ELSOA: Enhanced Locust Swarm Optimization for IoT Task Scheduling in Cloud–Fog Systems

Abstract

The increasing popularity of Internet of Things (IoT) applications highlights the demand for task scheduling in the cloud–fog scenarios, where low latency, short makespan, and minimal energy use are of the utmost concern. Although prior optimization methods solved the problems, limitations remain in convergence speed and overall scheduling performance. We present an Enhanced Locust Swarm Optimization Algorithm (ELSOA) for scheduling IoT tasks across fog nodes and cloud servers. ELSOA integrates Opposition-Based Learning (OBL) and chaotic sine mapping to improve the balance between exploration and exploitation, accelerating convergence and avoiding local optima. Experimental results using both simulated and real-world datasets (GoCJ) demonstrate that ELSOA achieves an average reduction of 19.3% in makespan and 17.7% in energy consumption compared to state-of-the-art methods. These findings confirm that ELSOA offers a scalable and effective solution for dynamic IoT task scheduling in large-scale fog–cloud environments.