ABSTRACT
Hybrid semiparametric models make simultaneous use of the parametric and nonparametric modeling paradigms to solve complex problems. The main advantage of the semiparametric over the parametric or nonparametric approaches lies in that it broadens the knowledge base that can be used to solve a particular problem. In many complex chemical or biological engineering problems, it is unlikely that the system can be fully described either by a mechanistic (parametric) or by an empirical (nonparametric) approach. Opting for one or the other framework will invariably promote reductionism. In contrast, the complementary use of both types of knowledge permits expansion of the system toward more global descriptions of the process at hand. In this chapter, the focus is on hybrid semiparametric model structures. The main structures presented in the literature will be overviewed in the context of integration of different forms of knowledge—first-principles, mechanistic, heuristic, and empirical knowledge. Finally, the application of semiparametric systems for knowledge integration will be illustrated with two examples.
