Recent research in Resource Utilization-based Scheduling and Allocation in Cloud Computing focuses on optimally matching tasks and resources so that physical and virtual machines are used more fully, idle resources are minimised, and overall system throughput is maximised. Studies employ task-scheduling strategies that account for resource utilisation metrics (CPU, memory, bandwidth) when allocating workloads, and often integrate predictive models or heuristics to dynamically adapt allocations as demand fluctuates. By designing schedulers that prioritise high-utilisation mappings and reclaim or reassign under-utilised resources, these approaches improve efficiency, reduce waste, and support better performance in heterogeneous, dynamic cloud environments.