Timing analysis of timed event graphs with bounded delays using algebraic techniques

Timed event graphs constitute an important class of discrete event systems that have a wide domain of applicability. Analysing the temporal behavior of these systems has proven to be efficient, primarily through the use of algebraic techniques. In this paper, we present a major extension to this work, in that systems with timing properties that are specified using delay ranges, instead of fixed delays, are considered. We analyse the nonstochastic behavior of timed event graphs, and present an efficient algorithm to find exact (tight) upper and lower bounds on the separation in time of an arbitrary pair of system events. Stochastic analysis may be more suitable for studying efficiency and utilization, but non-stochastic techniques and tight upper and lower bounds on separation times, are useful for verifying correct operation