This paper focuses on a framework for probabilistic supervisory control of probabilistic discrete event systems (PDES). PDES are modelled as generators of probabilistic languages, and the supervisors used are probabilistic. In our previous work, we presented and solved a number of supervisory control problems inside the framework. We also suggested a pseudometric to measure the behavioural similarity between PDES, and used the pseudometric in the solution of two optimal supervisory control problems defined in the framework. In this paper, we survey these results and introduce a real-world application of the framework. Further, we investigate a relationship between our framework and that of Markov Decision Processes, that could prove beneficial for both control synthesis and probabilistic model checking.