After a response surface method (RSM) experiment is performed, one of the critical issues becomes selection of a good empirical model. Automatic model selection tools have become very popular, especially in situations where there are a large number of factors. While these tools are fast and easy to use, some caution must be exercised, particularly if the design space is constrained (e.g. a mixture design) or irregularly shaped. In this webinar, Martin Bezener provides a brief overview to model selection in RSM, and discusses the pros and cons of several automatic model selection techniques and criteria.