[comp.text.tex] 2-dimensional bin packing problem

hoch@ipsi.UUCP (Rainer Hoch) (02/12/91)

I am interested in the following 2-dimensional bin packing  prob-
lem:

The problem requires to pack n rectangular items into  a  minimal
number  of rectangular unit capacity bins. I am interested in the
case when the length and the width  of  the  items  are  not  too
weird.  In  fact I can assume that the ratio of the length (resp.
															width) of
the unit and the length (resp. width)  of  an  item  is
equal to 1,2,3,4 or 5 (i.e. it is integral).

A possible application would be the problem of how to place n ar-
ticles on a minimal number of pages in a newspaper.

Any hints, solutions or pointers to algorithms in  the literature
are welcome.

Thanks Volker Turau

Adress:  e-mail hoch@darmstadt.gmd.dbp.de   or
                turau@darmstadt.gmd.dbp.de
        ean-address: s=hoch;ou=darmstadt;p=gmd;a=dbp;c=de or
                     s=turau;ou=darmstadt;p=gmd;a=dbp;c=de

xanthian@zorch.SF-Bay.ORG (Kent Paul Dolan) (02/14/91)

 hoch@ipsi.UUCP (Rainer Hoch) writes:

> I am interested in the following 2-dimensional bin packing problem:

> The problem requires to pack n rectangular items into a minimal number
> of rectangular unit capacity bins. I am interested in the case when
> the length and the width of the items are not too weird. In fact I can
> assume that the ratio of the length (resp. width) of the unit and the
> length (resp. width) of an item is equal to 1,2,3,4 or 5 (i.e. it is
> integral).

> A possible application would be the problem of how to place n ar-
> ticles on a minimal number of pages in a newspaper.

> Any hints, solutions or pointers to algorithms in the literature are
> welcome.

> Thanks Volker Turau

> Address: e-mail hoch@darmstadt.gmd.dbp.de or
> turau@darmstadt.gmd.dbp.de ean-address:
> s=hoch;ou=darmstadt;p=gmd;a=dbp;c=de or
> s=turau;ou=darmstadt;p=gmd;a=dbp;c=de

Considering where I encountered this problem, what a strange place
(comp.text.sgml) to find it again!

A very similar problem is studied under the names "floor space control",
"shop floor control", "floor space scheduling", and "platten
scheduling"; in each case, what is being scheduled is rectangular work
areas in which major mechanical equipment is constructed. The problem in
its full blown nature is very complex. Plattens, for example, are the
outdoor areas on which major component pieces of ships are welded up. In
this case, you are actually working in four dimensions, as you try to
schedule not only the flat space, but also clearances at particular
points in time to lift the finished pieces out over other,
under-construction pieces with a limited vertical clearance bridge
crane.

"Floor space" is more usually such items as massive trucks and other
earth moving equipment being constructed in parts within a shelter; the
scheduling concerns are similar to those for plattens (my own research
interest).

At any rate, eight years back, there were commercial heuristic schedulers
for such uses.  Consult journals and trade publications for the earthmoving,
shipbuilding, and oil exploration industries for ads or articles under the
above topics.

A generalization of your problem which may prove useful also is "nesting",
the science of cutting the most parts out of the least material.  In the
two dimensional case, this is interesting in the metal fabrication and
garment cloth layout tasks.  Again, all the work I know is in the area of
heuristics; the problem is considered too difficult for a closed solution.

I hope these rather general hints will provide a direction to your search.

Kent, the man from xanth.
<xanthian@Zorch.SF-Bay.ORG> <xanthian@well.sf.ca.us>