Research On The Bin-Packing Problem Based On Hybrid Group Genetic Algorithm

The Bin-packing problem (BPP) is a kind of very typical NP-hard problem. And it is of important theoretical value and practical significance. The common purpose of BPP is putting some "goods" into some "boxes" while the used "boxes" number is the smallest. It has a wide range of applications such as the resource allocation of computer operating system, the fabric cutting of apparel industry, the collation objects in our daily life etc. On the other hand, Bin-packing problem as one of the earlier study of NP-hard problem, the study of the issue provides an important theoretical research significance to research other NP-hard problem. So, there is an important value in studying the BPP.The BPP has aroused man's attention from the early 1970s, and it has been studied for nearly forty years. Many combinatorial optimization researchers have shown a great interest in the BPP and built up integrate theories and explored many algorithms for the problem. Even so, the study of the issue is far from over.The article utilizes Genetic Algorithm (GA) in solving BPP on the basic of the superiors, by integrating with the approximation algorithm of BF, FFD and group genetic algorithm, a hybrid group genetic algorithm is presented to solve the bin-packing problem. The main idea is to design a fitness function, using group genetic algorithm combined with the BF algorithm and FFD algorithm to optimize it, and an available bin-packing result is obtained. It is to realize the algorithm by C++ and experiments results based on the bin-packing instance show that it is an available method to solve the bin-packing problem.