I can't share the code I wrote since it was for a client, but I suppose I can talk in general terms about the approach. It may not interest most people on this list though, which is why I suggested that the OP email me privately.
There are a lot of variables to consider such as: Can the pieces be rotated at any angle, only by 90 degrees or only by 180 degrees? Rotation restrictions are related to the stock material having a "grain" that is significant. Do you need to leave a gap between the pieces to allow for a cutter that will consume some of the material? Do you have a large number of identical pieces to be cut or do they mostly differ in their shape? When there are lots of identical, non-rectangular pieces, it is useful to find an efficient way to arrange a pair of them and then find an efficient way to arrange the pairs. When the pieces vary in their shape and they are not rectangles, it becomes an NP-complete problem that I addressed with a set of tunable heuristics. My code created multiple candidate layouts using various tuning values and keep the smallest layout. Each layout can be created in parallel with the others. It seems to work best to place the largest pieces first and then fill in the gaps with smaller pieces. On Fri, Sep 10, 2010 at 12:59 PM, Robert Casto <[email protected]> wrote: > Are these proprietary algorithms or can they be shared with the group? I'm > interested in learning how you would go about it. I'm guessing others might > be too. Solving problems is the fun part of programming. > > On Fri, Sep 10, 2010 at 1:56 PM, Mark Volkmann <[email protected]> > wrote: >> >> I wrote software like this many years ago. Email me privately to discuss >> it. >> >> Are the pieces rectangular or are they arbitrary shapes? >> >> On Fri, Sep 10, 2010 at 7:48 AM, TC <[email protected]> wrote: >> > I'm re-writing an application that takes x amount of pieces and tries >> > to fit them best to the right number and size of stock material. So >> > given 2 pieces at one size 1 at a larger size and 3 at a larger size >> > find the most efficient cuts from the stock material. >> > >> > My Problem: >> > The way we currently handle this is to calculate linear footage and >> > then step through a piece array, twice. Attempting to match the >> > footage of the piece to the smallest core we can, along with as many >> > other pieces possibly. I would like to optimize this as much as >> > possible, and I'm wondering if there is some sort of equation to >> > figure this out. I did a few searches but its hard to know what to >> > search for. >> > >> > I apologize for not being able to post code, but its long and messy >> > anyways. Any help is appreciated.(Using Java) >> > >> > -- >> > You received this message because you are subscribed to the Google >> > Groups "The Java Posse" group. >> > To post to this group, send email to [email protected]. >> > To unsubscribe from this group, send email to >> > [email protected]. >> > For more options, visit this group at >> > http://groups.google.com/group/javaposse?hl=en. >> > >> > >> >> >> >> -- >> R. Mark Volkmann >> Object Computing, Inc. >> >> -- >> You received this message because you are subscribed to the Google Groups >> "The Java Posse" group. >> To post to this group, send email to [email protected]. >> To unsubscribe from this group, send email to >> [email protected]. >> For more options, visit this group at >> http://groups.google.com/group/javaposse?hl=en. >> > > > > -- > Robert Casto > www.robertcasto.com > > > -- > You received this message because you are subscribed to the Google Groups > "The Java Posse" group. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/javaposse?hl=en. > -- R. Mark Volkmann Object Computing, Inc. -- You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/javaposse?hl=en.
