Abstract: A Layout Pattern Based Particle Swarm Optimization algorithm (LPPSO) is presented herein for solving the two-dimensional packing problem with constraints. In the optimizing process of LPPSO, some individuals are constructed according to non-isomorphic layout pattern and these individuals are added into the current population of Particle Swarm Optimization (PSO) algorithm to replace the bad individuals, the new population is created as a result. Moreover, a non-isomorphic layout pattern is constructed based on exact boundary line approach to avoid premature convergence and improve the computational efficiency. This study also discussed the basic idea, key problem and the process of the proposed LPPSO algorithm. Two examples of constrained packing problems showed that LPPSO was feasible and effective in the experiments.