—Programmers spend a large portion of their time debugging the code they write. This is supported by a variety of debugging techniques such as pause/resume, the setting of breakpoints, stepping over functions, etc. Today, modelling and simulation become increasingly important enablers in the development of complex, reactive, often real-time, software-intensive systems, as they allow rapid prototyping and early validation of designs. Simulation models, though at a higher level of ion than code, can however still contain bugs. There is hence a need for model-level debuggers, that are adapted to the semantics of the modelling formalism(s) used, and can properly deal with the timed nature of many of these models. This paper presents a method for constructing model debugging environments for deterministic, operational formalisms. In order to manage the inherent complexity, the timed, reactive behaviour of the debugger is modelled explicitly. The feasibility of the approach is demonstrated...