity and uncertainty of production. These problems are mainly caused by the difficulty in controlling and coordinating the production process due to frequent product changes. Summary of the invention

[0003] The present invention provides a lean management rapid mold change system, which combines hardware to obtain production line and enterprise related information, and performs statistical analysis based on data to realize visualization of mold change processes and results, so as to facilitate targeted optimization of mold change operations, and combines artificial intelligence technology to realize automatic scheduling of mold change operations, automatically optimize mold change operations, and improve the efficiency of mold change operations, thereby solving the technical problem in related technologies that it is difficult for production enterprises to improve the efficiency of product mold change operations.

[0004] The present invention provides a lean management rapid changeover system, comprising:

[0005] A mold change process information storage module, which is used to store the process information required for the mold change;

[0006] Personnel information management module, which is used to manage personnel information;

[0007] A model change task release module is used to release model change tasks, which include product models before and after the model change, process information required for the model change, and personnel assigned to the model change tasks;

[0008] The model change operation tracing module is used to count the time used by operators in the model change operation steps when performing the model change operation;

[0009] The abnormal information analysis module for changing models is used to count the production process links with problems and the number of scrapped products after the changing model operation;

[0010] A time regularization module for the changeover operation is used to quantify the time required for the steps of the changeover operation in a standard manner to obtain a reference time consumption for the steps of the changeover operation;

[0011] The personnel skill evaluation module compares the time taken by the operator for the steps of the model change operation with the reference time of the steps of the model change operation to evaluate the operator's ability to operate the model change operation;

[0012] The graphical module is used to generate graphs based on the processing results of the personnel skill analysis module and the model change abnormality information analysis module.

[0013] Further, the process information includes material information and operation step information;

[0014] Material information includes information on equipment, equipment parts, and raw materials;

[0015] The operation step information includes step information of the type change operation described in text or image.

[0016] Furthermore, the operating ability is divided into three levels: high, medium and low. When the time used by the operator for the model changing operation step when performing the model changing operation is greater than the reference time consumption of the corresponding model changing operation step, the operating ability is evaluated as high; when the time used by the operator for the model changing operation step when performing the model changing operation is 1 to 1.5 times the reference time consumption of the corresponding model changing operation step, the operating ability is evaluated as medium; when the time used by the operator for the model changing operation step when performing the model changing operation is 1.5 to 3 times the reference time consumption of the corresponding model changing operation step, the operating ability is evaluated as medium.

[0017] Furthermore, a personnel transfer module for allocating personnel is also included, and the personnel transfer module includes:

[0018] The shift information extraction module is used to extract the operators of all production line stations and the production capacity information of the operators;

[0019] The operator's production capacity information includes the output per unit time of each model of product;

[0020] The workstation information includes information on the processing technology link corresponding to the workstation;

[0021] A current graph generation module generates current graph structure data based on the information of operators and workstations of the current production line;

[0022] The current graph structure data includes vertices and edges connecting vertices. One vertex maps to one operator or workstation. There is an edge between the vertex mapped to the operator and the workstation where he / she is located. There is an edge between the vertices mapped to any two workstations on the production line.

[0023] Generate vertex features for vertices. The vertex features of the mapping operator are generated by encoding the operator's production capacity information. The vertex features of the mapping station are obtained by combining the station information encoding with the state code.

[0024] There are five types of combined status codes: one is four-digit 1, indicating that the workstation belongs to a running workstation on the production line to be replaced, and it remains unchanged after the replacement; one is four-digit 2, indicating that the workstation belongs to a running workstation on the production line to be replaced, and it stops running after the replacement; one is four-digit 3, indicating that the workstation belongs to a non-running workstation on the production line to be replaced, and it starts running after the replacement; one is four-digit 4, indicating that the workstation belongs to a non-running workstation on the production line to be replaced, and it remains unchanged after the replacement; one is four-digit 0, indicating that the workstation does not belong to a running workstation on the production line to be replaced;

[0025] The transformation evolution module inputs the current graph structure data into the transformation evolution model. The transformation evolution model includes: graph encoding layer, sampling layer, graph generation layer and one-hot encoding layer. The calculation formula of the graph encoding layer is as follows:

[0026] O1=SReLU(SXW1)W2

[0027] O2=SReLU(SXW1)W3

[0028]

[0029] Where X represents the input feature matrix, a row vector of the input feature matrix represents a vertex feature, O1 represents the first encoding matrix, O2 represents the second encoding matrix, represents the sum of the vertex adjacency matrix and the identity matrix, express The degree matrix, W1, W2, W3 represent three weight parameters;

[0030] The calculation formula of the sampling layer is as follows:

[0031] Z=O1+O2 2 *ε,ε~N(0,I)

[0032] Where ε represents a random constant that conforms to the standard Gaussian distribution;

[0033] The calculation formula of the graph generation layer is as follows:

[0034] U=σ(ZZ T )

[0035] U represents the output matrix, and σ represents the sigmoid function.

[0036] One-hot encoding layer, which is used to binarize the output matrix and convert the values ​​of the output matrix into 0 or 1;

[0037] The decoding module decompiles the output matrix to obtain the allocation relationship between operators and workstations.

[0038] Furthermore, the connection relationship of the vertices is recorded in the form of an adjacency matrix, and the element in the i-th row and j-th column of the adjacency matrix indicates whether there is an edge between the i-th and j-th vertices. If so, the value of the element is 1, and if not, the value of the element is 0.

[0039] Furthermore, one method of binarization is to use a conditional function: if the element value is greater than or equal to 0.5, the output is 1, otherwise the output is 0.

[0040] Furthermore, the loss function of the transformation evolution model is:

[0041]

[0042] represents the cross entropy loss function, represents the output matrix after binarization processing, D represents the vertex adjacency matrix of the post-graph structure data generated by the information of the real production line after the model change with reference to the training samples, KL represents the KL divergence, and p(Z)=N(0,I).

[0043] Furthermore, the output matrix can represent the connection relationship of the vertices of the post-graph structure data after the transformation, that is, the output matrix can represent whether there are edges between the vertices, and the operator and the workstation to be assigned to are decoded according to the existence rules of the edges of the vertices. The existence rules of the edges of the vertices used for decoding are as follows: there is an edge between the operator and the vertex mapped to the workstation.

[0044] A lean management rapid changeover method is provided, which performs the following steps based on a personnel transfer module:

[0045] Step S101, extracting operators of all production line workstations and their production capacity information;

[0046] Step S102, generating current graph structure data based on the information of operators and workstations of the current production line;

[0047] Step S103, inputting the current graph structure data into the transformation evolution model to obtain an output matrix;

[0048] Step S104, obtaining the distribution relationship between operators and workstations by decompiling the output matrix.

[0049] A storage medium, characterized in that it stores non-transitory computer-readable instructions, when the non-transitory computer-readable instructions are executed by a computer, one or more steps in the aforementioned lean management rapid changeover method. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1It is a module schematic diagram of the lean management rapid changeover system of the present invention;

[0051] like Figure 2 The figure shows the human-machine interaction interface for managing the changeover plan steps of the lean management rapid changeover software;

[0052] like Figure 3 The figure shows the human-computer interaction interface for the resource management of lean management rapid changeover software;

[0053] like Figure 4 The figure shows the changeover performance human-computer interaction interface of the lean management rapid changeover software;

[0054] like Figure 5 The figure shows a partial human-machine interaction interface of the changeover analysis report of the lean management rapid changeover software;

[0055] like Figure 6 The figure shows the second part of the human-machine interaction interface of the lean management rapid changeover software for the changeover analysis report;

[0056] like Figure 7 The third part of the human-machine interaction interface of the lean management rapid changeover software for the changeover analysis report is shown;

[0057] like Figure 8 Shown is a partial view of the human-computer interaction interface of the lean management rapid changeover software's changeover analysis report. DETAILED DESCRIPTION

[0058] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0059] At least one embodiment of the present invention provides a lean management rapid changeover system, such as Figure 1 As shown, including:

[0060] A mold change process information storage module 101, which is used to store process information required for mold change;

[0061] Process information includes material information and operation step information;

[0062] Material information includes information on equipment, equipment parts, raw materials, etc.;

[0063] Operation step information includes but is not limited to step information of the type change operation described in text or images;

[0064] A personnel information management module 102, which is used to manage personnel information;

[0065] The personnel include operators, who are used to operate the assembly line and the changeover operation. The assembly line is a unit including a series of workstations for processing products;

[0066] A model change task release module 103 is used to release model change tasks, which include product models before and after the model change, process information required for the model change, and personnel assigned to the model change tasks;

[0067] A changeover operation tracing module 104 is used to count the time used by operators for the changeover operation steps when performing the changeover operation;

[0068] The model change abnormal information analysis module 105 is used to count the production process links with problems and the number of scrapped products after the model change operation;

[0069] For example, the production process link where problems occur after the mold change operation is the hot pressing link, the cutting link, or the gluing link.

[0070] A time adjustment module 106 for the type change operation is used to quantify the time required for the steps of the type change operation in a standard manner to obtain a reference time consumption of the steps of the type change operation;

[0071] The personnel skill evaluation module 107 compares the time taken by the operator for the step of the type change operation with the reference time taken for the step of the type change operation to evaluate the operator's ability to operate the type change operation;

[0072] In one embodiment of the present invention, the operating ability is divided into three levels: high, medium and low. When the time used by the operator for the step of the model changing operation when performing the model changing operation is greater than the reference time consumption of the corresponding step of the model changing operation, the operating ability is evaluated as high; when the time used by the operator for the step of the model changing operation when performing the model changing operation is 1 to 1.5 times the reference time consumption of the corresponding step of the model changing operation, the operating ability is evaluated as medium; when the time used by the operator for the step of the model changing operation when performing the model changing operation is 1.5 to 3 times the reference time consumption of the corresponding step of the model changing operation, the operating ability is evaluated as medium.

[0073] The graphical module 108 is used to generate graphs based on the processing results of the personnel skill analysis module and the model change abnormality information analysis module.

[0074] In one embodiment of the present invention, the steps of the mold change operation include waiting, transporting, hot pressing mold installation, and hot pressing dimension debugging.

[0075] In one embodiment of the present invention, a lean management rapid changeover system further includes a personnel mobilization module for allocating personnel, and the personnel mobilization module includes:

[0076] The shift information extraction module is used to extract the operators of all production line stations and the production capacity information of the operators;

[0077] The operator's production capacity information includes the output per unit time of each model of product;

[0078] For example, the output of the GHZ model product is 5 pieces / day.

[0079] The workstation information includes information on the processing technology link corresponding to the workstation;

[0080] A current graph generation module generates current graph structure data based on the information of operators and workstations of the current production line;

[0081] The current graph structure data includes vertices and edges connecting vertices. One vertex maps to one operator or workstation. There is an edge between the vertex mapped to the operator and the workstation where he / she is located. There is an edge between the vertices mapped to any two workstations on the production line.

[0082] The connection relationship of the vertices is recorded in the form of an adjacency matrix. The element in the i-th row and j-th column of the adjacency matrix indicates whether there is an edge between the i-th and j-th vertices. If so, the value of the element is 1, otherwise, the value of the element is 0.

[0083] Generate vertex features for vertices. The vertex features of the mapping operator are generated by encoding the operator's production capacity information. The vertex features of the mapping station are obtained by combining the station information encoding with the state code.

[0084] The encoding method may be a general semantic encoding method, and the aforementioned information is generally recorded in the form of text.

[0085] For example, the vector after the information encoding of the workstation is A, and the combined state code is divided into five types, one is four-bit 1, indicating that the workstation belongs to the working station on the production line to be replaced and remains unchanged after the replacement, and the combined vertex feature is represented by A1111, one is four-bit 2, indicating that the workstation belongs to the working station on the production line to be replaced and stops running after the replacement, one is four-bit 3, indicating that the workstation belongs to the non-operating workstation on the production line to be replaced and starts running after the replacement, and one is four-bit 4, indicating that the workstation belongs to the non-operating workstation on the production line to be replaced and remains unchanged after the replacement;

[0086] One is four 0s, indicating that the workstation does not belong to the running workstation on the production line to be replaced, and the combined vertex feature is represented by A0000;

[0087] The above workstation status is based on the status when the information is extracted, which is generally the moment before the start of the changeover.

[0088] The transformation evolution module inputs the current graph structure data into the transformation evolution model. The transformation evolution model includes: graph encoding layer, sampling layer, graph generation layer and one-hot encoding layer. The calculation formula of the graph encoding layer is as follows:

[0089] O1=SReLU(SXW1)W2

[0090] O2=SReLU(SXW1)W3

[0091]

[0092] Where X represents the input feature matrix, a row vector of the input feature matrix represents a vertex feature, O1 represents the first encoding matrix, O2 represents the second encoding matrix, represents the sum of the vertex adjacency matrix and the identity matrix, express The degree matrix, W1, W2, W3 represent three weight parameters;

[0093] The calculation formula of the sampling layer is as follows:

[0094] Z=O1+O2 2 *ε,ε~N(0,I)

[0095] Where ε represents a random constant that conforms to the standard Gaussian distribution;

[0096] The calculation formula of the graph generation layer is as follows:

[0097] U=σ(ZZ T )

[0098] U represents the output matrix, and σ represents the sigmoid function.

[0099] One-hot encoding layer, which is used to binarize the output matrix and convert the values ​​of the output matrix into 0 or 1;

[0100] One way to binarize is to use a conditional function: if the element value is greater than or equal to 0.5, the output is 1, otherwise the output is 0.

[0101] The loss function of the transformation evolution model is:

[0102]

[0103] represents the cross entropy loss function, represents the output matrix after binarization processing, D represents the vertex adjacency matrix of the post-graph structure data generated by the information of the real production line after the model change with reference to the training sample, KL represents the KL divergence, and p(Z)=N(0,I);

[0104] The training samples come from the historical records of model changeovers. The historical data records the information of the production line after the model changeover. Based on the operators of all the production line stations after the model changeover and the production capacity information of the operators, a post-graph structure data can be generated. The adjacency matrix of the vertices of the post-graph structure data is used as a reference to generate the loss.

[0105] It should be noted that the historical records of model change from which the training samples are sourced are records of good model change processes or results, which need to be manually screened. As an example of a screening condition, the model change time is lower than the set value;

[0106] Alternatively, the changeover time is lower than a set value, and the output per unit time of the production line that has not been changed decreases by less than 10%.

[0107] The adjacency matrix of the vertices of the graph structure data is generated with reference to the information of the actual production line after the model change.

[0108]

[0109] In the above formula, y i represents the value of the i-th element in D (0 or 1), Represents the generated graph matrix is the value of the i-th element in , and N is the total number of elements in D.

[0110] The decoding module decompiles the output matrix to obtain the allocation relationship between operators and workstations.

[0111] The output matrix can represent the connection relationship of the vertices of the post-graph structure data after the transformation, that is, the output matrix can indicate whether there are edges between the vertices. The operator and the workstation to be assigned are decoded according to the existence rules of the edges of the vertices. The decoding is based on the following rules: there is an edge between the operator and the vertex mapped to the workstation.

[0112] Operators are assigned based on the decoding results and the changeover is initiated.

[0113] The aforementioned personnel mobilization module can mobilize operators of production lines that do not need to change models to reasonably allocate human resources according to the capabilities of the operators, thereby minimizing the changeover time while reducing the impact on the production of other production lines and maximizing production speed.

[0114] The present invention provides a method for rapid mold change, which combines software and hardware technologies with professional knowledge of rapid mold change, and can effectively reduce mold change time and improve production efficiency. At the same time, in order to achieve rapid mold change, the present invention provides some hardware technology integration solutions that can make it easier for workers on the production line to view production tasks, equipment status, material information, etc., so as to perform mold change operations more quickly. It makes it easier for workers on the production line to perform production operations. It can also identify deviations between the mold change process and the standard movements, and identify safety operation risks during the mold change process (for example, it is capable of providing a goggle vision and helmet vision technology that can provide more safety features, which can be used to improve the safety of high-risk tasks such as identifying mold change operations).

[0115] In addition to the technologies mentioned above, there are also automation equipment, sensor technology, machine vision, etc. These technologies can help production lines become more automated and intelligent, thereby improving production efficiency and reducing changeover time.

[0116] At least one embodiment of the present invention provides a lean management rapid changeover software, such as Figure 2 The figure shows the human-machine interaction interface for the lean management of rapid changeover software. Figure 3 The figure shows the human-computer interaction interface of the lean management rapid changeover software for resource management. Figure 4 The figure shows the human-computer interaction interface of the lean management rapid changeover software. Figure 5 The figure shows a partial view of the human-machine interaction interface of the lean management rapid changeover software for the changeover analysis report. Figure 6 The following is a partial view of the human-machine interaction interface of the lean management rapid changeover software for the changeover analysis report. Figure 7 The following is a partial view of the human-machine interaction interface of the lean management rapid changeover software for the changeover analysis report. Figure 8 Shown is a partial view of the human-computer interaction interface of the lean management rapid changeover software's changeover analysis report.

[0117] The names of the people in the figure are all pseudonyms, and the numbers are all meaningless exemplary numbers.

[0118] At least one embodiment of the present disclosure provides a lean management rapid changeover method, which performs the following steps based on a personnel mobilization module:

[0119] Step S101, extracting operators of all production line workstations and their production capacity information;

[0120] Step S102, generating current graph structure data based on the information of operators and workstations of the current production line;

[0121] Step S103, inputting the current graph structure data into the transformation evolution model to obtain an output matrix;

[0122] Step S104, obtaining the distribution relationship between operators and workstations by decompiling the output matrix.

[0123] At least one embodiment of the present disclosure provides a storage medium storing non-transitory computer-readable instructions. When the non-transitory computer-readable instructions are executed by a computer, one or more steps of the aforementioned lean management rapid changeover method are performed.

[0124] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are protected by the present embodiment.

Claims

1. A lean management rapid changeover system, characterized in that: include: A mold change process information storage module, which is used to store the process information required for the mold change; Personnel information management module, which is used to manage personnel information; A model change task release module is used to release model change tasks, which include product models before and after the model change, process information required for the model change, and personnel assigned to the model change tasks; The model change operation tracing module is used to count the time used by operators in the model change operation steps when performing the model change operation; The abnormal information analysis module for changing models is used to count the production process links with problems and the number of scrapped products after the changing model operation; A time regularization module for the changeover operation is used to quantify the time required for the steps of the changeover operation in a standard manner to obtain a reference time consumption for the steps of the changeover operation; The personnel skill evaluation module compares the time taken by the operator for the steps of the model change operation with the reference time of the steps of the model change operation to evaluate the operator's ability to operate the model change operation; A graphical module, which is used to generate graphs based on the processing results of the personnel skill analysis module and the changeover abnormal information analysis module; A personnel transfer module for allocating personnel. The personnel transfer module includes: The shift information extraction module is used to extract the operators of all production line stations and the production capacity information of the operators; The operator's production capacity information includes the output per unit time of each model of product; The workstation information includes information on the processing technology link corresponding to the workstation; A current graph generation module generates current graph structure data based on the information of operators and workstations of the current production line; The current graph structure data includes vertices and edges connecting vertices. One vertex maps to one operator or workstation. There is an edge between the vertex mapped to the operator and the workstation where he / she is located. There is an edge between the vertices mapped to any two workstations on the production line. Generate vertex features for vertices. The vertex features of the mapping operator are generated by encoding the operator's production capacity information. The vertex features of the mapping station are obtained by combining the station information encoding with the state code. There are five types of combined status codes: one is four-digit 1, indicating that the workstation belongs to a running workstation on the production line to be replaced, and it remains unchanged after the replacement; one is four-digit 2, indicating that the workstation belongs to a running workstation on the production line to be replaced, and it stops running after the replacement; one is four-digit 3, indicating that the workstation belongs to a non-running workstation on the production line to be replaced, and it starts running after the replacement; one is four-digit 4, indicating that the workstation belongs to a non-running workstation on the production line to be replaced, and it remains unchanged after the replacement; one is four-digit 0, indicating that the workstation does not belong to a running workstation on the production line to be replaced; The transformation evolution module inputs the current graph structure data into the transformation evolution model. The transformation evolution model includes: graph encoding layer, sampling layer, graph generation layer and one-hot encoding layer. The calculation formula of the graph encoding layer is as follows: O1=SReLU(SXW1)W2 O2=SReLU(SXW1)W3 Where X represents the input feature matrix, a row vector of the input feature matrix represents a vertex feature, O1 represents the first encoding matrix, O2 represents the second encoding matrix, represents the sum of the vertex adjacency matrix and the identity matrix, express The degree matrix, W1, W2, W3 represent three weight parameters; The calculation formula of the sampling layer is as follows: Z=O1+O2 2 *ε,ε~N(0,I) Where ε represents a random constant that conforms to the standard Gaussian distribution; The calculation formula of the graph generation layer is as follows: U=σ(ZZ T ) U represents the output matrix, σ represents the sigmoid function; One-hot encoding layer, which is used to binarize the output matrix and convert the values ​​of the output matrix into 0 or 1; A decoding module, which decompiles the output matrix to obtain the distribution relationship between operators and workstations; The connection relationship of the vertices is recorded in the form of an adjacency matrix. The element in the i-th row and j-th column of the adjacency matrix indicates whether there is an edge between the i-th and j-th vertices. If so, the value of the element is 1, and if not, the value of the element is 0. One way to binarize is to use a conditional function: if the element value is greater than or equal to 0.5, the output is 1, otherwise the output is 0; The loss function of the transformation evolution model is represents the cross entropy loss function, represents the output matrix after binarization processing, D represents the vertex adjacency matrix of the post-graph structure data generated by the information of the real production line after the model change with reference to the training sample, KL represents the KL divergence, and p(Z)=N(0,I); The output matrix can represent the connection relationship of the vertices of the post-graph structure data after the transformation, that is, the output matrix can indicate whether there are edges between the vertices. The operator and the workstation to be assigned are decoded according to the existence rules of the edges of the vertices. The decoding is based on the following rules: there is an edge between the operator and the vertex mapped to the workstation.

2. The lean management rapid changeover system according to claim 1, characterized in that: Process information includes material information and operation step information; Material information includes information on equipment, equipment parts, and raw materials; The operation step information includes step information of the type change operation described in text or image.

3. The lean management rapid changeover system according to claim 1, characterized in that: The operating ability is divided into three levels: high, medium and low. When the time used by the operator for the model changing operation step when performing the model changing operation is greater than the reference time consumption of the corresponding model changing operation step, the operating ability is evaluated as high; when the time used by the operator for the model changing operation step when performing the model changing operation is 1 to 1.5 times the reference time consumption of the corresponding model changing operation step, the operating ability is evaluated as medium; when the time used by the operator for the model changing operation step when performing the model changing operation is 1.5 to 3 times the reference time consumption of the corresponding model changing operation step, the operating ability is evaluated as low.

4. A lean management rapid changeover method, characterized in that: The following steps are performed based on the lean management rapid changeover system as described in any one of claims 1 to 3: Step S101, extracting operators of all production line workstations and their production capacity information; Step S102, generating current graph structure data based on the information of operators and workstations of the current production line; Step S103, inputting the current graph structure data into the transformation evolution model to obtain an output matrix; Step S104, obtaining the distribution relationship between operators and workstations by decompiling the output matrix.

5. A storage medium, characterized in that: Non-transitory computer-readable instructions are stored, and when the non-transitory computer-readable instructions are executed by a computer, one or more steps in the lean management rapid changeover method as claimed in claim 4 can be executed.

Images