A hybrid ant colony algorithm for solving container loading

Container loading is a combinatorial optimization problem in three-dimensional packing which is an area of cutting and packing. It belongs to a NP-hard problem. To solve this problem, a hybrid approach is proposed in this paper. The approach integrated a loading heuristic into an ant colony optimization algorithm. The loading sequence is generated by the ant colony algorithm which configures the quantity of pheromone trails, transition probability and objective function. The loading sequence is transformed into the loading layout of boxes by using a loading heuristic which combines the heuristic strategy of generated blocks and the method for handling remaining spaces. The approach is tested through the test instance. The experimental result has shown that the approach is superior to the DNA generic algorithm.