CN-122029551-A - Method for classifying predicted scraps in industrial manufacturing process and related system

Abstract

The invention relates to a method for handling material offcuts in an industrial manufacturing process, comprising ■ acquisition (ACQ 1 ) of an initial schedule (PL 1 ), ■ acquisition (ACQ 2 ) of an initial stock layout plan (PIM A ), ■ characterization (CAR 1 ) of at least one predicted offcut (CH k ) defining at least one area of the stock layout plan (PIM A ), ■ querying an initial item database (BD 1 ) comprising a set of parts to be produced (P n ), ■ performing a first comparison (COMP 2) between the size (K c ) of the at least one predicted offcut (CH k ) and the size of the parts to be produced (P n ) extracted from the first item database (BD 1 ), n determining a LIST of compatible parts to be produced (LIST n ), n classifying each predicted offcut (CH n ), and n updating the offcut database (BD n ).

Inventors

• F. Klein
• S. Moller
• J. Ren e - Alexander

Assignees

• 里弗斯系统公司

Dates

Publication Date

20260512

Application Date

20240814

Priority Date

20230816

Claims (20)

1. 1. A computer-implemented method for classifying predicted material scrap in an industrial manufacturing process, the method aiming at performing cutting of one or more sheets of material based on a plan (PL 1'), the method comprising: ■ Obtaining (ACQ 1 ) a first plan (PL 1 ) for a group of parts to be manufactured (P i ) constituting a step in an industrial production process for a plurality of parts to be produced, said parts (P i ) being intended to be cut from a group of material sheets (TOL k ); ■ Obtaining (ACQ 2 ) at least one initial layout plan (PIM A ) of parts to be cut (P i ) on each sheet of material (TOL k ), each part to be cut (P i ) having a two-dimensional profile and thickness and its profile inscribed within a first surface comprised by the surface of said sheet of material (TOL k ), each part to be manufactured represented in said first layout plan (PIM A ) being referred to as a side (FL i ); ■ -characterizing (CAR 1 ) at least one predicted scrap (CH k ), the predicted scrap (CH k ) defining at least one area of the layout plan (PIM A ) not occupied by a side face (FL 1 ), the characterizing (CAR 1 ) comprising at least: ■ An identifier (ID c ) for each predicted scrap (CH k ); ■ The size (K c ) of each predicted scrap (CH k ); ■ Querying (REQ 1 ) a first item database (BD 1 ) containing a set of parts (P n ) to extract geometric descriptors of the parts (P n ); ■ -making a first comparison (COMP 1 ) between the size (K c ) of the at least one predicted scrap (CH k ) and the size of at least one part (P n ) extracted from the first item database (BD 1 ), the first comparison (COMP 1 ) generating a matching degree index (IND 1 ); ■ Determining a first LIST (LIST 1 ) of compatible parts (P n' ) based on the value of the matching degree index (IND 1 ); -selecting (SEL 1 ) at least one part to be produced (P n' ) from said first LIST (LIST 1 ) according to a given criterion (C 1 ), and reassigning milestones in a schedule (PL 1 ,PL i ); ■ Classifying (CLASS 1 ) each predicted scrap (CH k ) according to a plurality of labels (LB 1 、LB 2 、LB 3 ) based on a calculation of a characteristic score of the predicted scrap (CH k ), the score being calculated in particular based on at least the first criterion (C 1 ); ■ Generating (GEN 1 ) a new first plan (PL 1' ) comprising generating at least one new stock plan (PIM B ) for at least one sheet of material (TOL k ), the new stock plan (PIM B ) comprising at least one new part selected from the first LIST (LIST 1 ) within an area corresponding to a predicted scrap (CH k ) having a reuse label (LB 1 ); ■ The scrap database (BD 2 ) is updated (MAJ 1 ) with at least one marked predicted scrap (CH k ) recorded therein.
2. 2. The method according to claim 1, characterized in that the first plan (PL 1 ) comprises at least one milestone corresponding to a predicted cut of at least one sheet of material (TOL k ), constituting a step in an industrial production flow of a plurality of parts to be produced.
3. 3. The method according to any one of claims 1 to 2, wherein the initial plan (PL 0 ) comprises at least one product identifier and an association with a plurality of first plans (PL 1 ), each first plan (PL 1 ) being associated with a sheet of material (TOL k ).
4. 4. A method according to any one of claims 1-3, characterized in that the layout plan (PIM A ) comprises a set of identifiers of parts to be produced, and for each identifier the parts to be produced are oriented and positioned within a surface representing the surface of a sheet of material (TOL k ) associated with the layout plan (PIM A ).
5. 5. The method according to any one of claims 1 to 4, characterized in that the first surface of the layout plan (PIM A ) is identical to the surface of at least one sheet of material (TOL k ).
6. 6. The method according to any one of claims 1 to 5, characterized in that the dimensions (K c ) of each scrap (CH i ) comprise in particular the area and the thickness.
7. 7. The method according to any one of claims 1 to 6, characterized in that the set of scrap (CH k ) is generated by processing all areas of the layout plan not occupied by sides (FL 1 ), said processing comprising a step (SEG 1 ) of dividing each predicted scrap surface based on the cut lines of the sides of the layout plan (PIM A ) and the layout plan edges.
8. 8. The method according to claim 7, wherein the step of segmenting each predicted scrap surface (SEG 1 ) comprises considering a set of exclusion criteria that allow assigning areas of the layout plan (PIM A ) surface not occupied by sides (FL 1 ) to scrap areas that do not participate in the generation of predicted scrap.
9. 9. The method according to any one of claims 7 to 8, wherein the step of segmenting each predicted scrap surface (SEG 1 ) comprises a first characterization (CAR 10 ) of the scrap taking into account the size of the predicted scrap (CH k ), the location of the predicted scrap (CH k ), the surface condition of the predicted scrap (CH k ) and/or the material of the predicted scrap (CH k ).
10. 10. The method according to any one of claims 1 to 9, wherein each predicted scrap (CH k ) comprises a set of geometric descriptors, in particular comprising: ■ Minimum rectangle and/or; ■ Polygons that contain the largest number of edges, and/or; ■ An image of the predicted trim, and/or; ■ A two-dimensional outer frame for pre-measuring the outline of the leftover materials and/or; ■ Data characterizing the precise surface area of the predicted trim.
11. 11. The method according to any one of claims 1 to 10, wherein said first comparison (COMP 1 ) comprises comparing the surface area of the predicted scrap (CH k ) with the surface area of the part identified in the item database (BD 1 ).
12. 12. The method according to any one of claims 1 to 11, wherein said comparing step (COMP 1 ) comprises comparing the area of the polygon inscribed in the outer frame defining the area of the predicted scrap (CH k ) with the perimeter and/or area of the circumscribing polygon containing the area selected from the parts of the item database (BD 1).
13. 13. The method according to any one of claims 1 to 12, wherein said comparing step (COMP 1 ) comprises comparing the perimeter and/or area of a plurality of polygons generated and inscribed within the outer frame defining the area of the predicted scrap (CH k ) with the perimeter and/or area of the circumscribing polygon containing the area selected from the item database (BD 1 ) parts.
14. 14. The method according to any one of claims 1 to 13, wherein the first criterion (C 1 ) comprises: ■ Consider a first occurrence of a part extracted from the item database (BD 1 ) and associated with a first plan (PL 1 ) or another plan (P i ), and/or ■ Consider a collection of identical parts extracted from the item database (BD 1 ) and associated with at least one sheet of material (TOL k ) from the first schedule (PL 1 ), and/or ■ At least a first time interval between two milestones in the same plan (PL 1 ), wherein the part extracted from the item database (BD 1 ) is associated with two associated milestones each, each milestone being associated with a different sheet of material (TOL k ), and/or ■ At least one second time interval between two plans (PL 1 、PL i ), wherein the part extracted from the item database (BD 1 ) is associated with at least one milestone in each plan.
15. 15. The method according to any one of claims 1 to 14, wherein the classification (CLASS 1 ) comprises: ■ -assigning a first tag (LB 1 ) to the predicted scrap (CH k ) for reuse when the first criterion (C 1 ) comprises: ■ In the first planning schedule (PL 1 ) or in another planning schedule (PL i ), the number of occurrences or batches of parts selected from the first LIST (LIST 1 ) is greater than a predetermined threshold, and ■ The first time interval or the second time interval is less than a predetermined duration; ■ -assigning a second tag (LB 2 ) to the predicted scrap (CH k ) for reservation when the first criterion (C 1 ) comprises: ■ In the first plan (PL 1 ) or another plan (PL i ), a number of occurrences or batches of parts selected from the first LIST (LIST 1 ) is greater than a predetermined threshold, and ■ The first time interval or the second time interval is greater than a predetermined duration, and/or; -when the first criterion (C 1 ) corresponds to a structural, dimensional or quality attribute of the predicted scrap (CH k ) corresponding to a predetermined criterion, assigning a third label (LB 3 ) to the predicted scrap for retention, and/or; ■ -assigning a fourth tag (LB 4 ) to the predicted scrap (CH k ) for cleaning when the first criterion (C 1 ) comprises: ■ The presence of the selected part being below a predetermined threshold, and ■ The first time interval or the second time interval being greater than a predetermined length of time, and ■ The third tag (LB 3 ) is assigned to the predicted scrap.
16. 16. The method according to any one of claims 1 to 15, characterized in that the generating of the new stock plan (PIM B ) involves superimposing the first stock plan (PIM A ) and an intermediate stock plan (PIM i ), the intermediate stock plan (PIM i ) comprising at least one new part to be produced selected from the first LIST (LIST 1 ) and being positioned and oriented in an area corresponding to predicted scrap with a multiplexing label (LB 1 ).
17. 17. The method according to any one of claims 1 to 16, characterized in that the scrap database (BD 2' ) comprises scrap with the second tag (LB 2 ) and/or the third tag (LB 3 ) in order to provide a user with a communication interface accessible from a data network for accessing the scrap so marked.
18. 18. A computer program product comprising instructions which, when executed by a computer, cause the computer to carry out the steps of the method according to any one of claims 1 to 17.
19. 19. A system (ST 1 ) for generating a label for a predicted scrap, comprising: ■ At least one communication interface (INT 1 ) for retrieving planning data, planning and parts from a first database (BD 1 ), ■ A memory (MEM 1 ) for storing said data, and ■ -A calculator (K 1 ) for generating a label for the predicted scrap according to the method of any one of claims 1 to 17, and further comprising an output communication interface (INT 2 ) for transmitting update data of the marked scrap to a computer device.
20. 20. The system (ST 1 ) according to claim 19, characterized in that it comprises a material sheet state sensor for calculating a state criterion considered in formulating the first criterion (C 1 ), which can be associated in particular with the surface condition of the material sheet, such as the degree of tarnishing, impact, paint treatment or quantification of orientation criteria, such as the rolling direction or the fibre direction.

Description

Method for classifying predicted scraps in industrial manufacturing process and related system Technical Field The present invention relates to a computer-implemented method and system for classifying scrap material generated during the manufacture of industrial parts. More particularly, the field of the invention relates to methods aimed at reducing the generation of scrap in material cutting operations. The field of the invention relates more particularly to the cutting of sheet metal or plates of material such as metal, but can also be applied to the processing of glass, plastics (such as polymers), wood, composite materials and textiles. Background There are a number of production schemes currently available for planning and optimizing the cutting of groups of industrial parts from a sheet of material. This operation is typically optimized by a manufacturing layout plan (plan' imbrication), i.e., optimizing the layout of the parts within the sheet to minimize material waste. There are a number of algorithms to optimize this layout based on constraints of part size, frequency of occurrence, necessary margins, cutting techniques, etc. However, a significant amount of material waste still occurs when planning industrial part production. The creation of such material wastage is sometimes unavoidable because the planning involves non-optimized final material sheets, or the part order may change over time. Furthermore, the size of the material plates may vary from supply to supply during planning, and during the life cycle of the industrial production line, there is a heterogeneity of the different products and parts to be produced, which can lead to the generation of scrap. Disclosure of Invention The present invention overcomes the above-described drawbacks of the prior art. According to a first aspect, the invention relates to a computer-implemented method for classifying predicted material scrap in an industrial manufacturing process, comprising: ■ Obtaining an initial plan of a set of parts to be manufactured, the parts being intended to be cut from a set of sheets of material; ■ Acquiring at least a first layout plan of parts to be cut for each sheet of material, each part to be cut having a two-dimensional profile and a thickness, and its profile inscribed within a first surface comprised by the surface of the sheet of material, each part to be manufactured represented in the first layout plan being referred to as a side; ■ Characterizing at least one predicted scrap area defining at least one area not occupied by a side in a layout plan, said characterizing comprising at least: ■ An identifier of each predicted scrap; ■ The size of each predicted scrap area; ■ Querying a first article database containing a set of parts to extract geometric descriptors of the parts; ■ Performing a first comparison of the size of the at least one predicted scrap and the size of the at least one part extracted from the first item database, the first comparison producing a match index; ■ Determining a first list of compatible parts according to the value of the matching degree index; ■ Selecting at least one part from the first list according to a given criterion; ■ Classifying each predicted scrap according to a plurality of labels based on a feature score calculation for the predicted scrap, the score being calculated in particular according to at least a first criterion; ■ Generating a new initial plan including generating at least one new stock plan for the at least one sheet of material, the stock plan including selecting at least one new part from the initial list in an area corresponding to predicted scrap with a reusable label; ■ Updating the scrap database, wherein at least one marked predicted scrap is recorded. According to one embodiment, such a computer-implemented method for predicting material scrap classification in an industrial manufacturing process aims at effecting cutting of a material sheet(s) based on at least one plan for a part to be produced. According to one embodiment, the obtaining of the first plan constitutes a step for a plurality of parts to be produced in an industrial manufacturing process. One advantage is that it allows the use of the predicted scrap for different purposes, in particular by repositioning the new part to be produced in the position of the predicted scrap, so that the scrap sheet is reused. According to one embodiment, the first plan includes at least one milestone corresponding to a predicted cut to at least one sheet of material, constituting a step in an industrial production flow of a plurality of parts to be produced. One advantage is that it optimizes a production line containing multiple milestones, wherein different material cutting plans are provided. According to one embodiment, the first plan contains at least one sheet identifier, at least one cutting date for each sheet of material of the first plan, and the number of parts to be produced. According to one e