Priority-aware CO–MRFO-based task scheduling for energy-efficient mobile ad hoc clouds

Abstract

Mobile ad hoc cloud computing allows a mobile task to run on its source device, a nearby cooperative mobile node, an edge server, or a remote cloud. This paper presents a fully specified hybrid Cheetah Optimizer–Manta Ray Foraging Optimization (CO–MRFO) scheduler for priority-aware task assignment. MRFO performs global exploration, CO refines promising schedules, and a linearly decreasing individual-level probability controls their use. Each candidate is decoded into a one-hot task–node assignment and repaired when resource or connectivity constraints are violated. The minimization objective combines normalized response time, node-dependent energy, priority-weighted deadline violations, and feasibility penalties. Over 30 paired runs with 100 tasks, CO–MRFO obtained an objective value of 12.84 ± 0.31, runtime of 8.74 s, total energy of 412.6 J, deadline-miss ratio of 0.42%, and throughput of 114.8 tasks/s. Relative to GWO, it reduced the objective by 14.7%, runtime by 14.6%, energy by 10.8%, and deadline misses by 78.1%. Statistical, ablation, scalability, sensitivity, and service-level analyses support the contribution of the hybrid switching, priority term, and node-dependent energy model within the simulated environment.