ABSTRACT Wireless sensor networks constitute a foundational technology for ubiquitous monitoring and data acquisition across diverse application domains ranging from environmental surveillance to critical infrastructure management. The operational efficacy and longevity of these networks critically depend on strategic configuration of multiple design parameters including field coverage, sensors per cluster in‐charge, sensor out‐of‐range error, overlaps per cluster in‐charge, and network energy consumption. These objectives exhibit inherent trade‐offs, rendering the optimization problem a complex multi‐objective challenge characterized by conflicting criteria and high‐dimensional search spaces. This research presents a novel adaptive hybrid multi‐objective evolutionary algorithm that synergistically integrates opposition‐based learning for enhanced population diversity and initialization, Levy Flight mutation for effective escape from local optima, and adaptive operator selection for dynamic adjustment of genetic operator probabilities. We conducted exhaustive empirical evaluation comprising independent runs with individuals evolved over multiple generations, benchmarking the proposed algorithm against three state‐of‐the‐art approaches. Performance metrics were computed using global normalization with respect to theoretical problem bounds to ensure measurement validity and cross‐algorithm comparability. Statistical analysis including non‐parametric rank tests, pairwise comparisons, and effect size quantification confirm the proposed algorithm achieves statistically significant improvements with very large practical significance. The algorithm demonstrates superior convergence characteristics, solution diversity, and Pareto front quality, establishing a robust framework for automated wireless sensor network configuration in resource‐constrained environments.
Deshmukh et al. (Mon,) studied this question.