The work aims to solve the low efficiency of the steel-coil packaging production line with improved genetic algorithm. Self-adaption was used to change the crossover probability and mutation probability. When the fitness of the populations tended to be consistent, the crossover probability and mutation probability should be increased; when the fitness of the populations differed greatly, the smaller crossover probability and mutation probability should be used. The balance rate of the optimized packaging production line reached by 88.46%. The improved packaging production line had a better balancing effect. The improved genetic algorithm adopted can effectively solve the balance problem of single-variety steel-coil packaging production line.