We report on a project with a German car manufacturer. The task is to compute (approximate) solutions to a specific large-scale packing problem. Given a polyhedral model of a car trunk, the aim is to pack as many identical boxes of size 4 × 2 × 1 units as possible into the interior of the trunk. This measure is important for car manufacturers, because it is a standard in the European Union. First, we prove that a natural formal variant of this problem is NP-complete. Further, we use a combination of integer linear programming techniques and heuristics that exploit the geometric structure to attack this problem. Our experiments show that for all considered instances, we can get very close to the optimal solution in reasonable time. © Springer-Verlag 2003.
CITATION STYLE
Eisenbrand, F., Funke, S., Reichel, J., & Schömer, E. (2003). Packing a Trunk. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2832, 618–629. https://doi.org/10.1007/978-3-540-39658-1_56
Mendeley helps you to discover research relevant for your work.