Trust-Based Petri Net Model for Fault Detection and Treatment in Automated Manufacturing Systems

Automated manufacturing systems (AMSs) are vulnerable to failures. Trust evaluation is becoming a novel technique for detecting faults in AMSs. In this article, a two-step robust deadlock control strategy for systems with unreliable resources is proposed. Recovery subnets are added for all failure-prone resources of the system model in the first step based on Petri nets. The recovery subnets are applied to a system to ensure the system works in a reliable model. In the second step, a novel hybrid technique combining a trust-model and colored Petri nets is proposed to the detection and handling of faults obtained in the first step. The proposed approach includes the descriptive features of a modular Petri net integration with the trust method. It provides the combination of three kinds of procedures: management of all resource failures, detection, and treatment of faults in a system. Consequently, the proposed model considers not only resolving resource failures in AMSs, but also their treatment once they are detected. Finally, an example from the literature is used to test the proposed approach.