A modified tabu search with the diversification perturbation operator and mutation operator is developed based
on the mounting process optimization problem for the multi-head surface mounting machine. Based on the traditional tabu search algorithm, the diversification perturbation based on long-term frequency information and a mutation operator are used to expand the search place. A local descent search strategy is embedded into the algorithm to optimize the feeder assignment. Experimental results show that the proposed algorithm can obtain satisfied near-optimal solutions to the mounting process optimization in a short length of time and is superior to the algorithms presented in literature in terms of both solution quality and computation speed.