Log in
E-mail
Password
Remember
Forgot password ?
Become a member for free
Sign up
Sign up
Settings
Settings
Dynamic quotes 
OFFON

MarketScreener Homepage  >  Equities  >  Euronext Paris  >  Saint-Gobain    SGO   FR0000125007

SAINT-GOBAIN

(SGO)
  Report  
SummaryQuotesChartsNewsRatingsCalendarCompanyFinancialsConsensusRevisions 
News SummaryMost relevantAll newsOfficial PublicationsSector newsMarketScreener StrategiesAnalyst Recommendations

Researchers Submit Patent Application, “Method And Device For Optimizing A Plan For Cutting By Guillotine Of Pieces Of Glass”, for Approval (USPTO 20190227514): Saint-Gobain Glass France

share with twitter share with LinkedIn share with facebook
share via e-mail
0
08/14/2019 | 05:00pm EDT

2019 AUG 14 (NewsRx) -- By a News Reporter-Staff News Editor at Computer Business Daily -- From Washington, D.C., NewsRx journalists report that a patent application by the inventors LUCAS, Claire (Paris, FR); SAUSSET, Francois (Montrouge, FR); TLILANE, Lydia (Paris, FR), filed on September 7, 2017, was made available online on July 25, 2019.

The patent’s assignee is Saint-Gobain Glass France (Courbevoie, France).

News editors obtained the following quote from the background information supplied by the inventors: “The present invention lies in the field of methods of cutting glass, and it relates more particularly to a method and to a device for optimizing such cuttings.

“The method and the method of the invention serve in particular to reduce losses of glass when creating batches for cutting in a factory.

“It should be recalled that a plan for cutting a batch of rectangular pieces of glass by guillotine needs to take account of the fact that the pieces are for stacking on one or more frames, in a predetermined sequence that is specific to each frame.

“In the present state of the art, there does not exist any software for optimizing the design of a cutting plan and capable of complying with placing constraints, while minimizing losses of glass.

“Specifically, the software used at present for this purpose is capable of constructing only a limited number of cutting plans that comply with placement constraints, thus making it necessary in practice for the operator to use a plurality of pieces of software in parallel, and then to select from among the cutting plans generated by those various pieces of software, the plan that minimizes losses of glass.

“The document ‘An exact algorithm for general, orthogonal, two-dimensional knapsack problems, European Journal of Operational Research, Vol. 83, No. 1, May 18, 1995, Hadjisconstantinou et al.)’ describes a method of cutting pieces from a single sheet. That method is not applicable to a method of cutting by guillotine in which the pieces may be of dimensions that are very different, the pieces possibly being for stacking on a plurality of frames with a predefined sequence for each frame.”

As a supplement to the background information on this patent application, NewsRx correspondents also obtained the inventors’ summary information for this patent application: “The invention seeks to provide a method and a device for determining a cutting plan that does not present the above-mentioned drawbacks.

“More precisely, and in a first aspect, the invention provides a method performed by computer to determine an optimized cutting plan for using a guillotine to cut a batch of rectangular pieces of glass out from at least one sheet of glass, the pieces, once cut out, being for stacking on one or more frames, the pieces for any one frame being for placing on a sheet that is to be cut up in a sequence that is predetermined for that frame, said method comprising: an initialization step for defining cutting constraints and positioning constraints for said pieces together with an optimization criterion; creating a tree comprising a root, leaves, each presenting a complete cutting plan enabling all of the pieces of the batch to be cut out, each other node of the tree representing a partial cutting plan, the cutting plan associated with a node of the tree being obtained by adding to the partial cutting plan associated with the parent node of said node, and in compliance with predetermined constraints, the next piece for a said frame determined in compliance with the sequence predetermined for the frame; and a selection step for selecting a complete cutting plan associated with a leaf of the tree as a function of said optimization criterion.

“The invention has a preferred application when the pieces are for stacking on at least two frames, with a predefined sequence for each frame.

“The invention also has a preferred application when using at least two sheets of glass.

“Thus, and in general manner, the invention seeks to provide a method making it possible to devise a large number of complete cutting plans, including for batches that are complex. The method is remarkable in that it constructs a tree of partial cutting plans progressively by placing the pieces one by one in compliance with constraints.

“In accordance with the invention, the pieces of partial and of complete cutting plans have the dimensions of the pieces that are indeed to be stacked on the frames; each complete cutting plan having exactly the number of pieces required for all of the frames. This constitutes a fundamental difference compared with the method described in the document by Hadjisconstantinou et al., which that allows for cutting out a number of pieces that varies in compliance with a predefined limit.

“It is usual practice for the person skilled in the art of trees to make use of the concept of a ‘level’. The root constitutes the first level of the tree, the child nodes of the root constitute the second level of the tree, the child nodes of those child nodes constitute the third level of the tree, etc.

“This method also presents the advantage of being easy to perform in parallel.

“The method greatly facilitates the task of the operator, which consists essentially in inputting constraints and the optimization criterion.

“The optimization criterion may seek to minimize a total area lost as a result of the cutting. In a variant, the operator may define some other optimization criterion, e.g. minimizing the number of sheets of glass that are used.

“In a particular implementation, the cutting constraints may be selected from among: a maximum number of cutting levels; a minimum width of scrap, e.g. greater than twice the thickness of the sheet; a direction for the first break, e.g. in the width direction of the sheet, or the maximum width of a cutting level, e.g. three meters, because of handling constraints.

“It should be observed that it is important not to confuse the concepts of ‘level in the tree’ and of ‘cutting level’, which are completely independent.

“In a particular implementation, the positioning constraints may be selected from among the orientations of the pieces in a sheet, the relative positions of the pieces in a single sheet as a function of their level, the maximum number of said at least one sheet of glass.

“In a particular implementation, a node of the tree possesses a maximum of 9.m child nodes, where m is the number of frames. Specifically, for each frame, a new piece selected in compliance with the positioning constraints may be added to a partial cutting plan and this can be done in nine different manners, namely: to the right of the preceding piece (in a cutting level 3), and either horizontally or vertically; above the preceding piece if these last two pieces have the same width (in a new cutting level 4), a single position suffices among horizontal and vertical, namely the position of width that is equal to the width of the preceding piece; above the preceding piece, at the left end of the current breadth (in a new cutting level 2), either horizontally or vertically; in a new breadth (in a new cutting level 1), either horizontally or vertically; and in a new sheet, either horizontally or vertically.

“In an optimized variant implementation, the creation of the tree comprises: a step consisting in creating, under the root and for each of the frames, a node associated with a partial cutting plan for each of the acceptable positions of the first piece of the frame while complying with said constraints; and at least one iteration, each iteration comprising: a selection step for selecting a current node of the tree as a function of characteristics of the partial cutting plan represented by that node; and a creation step for creating at least one child node of said current node, the cutting plan associated with said child node being obtained by adding to the partial cutting plan associated with said current node, and in compliance with the constraints, the next piece for a frame taken in accordance with the sequence predetermined for that frame.

“This variant is remarkable in that it comprises, on each iteration, a step of selecting the current node for which child nodes are to be created as a function of the characteristics of the partial cutting plan represented by this node.

“This characteristic makes it possible to select the current node that is the most promising for reaching the optimum solution, such that the method normally reaches the optimum solution more quickly.

“It is therefore advantageous, in a particular implementation, for the method to include a step of stopping the iterations if the duration of execution of the method is greater than a predefined duration so as to enable the operator to take cognizance of an optimized solution or of a good solution within a reasonable length of time.

“In a particular implementation, the method enables the user to have knowledge of the best solution at any time, while allowing the method to continue executing, potentially discovering better solutions.

“In a particular implementation, the current node is selected: in compliance with a first criterion referred to as the ‘minimum scrap criterion’ consisting in selecting the node associated with the cutting plan having the smallest ratio of the area lost divided by useful area; or in compliance with a second criterion referred to as the ‘maximum area criterion’ consisting in selecting the node associated with the cutting plan having the largest useful area.

“This strategy of exploring the tree, in other words of selecting the current node, consists in alternating between the two criteria of ‘minimum scrap’ and of ‘maximum area’.

“For example, the ‘minimum scrap criterion’ may be used during a defined length of time or else during a predetermined number of iterations or else until a certain fraction of the RAM of the computer is occupied (about two million nodes open).

“Applying the ‘minimum scrap’ criterion makes it possible to explore promising regions of the tree in which the partial plans that are created have minimum scrap and can lead to complete cutting plans that are close to the optimum solution. Nevertheless, applying this criterion leads to a plurality of new nodes being created, and if no limit is set, that runs the risk of saturating the memory of the machine.

“Thus, advantageously, in this implementation of the invention, when the above-mentioned limit is reached (in terms of time, number of iterations, or RAM occupation fraction), the exploration strategy is changed in order to apply the ‘maximum area’ criterion.

“This criterion advantageously tends towards selecting nodes for which the partial plan is close to the complete cutting plan, by selecting partial plans with the greatest area of pieces that have already been placed. This makes it possible to reach complete cutting plans more quickly, and to improve the best complete cutting plan that has been obtained so far.

“In order to save on memory space, in a particular implementation, only the leaf associated with the complete cutting plan that maximizes the optimization criterion is stored, the other leaves being deleted. In other words, the first leaf to be obtained is stored, and when a new leaf is obtained, only the leaf associated with the better complete cutting plan is retained in memory.

“In a particular implementation, the method includes a step of deleting the nodes of the tree that are associated with partial cutting plans for which the acquired scrap area is greater than the area lost from a complete cutting plan associated with a said leaf.

“This implementation may advantageously be combined with the above-described implementation that alternates between the ‘minimum scrap criterion’ and the ‘maximum area criterion’.

“Specifically, if a better complete cutting plan is obtained, then a pruning operation is performed on all of the current nodes in an attempt to delete unpromising nodes that cannot make it possible to reach a better complete cutting plan. This is done by comparing the geometrical loss of a node that has not yet been explored with the loss of the best solution obtained so far; in other words, if the loss of a node is greater than the loss of the best node, then that node is deleted. This makes it possible to reduce the number of nodes that remain to be explored, thereby releasing memory. This criterion is also applied for a certain length of time or a predefined number of iterations, and then the ‘minimum scrap’ criterion is applied once more in order to create new promising nodes.

“The procedure consists in alternating between these two criteria, i.e. in alternating between creating new promising nodes and in obtaining improved new solutions that enable less-promising nodes to be deleted.

“This concept of ‘acquired scrap’ is summarized together with other concepts with reference to FIG. 7, which shows a partial cutting plan PDP.

“In this partial plan, it is known to define three types of area, namely: the useful area containing pieces, which in this figure are numbered with the number of the frame for which they are intended; acquired scrap (shaded), i.e. area that is unused and in which it is not possible to add any new piece; and the area (stippled) that can still be used for placing subsequent additional pieces (pale gray).

“Thus, in a particular implementation, when it is determined that a node of the tree is associated with a partial cutting plan for which the acquired scrap area is already greater than the lost area of a complete cutting plan, it is known that this node can be deleted since there are no circumstances in which it can give rise to a complete cutting plan that offers a better solution.

“This particular implementation makes it possible to prune the tree and to reduce considerably the execution time of the method.

“In a particular implementation, the optimization method of the invention avoids or minimizes creating nodes in the tree that corresponded to partial cutting plans that are isomorphic, i.e. cutting plans comprising the same pieces and presenting the same acquired scrap area. By way of example, the partial cutting plans PDPA and PDBP of FIG. 9 are isomorphic.

“Thus, in a particular implementation, said positioning constraints include at least one lexicographic constraint relating to a number of said frames in order to avoid or to minimize creating nodes that correspond to partial cutting plans that are isomorphic.

“For example, the positioning constraints may include two lexicographic rules, according to which: if the last piece is placed above the preceding piece, the frame number of the last piece must be less than the frame number of the preceding piece; if the last piece is placed to the right the preceding piece, the frame number of the last piece must be greater than or equal to the frame number of the preceding piece.

“In a particular implementation, each time a node is created, the node is classified as a function of at least one characteristic of the cutting plan represented by that node, this or these characteristics being sufficient for selecting the complete cutting plan.

“This characteristic is preferably the characteristic used during said step of selecting the current node of the tree in the optimized implementation variant.

“This implementation can avoid the need to scan once more through the entire tree in order to select the optimized complete cutting plan and in order to select the current node in the optimized implementation variant.

“In a particular implementation, the other nodes of the tree, including its leaves, are represented in memory by the frame number of the last piece to be placed, and by the direction in which it was placed.

“This implementation serves to minimize memory occupation.

“The invention also provides a device for determining an optimized cutting plan for using a guillotine to cut a batch of rectangular pieces of glass out from at least one sheet of glass, the pieces, once cut out, being for stacking on at least one frame, the pieces for any one frame being for placing on a sheet for cutting in a sequence that is predetermined for that frame, said device comprising: an initialization module for defining cutting constraints and positioning constraints for the pieces together with an optimization criterion; a module for creating a tree comprising a root, leaves, each presenting a complete cutting plan enabling all of the pieces of the batch to be cut out, each other node of the tree representing a partial cutting plan, the cutting plan associated with a node of the tree being obtained by adding to the partial cutting plan associated with the parent node of said node, and in compliance with the constraints, the next piece for a frame determined in compliance with the sequence predetermined for the frame; and a selection module for selecting a complete cutting plan associated with a leaf of the tree as a function of said optimization criterion.

“The cutting plans obtained by the method of the invention may be used in particular: for creating optimized cutting batches upstream from the cutting line; during cutting proper, where cutting consists mainly in propagating cracks across the glass, in an order that complies with the cutting plan; and for assisting the operator while breaking the sheet of glass in order to obtain pieces that are to be placed on the various frames.

“Consequently, the invention also provides a cutting method for using a guillotine to cut a batch of rectangular pieces of glass out from at least one sheet of glass, the method being characterized in that it comprises: performing a method of determining an optimized cutting plan as mentioned above; said optimized plan being used during a stage of cutting said sheet and during a stage of breaking said sheet.

“In a particular implementation of this method, the batch is itself determined by performing a method of determining an optimized cutting plan as described above.

“In a particular implementation, the various steps of the method of determining an optimized cutting plan of the invention are determined by computer program instructions.

“Consequently, the invention also provides a computer program on a data medium, the program including instructions adapted to performing steps of a method of determining an optimized cutting plan of the invention.

“The program may use any programming language, and it may be in the form of source code, object code, or code intermediate between source code and object code, such as in a partially compiled form, or in any other desirable form.

“The invention also provides a computer readable data medium, including instructions of a computer program as mentioned above.

“The data medium may be any entity or device capable of storing the program. For example, the medium may comprise storage means, such as a read only memory (ROM), e.g. a compact disk (CD) ROM or a microelectronic circuit ROM, or indeed magnetic recording means, e.g. a hard disk.

“Furthermore, the data medium may be a transmissible medium such as an electrical or optical signal that can be conveyed via an electrical or optical cable, by radio, or by other means. The program of the invention may in particular the downloaded from a network of the Internet type.

“Alternatively, the data medium may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method in question.”

The claims supplied by the inventors are:

“1. A method performed by computer to determine an optimized cutting plan for using a guillotine to cut a batch of rectangular pieces of glass out from at least one sheet of glass, the pieces, once cut out, being for stacking on at least one frame, the pieces for any one frame being for placing on said at least one sheet that is to be cut up in a sequence that is predetermined for that frame, said method comprising: initializing including defining cutting constraints and positioning constraints for said pieces together with an optimization criterion; creating a tree comprising a root, leaves, each presenting a complete cutting plan enabling all of the pieces of said batch to be cut out, each other node of the tree representing a partial cutting plan, the cutting plan associated with a node of the tree being obtained by adding to the partial cutting plan associated with the parent node of said node, and in compliance with said constraints, the next piece for said frame determined in compliance with the sequence predetermined for the frame; and selecting a complete cutting plan associated with a leaf of the tree as a function of said optimization criterion.

“2. The method according to claim 1, wherein the optimization criterion is selected from among: a criterion seeking to minimize the number of sheets of glass used; and a criterion seeking to minimize the total area lost as generated by cutting.

“3. The method according to claim 1, wherein said cutting constraints may be selected from among: a maximum number of cutting levels; a minimum width of scrap; a maximum width of a cutting level; and a direction for the first break.

“4. The method according to claim 1, wherein said positioning constraints are selected from among orientations of the pieces in a sheet, relative positions of the pieces in a single sheet as a function of their level, and a maximum number of said at least one sheet of glass.

“5. The method according to claim 1, wherein a node of the tree possesses a maximum of 9.m child nodes, where m is the number of frames, a new piece being selectable for each frame while complying with the positioning constraints of that frame and is configured to be added to a partial cutting plan in nine different manners, namely: to the right of the preceding piece (in a cutting level 3), either horizontally or vertically; above the preceding piece if these last two pieces have the same width (in a new cutting level 4), a single position suffices among horizontal and vertical, namely the position of width that is equal to the width of the preceding piece; above the preceding piece, at the left end of the current breadth (in a new cutting level 2), either horizontally or vertically; in a new breadth (in a new cutting level 1), either horizontally or vertically; and in a new sheet, either horizontally or vertically.

“6. The method according to claim 1, wherein said creating the tree comprises: creating, under said root and for each of said frames, a node associated with a partial cutting plan for each of the acceptable positions of the first piece of said frame while complying with said constraints; and at least one iteration, each iteration comprising: selecting a current node of the tree as a function of characteristics of the partial cutting plan represented by that node; and creating at least one child node of said current node, the cutting plan associated with said child node being obtained by adding to the partial cutting plan associated with said current node, and while complying with said constraints, the next piece of a said frame taken in accordance with the sequence predetermined for that frame.

“7. The method according to claim 6, further comprising stopping said iterations if the duration of execution of the method is greater than a predefined duration.

“8. The method according to claim 6, wherein said current node is selected: in compliance with a first criterion referred to as the “minimum scrap criterion” consisting in selecting the node associated with the cutting plan having the smallest ratio of the area lost divided by the total area occupied by the pieces of said plan; or in compliance with a second criterion referred to as the “maximum area criterion” consisting in selecting the node associated with the cutting plan having the largest useful area.

“9. The method according to claim 1, wherein only the leaf associated with the complete cutting plan that maximizes said optimization criterion is stored, the other leaves being deleted.

“10. The method according to claim 1, further comprising deleting the nodes of the tree that are associated with partial cutting plans for which the acquired scrap area is greater than the area lost from a complete cutting plan associated with said leaf.

“11. The method according to claim 1, wherein said positioning constraints include at least one lexicographic constraint relating to a number of said frames in order to avoid or to minimize creating nodes that correspond to partial cutting plans that are isomorphic.

“12. The method according to claim 1, wherein each time a node is created, said node is classified as a function of at least one characteristic of the cutting plan represented by that node, said at least one characteristic being sufficient for selecting said complete cutting plan.

“13. The method according to claim 6, wherein each time a node is created, said node is classified as a function of at least one characteristic of the cutting plan represented by that node, said at least one characteristic being sufficient for selecting said complete cutting plan, and said at least one characteristic used for said classification is said at least one characteristic used during said step of selecting the current node of the tree.

“14. A device for determining an optimized cutting plan for using a guillotine to cut a batch of rectangular pieces of glass out from at least one sheet of glass, the pieces, once cut out, being for stacking on at least one frame, the pieces for any one frame being for placing on said at least one sheet for cutting in a sequence that is predetermined for that frame, said device comprising: an initialization module for defining cutting constraints and positioning constraints for said pieces together with an optimization criterion; a module for creating a tree comprising a root, leaves each presenting a complete cutting plan enabling all of the pieces of said batch to be cut out, each other node of the tree representing a partial cutting plan, the cutting plan associated with a node of the tree being obtained by adding to the partial cutting plan associated with the parent node of said node, in compliance with said constraints, the next piece of said frame determined in compliance with the sequence predetermined for the frame; and a selection module for selecting a complete cutting plan associated with a leaf of the tree as a function of said optimization criterion.

“15. (canceled)

“16. A non-transitory computer readable medium and storing a computer program including instructions for executing steps of the method according to claim 1 for determining an optimized cutting plan.

“17. A cutting method for using a guillotine to cut a batch of rectangular pieces of glass out from at least one sheet of glass, the method comprising: performing the method according to claim 1 for determining an optimized cutting plan; and using said optimized plan during a stage of cutting said sheet and during a stage of breaking said sheet.

“18. The cutting method according to claim 17, wherein said batch is determined by determining an optimized cutting plan.”

For additional information on this patent application, see: LUCAS, Claire; SAUSSET, Francois; TLILANE, Lydia. Method And Device For Optimizing A Plan For Cutting By Guillotine Of Pieces Of Glass. Filed September 7, 2017 and posted July 25, 2019. Patent URL: http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.html&r=1&f=G&l=50&s1=%2220190227514%22.PGNR.&OS=DN/20190227514&RS=DN/20190227514

(Our reports deliver fact-based news of research and discoveries from around the world.)

Copyright © 2019 NewsRx LLC, Computer Business Daily, source Technology Newsletters

share with twitter share with LinkedIn share with facebook
share via e-mail
0
Latest news on SAINT-GOBAIN
08/14Adhesives Market Is Expected To Grow At A CAGR Of 9.0% Till 2022
AQ
08/13SAINT GOBAIN : Fifth weber plant opened in indonesia
AQ
07/313M : and SGA Settle Patent Dispute Relating to Paint Spray Technology
AQ
07/25SAINT GOBAIN : Construction materials group St. Gobain's first half profits rise
RE
07/25SAINT GOBAIN : H1-2019 Results
PU
07/25Saint-Gobain Backs guidance After 1st Half Net Profit Fell
DJ
07/22Investors With $2 Trillion in Assets Urge Cement Makers to Cut Emissions
DJ
07/22Funds managing $2 trillion urge cement makers to act on climate impact
RE
07/18SAINT GOBAIN : enters into exclusive negotiations to sell its civil engineering ..
PU
07/18SAINT GOBAIN : in Talks to Sell French Concrete-Products Distribution Business
DJ
More news
Financials (EUR)
Sales 2019 42 964 M
EBIT 2019 3 336 M
Net income 2019 1 740 M
Debt 2019 9 426 M
Yield 2019 4,39%
P/E ratio 2019 9,92x
P/E ratio 2020 9,01x
EV / Sales2019 0,62x
EV / Sales2020 0,60x
Capitalization 17 335 M
Chart SAINT-GOBAIN
Duration : Period :
Saint-Gobain Technical Analysis Chart | MarketScreener
Full-screen chart
Technical analysis trends SAINT-GOBAIN
Short TermMid-TermLong Term
TrendsBearishNeutralNeutral
Income Statement Evolution
Consensus
Sell
Buy
Mean consensus OUTPERFORM
Number of Analysts 22
Average target price 40,34  €
Last Close Price 31,91  €
Spread / Highest target 56,7%
Spread / Average Target 26,4%
Spread / Lowest Target -12,3%
EPS Revisions
Managers
NameTitle
Pierre-André de Chalendar Chairman & Chief Executive Officer
Benoît Bazin Chief Operating Officer
Sreedhar Natarajan Chief Financial Officer
Armand Ajdari Vice President-Innovation, Research & Development
Frédéric Verger Chief Information Officer
Sector and Competitors
1st jan.Capitalization (M$)
SAINT-GOBAIN9.39%18 965
ASSA ABLOY36.07%24 408
MASCO29.96%10 999
FORTUNE BRANDS HOME & SECURITY, INC32.98%6 956
TOTO LTD9.63%6 443
AGC INC-12.26%6 207