Evolutionary Meta-Heuristic Offloading and Scheduling Schemes Enabled Industrial Cyber-Physical System

The industrial cyber-physical system (ICPS) is a paradigm that connects conventional machines to cutting-edge technologies through the Internet of Things (IoT). The IoT encompasses applications, such as smart healthcare, intelligent transport, and smart homes that need to be connected to various cloud servers. These IoT applications offload their data workloads to the servers for processing and storage. However, existing ICPS paradigms for IoT applications must be rectified due to security, failure, execution, and communication delays. This study proposes an optimal, delay-efficient ICPS paradigm for IoT applications. The research considers task offloading and scheduling problems and presents the evolutionary meta-heuristics offloading scheduling (EMOS) algorithm framework. The EMOS algorithm framework comprises several delay-optimal schemes to address the offloading and scheduling challenges of IoT applications. The primary objective is to execute IoT applications on different nodes while adhering to the specified constraints. Simulation results demonstrate that the proposed meta-heuristic reduces local processing delays by 39%, offloading delays by 41%, execution and propagation delays by 40%, failure delays by 43%, and security and migration delays by 50% compared to existing ICPS meta-heuristics.