Increasing the automation level of systems has led to more attention being paid to human control roles and the dependability of systems that comprise operators and machines. When humans build mental models of complex automated systems, they can be faced with problems regarding situation awareness. Moreover, these complex systems may lead to knowledge-based errors in situations where decision-making has to be performed. To cope with such situations, one may wish to consider advisory systems. An advisory system is a type of decision support system that tackles the complexity of systems with regard to human judgement for decision-making. This paper describes an approach to systems, advisory systems and their human use for safe operation. The intention is to initiate a future path for a formal description of systems in which an advisory system might play a role for their dependability.