Two-dimensional irregular polygons packing problem is very difficult to be solved in traditional opti-mal way. Simulated annealing (SA) algorithm is a stochastic optimization technique that can be used to solve packing problems. The whole process of SA is introduced firstly in this paper. An extended neighborhood searching method in SA is mainly analyzed. A general module of SA algorithm is given and used to lay out the irregular polygons. The judgment of intersection and other constrains of irregular polygons are analyzed. Then an example that was used in the paper of Stefan Jakobs is listed. Results show that this SA algorithm shortens the computation time and improves the solution.