Multi-Objective Optimization Approaches for Resource-Constrained Iot Environments

In this paper came up with a multi-task-based optimization framework that addresses the contradiction of energy efficiency, latency reduction, reliability enhancement, and computational cost minimization in resource-constrained IoT environments. This paper proposes a three-tier device–edge–cloud architecture with an adaptive decision engine leveraging context-aware offloading, dynamic routing, and lightweight computation balancing as its framework for resource-limited IoT contexts. A group of typical IoT scenarios were investigated and these results show significant improvements in data-driven performance: Up to 39% energy savings, 46% latency reduction, and 13% better reliability against baseline strategies as demonstrated by the results obtained. We demonstrate that using multi-objective optimization is a valid and scalable technique for realizing Pareto-efficient performance in heterogeneous IoT systems with the ability to address dynamic operating scenarios in the dynamic and resource-limited environment.