A Practical Approach for the Online Diagnosis of Industrial Transportation Systems

Model-based diagnosis enables the identification of minimal faulty component sets that explain encountered inconsistencies between a system's observed behavior and that of a “golden” system model. In this paper, we present a compositional model for the (online-)diagnosis of transient faults (like malfunctioning transportation segments, sensor errors, misrouting, …) in industrial transportation systems. Instead of analyzing flow parameters or modeling temporal behavior via finite state machines, we consider the temporal and logic constraints about a distributed item's progress separately, and at different abstraction levels. Initial results from an industrial facility show the applicability of our consistency-based diagnosis approach.