AI-Driven Optimization of Multi-Objective Task Scheduling in Fog-Assisted IoT Environments

As Internet of Things (IoT) demands increase, optimizing data storage and processing in close proximity to devices becomes crucial, leading to the prominence of fog computing as a promising architecture. Task offloading to fog nodes enables IoT devices to alleviate latency constraints, resulting in considerable reductions in energy consumption. However, designing effective task offloading criteria presents challenges, particularly in efficiently utilizing resources and selecting suitable fog nodes. Existing models prioritize parameters such as energy consumption and fog node priorities. However, they often face limitations in achieving a balanced trade-off between efficiency and energy optimization. Inadequate resource allocation or inefficient task offloading can result in underutilization or overloading of fog nodes. Therefore, optimizing task allocation to ensure balanced workload distribution across fog nodes is crucial for maximizing system performance. Moreover, minimizing communication delays across IoT and fog nodes as well as across the fog and the cloud remains challenging due to varying network conditions and distances. Motivated by this, an artificial intelligence (AI)-driven solution, termed the Multi-Objective Task-scheduling in Fog-assisted IoT based on Mountain Gazelle Optimizer (MTFMGO), is developed to achieve efficient and balanced scheduling of tasks across fog-enabled IoT environments. When compared to existing approaches, the MTFMGO method considerably improves IoT application response times while also successfully balancing the workload.