In laser cutting, many factors influence cutting quality. In this paper an intelligent system has been developed in order that the cutting parameters can be created quantitatively. This system is different from database and normal expert system. It combines factor experiment method and artificial neural network (ANN). Only a few tests are required to obtain basic data and then the test results are input into the ANN system. After the ANN is trained and learns from the test results, the system can produce laser cutting knowledge and optimized parameters. Based on test data and knowledge the cutting parameters given by the system will be more accuracy and reliable. Finally the intelligent system is applied to select 3D position laser cutting parameters. It shows the system being very effective and useful.