Entities with Combined Discrete-continuous Attributes in Discrete- Event-driven Systems

Whether a real system is approximated in a discrete or continuous simulation model depends on the more predominating type of variables determining the system’s behavior over time. If both discretely and continuously changing variables have to be considered, hybrid simulation techniques have to be applied, where discrete events have an influence on the otherwise continuously changing system state. But in hybrid as well as continuous models the involved entities are usually seen as some kind of evolving fluid. If the individual entities of the system and their attributes are of particular interest this approach might be inappropriate. The paper presents a different approach to hybrid simulation based on colored stochastic Petri nets. It enables the modeling of distinguishable entities with both discrete and continuous attributes in an otherwise discrete-event driven system. An example is used to demonstrate the principles and show possible application fields.