New Project: Visualization of the Best-Fit Placement Heuristic

admin — Sun, 19 Jul 2009 11:00:45 +0000

Often one has to solve the problem of arranging objects so that they use as less room as possible. In the days before Christmas, for example, people try to utilize the expensive wrapping paper fully in order to save to money for a second roll. Depending on the arrangement of the presents this may work, or not. In the industry, a can factory tries to stamp different sized cans out of a roll of sheet metal. To minimize the waste (and to save money) the company has to think about the cutting of the metal sheet.
Both of these examples describe a problem known as the [two-dimensional cutting stock problem](http://en.wikipedia.org/wiki/Cutting_stock_problem).

Although there is no algorithm available which produces an optimal solution to this problem several heuristics can identify usable solutions within an acceptable timeframe.
One of these heuristics is the *Best-Fit* heuristic (described [here](http://or.journal.informs.org/cgi/content/abstract/52/4/655)), developed by [E. K. Burke](http://www.cs.nott.ac.uk/~ekb/), [G. Kendall](http://www.cs.nott.ac.uk/~gxk/) and [G. Whitwell](http://www.cs.nott.ac.uk/~gxw/home-gxw.shtml) of the [University of Nottingham](http://www.nottingham.ac.uk/).

[This project](/projects/visualization-of-the-best-fit-placement-heuristic/) visualizes the different steps the *Best-Fit* heuristic performs in order to gain an acceptable solution (placing rectangles in a way so that waste is reduced) for a given set of different sized rectangles.

**Update:** Source code is [now available](http://github.com/simonboots/BestFitPlacementHeuristic/).

stiefels.net » cutting stock problem

New Project: Visualization of the Best-Fit Placement Heuristic