SPOT is a (open source) toolbox for surrogate model based optimization. SPOT uses Design of Experiment methods, various statistical modeling techniques, classical optimization algorithms and meta-heuristics to solve expensive (ressource/time-consuming) optimization problems. The sequential parameter optimization (SPO) concept was originally developed by Thomas Bartz-Beielstein for the purpose of algorithm tuning, but has since been applied to a large variety of problems.
The R-version combines contributions of several people (see: external link: https://cran.r-project.org/package=SPOT) and is currently maintained by myself (i.e., feel free to send me your questions w.r.t. the package). Recently, we uploaded a completely rewritten version of the package (v2.0.1), to create a more intuitive, modular interface and enhance the user experience.