An Internet-of-Things Based on ILPP, Petri Nets, and Artificial Neural Networks for Controlling Tool Failures in Flexible Manufacturing Systems Under Complex Operational Conditions

Petri nets are utilized for modeling and evaluating tool failures in flexible manufacturing systems (FMSs) characterized by “concurrency,” “conflicts,” “resource sharing,” and “sequential operations.” The effectiveness of the development controller for complicated FMSs depends on the quality of tool maintenance techniques that can minimize costs, increase productivity, reduce unscheduled downtime, and minimize disturbances to other FMS processes. This paper develops a new online controller that combines “colored Petri nets” (CPNs), the “internet of things” (IoT), and “artificial neural networks” (ANNs) to prevent early tool failures in FMSs. First, a structurally minimal approach-based PN, “vector covering approach” (VCA), and “place invariants” (PIs) are employed to generate the “integer linear programming problem” (ILPP); solving this ILPP enables the obtaining of an optimal set of monitors to ensure the FMS’s liveness. Second, in the first stage, the algorithm achieves a model with maximal permissiveness; nonetheless, its structure is complex. Consequently, CPNs are used to achieve minimal structural complexity by merging all control places into one controller. Next, a method that integrates the resultant network from the second step with IoT and ANN is developed for “fault detection and treatment,” ensuring the online control of tool failures in the FMS. Furthermore, a PN simulator is employed for verification, validation, and assessment of the liveness and reliability of the proposed model. Comparative analysis with existing literature indicates that the models are more effective in preventing tool failures in FMSs.