Optimization and Simulation: Sequential Packing of Flexible Objects Using Evolutionary Algorithms
We want to fill a given two-dimensional closed contour as accurately as possible with a fixed number of identical, flexible objects. These objects have to be packed sequentially. They adapt themselves to the surface they are packed on, but their deformation can only be simulated. This type of problem is the two-dimensional cross-section of manufacturing processes where soft material is wound onto a mandrel. We formulate this as a problem of dynamic programming with simulated law of motion. It allows an evolutionary algorithm approach that successfully produces approximate solutions in a non-sequential fashion.
KeywordsEvolutionary algorithms optimization and simulation packing problems dynamic programming
Unable to display preview. Download preview PDF.
- 4.Scheithauer, G.The Solution of Packing Problems with Pieces of Variable Length and Additional Allocation Constraints Optimization, 34 (1995), 81–86.Google Scholar
- 5.Z. Michalewicz. Genetic Algorithms + Data Structures = Evolutionary Programs. Springer Verlag, 1992.Google Scholar
- 6.T. J. Reinhart. Composites. Engineered materials handbook, ASM International /Handbook Committee, Metals Park, Ohio: ASM International, 1988.Google Scholar