The problem of real-time clustering has gained considerable attention in recent years in conjunction with the advances in the areas of multiple model representation of complex systems, summarization of information, and novelty detection for diagnostics and prognostics. This chapter deals with two main approaches for real-time clustering – the first algorithm is density based and is derived from the Mountain/Subtractive clustering method while the second one is distance based and has its roots in the k-nearest neighbors (k-NN) and self-organizing maps (SOM) clustering methods. Applications of these algorithms for extraction of rules from data, for control of complex systems with multiple operating modes, fault detection, and prognostics are presented in the chapter. (c) John Willey and Sons