The paper addresses the fault detection problem for large discrete event systems that can be modelled by a set of interacting Petri Net (PN) modules. Each system module is monitored by a PN diagnoser that has local information on the module structure and shares information by some places that are coupled with the other modules of the system. Each diagnoser works on-line: it waits for the firing of an observable transition and employs an algorithm based on the definition of some integer linear programming problems to decide whether the system behaviour is normal or exhibits some possible faults. A case study presented in the related literature shows the distributed fault detection strategy and points out the efficiency of the proposed approach.
