Handling constraints with Derivative-Free Optimization X.Fu, A. R. Almomani, M. Minick, K.R.Fowler Department of mathematics and computer science Derivative free optimization (DFO) methods are designed to solve optimization problems whose gradient information is unavailable or unreliable and thus the minimization is guided solely by function values. Nelder-Mead, Genetic Algorithm, Implicit Filtering and Particle Swarm Optimization are some of the commonly used DFO methods. On the other hand, there is little guidance for DFO methods to solve problems with linear and non-linear constraints which define the feasible search region. Typical methods to handle constraints include the penalty and the barrier methods. In this study, we seek to identify the best-performing pair of DFO and constrained methods based on numerical experiments using a suite of constrained optimization test problems. Our current focus is on collecting performance measures to assess algorithm behavior based on properties of objective functions such as non-smoothness, local minima, low amplitude noise and discontinuous domains. {2011}, {Applied Mathematics}, {Honors Program}, {K.R.Fowler}