一类0-1背包问题的贪婪遗传算法

new HGA for a class of 0-1KP called Greedy Genetic Algorithm is presented. GGA turns greedy principle to advantage. The greedy approach is used for generating the first chromosome, and made use of the greedy-operator to improve some feasible solutions further in the course of evolution. GGA uses larger mutation probability and smaller crossover probability at the same time to overcome GA’s limitation, mutation operation in GA plays an important and fundamental role in. Numerous computational experiments have been carried out on about 400,000 well-structured new classes of instances, and a comparative analysis with some recent state-of-art exact algorithms for the 0-1KP is given. It is found that the average computing time of GGA grow to be less than the time of the exact algorithms in company with an increase in coefficients. The average computing time to search for an optimal solution is between 0.3ms and 430.6ms, and the average generation is 1 to 40.