Review and comparison of algorithms and software for mixed-integer derivative-free optimization

Abstract This paper reviews the literature on algorithms for solving bound-constrained mixed-integer derivative-free optimization problems and presents a systematic comparison of available implementations these large collection test problems. Thirteen solvers are compared using set 267 The testbed includes: (i) pure-integer problems, (ii) small, medium, covering wide range characteristics found in applications. We evaluate according to their ability find near-optimal solution, best solution among currently solvers, improve given starting point. Computational results show that all obtain good solutions diminishes with increasing problem size, but evaluated collectively optimal 93% our set. open-source MISO NOMAD were performers tested. outperformed other binary while was performer mixed-integer, non-binary discrete, medium-sized