The rapid expansion of the Internet of Things (IoT) has led to a significant increase in data generation, necessitating efficient computational frameworks. Cloud computing provides large-scale processing capabilities but suffers from high latency, while fog computing extends cloud resources closer to the network edge, reducing delay and enhancing efficiency. This research focuses on optimized task allocation in a cloud-fog architecture to improve execution time, cost efficiency, and energy consumption. Existing scheduling approaches often fail to consider the trade-offs between execution speed, energy efficiency, and violation costs, resulting in suboptimal performance. This study proposes the Cognitive-Guided Coati Optimization Algorithm (CGCOA), an advanced scheduling mechanism incorporating cognitive factors to enhance search exploration and convergence speed. By leveraging hunting and attacking strategies, CGCOA dynamically adjusts resource allocation based on real-time computational demands. The research methodology evaluates the effectiveness of CGCOA in different environments, including fog-only, cloud-only, and hybrid cloud-fog settings. Comparative analysis with existing algorithms such as genetic algorithms (GA), antlion optimization (ALO), and grey wolf optimization (GWO) demonstrates the superior performance of CGCOA in minimizing execution time, cost, and energy consumption. The findings highlight the potential of intelligent scheduling frameworks in optimizing cloud-fog resource utilization while maintaining economic feasibility. This research contributes to the development of adaptive scheduling techniques, en-suring enhanced performance in modern IoT-driven computing environments. Future work may focus on refining the algorithm for dynamic workload variations and real-time scalability in diverse applications.

Task Scheduling, Internet of Things, Cognitive-guided Coati Optimization Algorithm, Resource Allocation, Execution Time, Energy Efficiency, Cost Optimization, Metaheuristic Algorithms