Applying Bayesian networks for intelligent adaptable printing systems

Bayesian networks are around more than twenty years by now. During the past decade they became quite popular in the scientific community. Researchers from application areas like psychology, biomedicine and finance have applied these techniques successfully. In the area of control engineering however, little progress has been made in the application of Bayesian networks. We believe that these techniques are useful for systems that dynamically adapt themselves at runtime to a changing environment, which is usually uncertain. Moreover, there is uncertainty about the underlying physical model of the system, which poses a problem for modelling the system. In contrast, using a Bayesian network the needed model can be learned, or tuned, from data. In this paper we demonstrate the usefulness of Bayesian networks for control by case studies in the area of adaptable printing systems and compare the approach with a classic PID controller. We show that it is possible to design adaptive systems using Bayesian networks learned from data.