As industrial processes have become more and more complex,the traditional sensors are unavailable,and it is difficult to properly model the critical variables and efficiently optimize or diagnose the important parts of a process.The Gaussian process model provides an alternative to the modeling,optimization and control of industrial processes under the constraints of uncertainties.In this paper,aiming at the complexity of industrial processes,the pros and cons of the Gaussian process model are investigated,and its application to the modeling,optimization,control and fault diagnosis of industrial processes as well as the corresponding research results is generalized.Final- ly,the application prospects and development orientations of the Gaussian process model in industrial processes are summarized and forecasted by combining the international research results with the authors' practical experience.