Multi-objective sequence optimization of PCB component assembly with GA based on the discrete Frechet distance

A new mechanism,namely a combination of curve matching method based on the discrete Frechet distance and evolutionary algorithms,is proposed to solve pick-and-place sequence optimisation problems as a multi-objective optimisation problem. The essence of the mechanism is to accomplish the comparison of objective vectors with curve matching method. The objective vector is mapped into the array of points with a binary mapping operator and the discrete Frechet distance is utilised to measure the similarity between the reference array of points and the comparison array of points. The genetic algorithm based on the discrete Frechet distance (FGA) is proposed. To test the new mechanism, together with FGA, three other test algorithms are selected to solve the sequence optimisation problem. The simulation results indicate that FGA outperforms other algorithms. This new mechanism is rational and feasible for multi-objective pick-and-place sequence optimisation problems.