Performance Evaluation of Metaheuristic Algorithms for IIoT Device Deployment in Obstacle-Constrained Environments

Abstract

The Industrial Internet of Things (IIoT) has transformed industrial systems through the integration of sensing technologies and automation. Despite these advantages, inadequate device placement leads to coverage holes that degrade system performance in complex environments. Optimizing IIoT device deployment is therefore essential to ensure reliable monitoring and communication. Due to the NP-hard nature of the problem, metaheuristic algorithms have been adopted for exploring complex search spaces. This paper presents a simulation-based performance evaluation of seven metaheuristic algorithms, including Antlion Optimization (ALO), Artificial Bee Colony (ABC), Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Water Cycle Algorithm (WCA), Bat Algorithm (BA), and PSO with Learning Strategy and Crossover (PSOLC) for IIoT device deployment in obstacle-constrained industrial environments. The experimental environment is designed to simulate the characteristics of a factory layout. All experiments are conducted on a 100 × 100 m area under two obstacle configurations corresponding to 14% and 44% obstacle density to ensure statistical robustness. Coverage is defined as the percentage of the deployable area covered by the sensing range of at least one device, excluding obstacle regions, and is computed using a grid-based spatial coverage model. Algorithm performance is evaluated in terms of average coverage and the number of function evaluations required for convergence. The results indicate that ALO achieves the highest coverage (91.13%) with rapid convergence, outperforming the others, whereas BA exhibits the weakest performance. The study also outlines future research directions for improving metaheuristic-based IIoT deployment strategies in industrial environments.