Assembly planning based on genetic algorithms

A genetic algorithm that generates and evaluates assembly plans is proposed. This algorithm is able to generate near-optimal assembly plans through purposely developed crossover and mutation operators starting from a randomly initialized population of assembly sequences. The optimization criteria we used to assess the quality of feasible assembly sequences are: (i) to minimize the orientation changes of the product; (ii) to minimize the gripper changes, and (iii) to group technologically similar assembly operations together as much as possible. Experimental results that confirm the validity of our approach are also included.