The modern modeling of flotation processes has been burdened for years with limitations of classical mathematics and modeling. The emergence of soft computing methods has also been looked upon in many of the industry's branches as promising, and up to a certain point, those expectations have been met. Today, these soft computing methods are used regularly in research, and in some cases, within different industrial practices. This paper gives a review regarding the most common soft computing methods and their use in flotation processes modeling. Artificial neural networks have received the widest application in this area, and are followed by fuzzy logic, genetic algorithms, support vector machines and learning decision trees. Over the last five years, the number of reported studies within this field, has steadily increased. And although several classes of flotation problems are being successfully modeled with soft computing methods, there still remain a number of unresolved issues and obstacles. This paper thus attempts to provide an explanation for the current state and use of soft computing methods, as well as to present some ideas on future initiatives and potential developments within the area. (C) 2015 Elsevier Ltd. All rights reserved.