A cloud-supported cps approach to control decision of process manufacturing: 3D ONoC

Abstract

The Cyber-Physical System (CPS) concept is now attracting attention in systems engineering, and it is being applied to a fully automated factory control in processes such as semiconductor fabrication. In this paper, we propose a novel control decision structure for process manufacturing, designated as the 3D Optical Network-on-Chip (ONoC) multi-core system, based on the cloud-supported CPS concept. We first construct a task graph — which includes interconnected Virtual Machines (VMs)—to represent the interaction between industrial-physical processes and cyber states. Given the task graph, the control decision process becomes into the problem of the on-chip VM placement. We then design a highly reliable on-chip VM placement scheduling to find the optimal control strategy while guaranteeing the reliability of the 3D ONoC structure. The simulation results demonstrate that our scheme achieves a higher reliability of the 3D ONoC structure when we make the control decision for process manufacturing.