Timing constraints and timing correctness are not only software implementation problems but are also, to a large extent, specification and design problems. The most widely studied items-scheduling algorithms and performance-are typical implementation problems. Interest in time-constraint elicitation from the application domain is increasing. Comparatively little has been done in the area of consistency and integrity checking of the elicited time constraints. The paper suggests a way to enhance the OMT methodology with a set of methods (the Q-model) for handling the consistency, non-contradiction and integrity analysis of time constraints imposed on the system. The Q-model is well supported by related studies on the required timing properties, time modelling requirements in a real-time systems, and consistency of its own axiomatic basis. The message of the paper is that real-time systems cannot be satisfactorily handled by extending existing data processing methods. More radical changes are needed based on new paradigms and research on time modelling.