Hybrid Nature-Inspired Metaheuristic Scheduling For Dynamic Cloud And Flexible Job Shop Environments: A Unified Optimization Framework

Abstract

The rapid convergence of cloud computing infrastructures and advanced manufacturing paradigms has intensified the need for intelligent scheduling mechanisms capable of addressing multi-objective, dynamic, and large-scale optimization challenges. Traditional deterministic scheduling approaches are increasingly inadequate in environments characterized by resource heterogeneity, uncertainty, and fluctuating workloads. Inspired by foundational scheduling theory and contemporary developments in nature-inspired metaheuristics, this study proposes a unified hybrid optimization framework that integrates Grey Wolf Optimization (GWO), Whale Optimization Algorithm (WOA), Particle Swarm Optimization (PSO), and emerging biologically motivated algorithms such as Dwarf Mongoose Optimization (DMO), Butterfly Optimization Algorithm (BOA), Cheetah Optimizer (CO), and Lungs Performance-based Optimization (LPO). Drawing on established scheduling theory (Leung, 2004; Morton et al., 1993), flexible job shop optimization advances (Bissoli et al., 2018; Gong et al., 2019; Zarrouk et al., 2019), and cloud scheduling surveys (Tsai et al., 2013; Garg et al., 2018; Kaur et al., 2016), the framework addresses multi-objective trade-offs including makespan minimization, load balancing, energy efficiency, and Quality of Service (QoS) assurance.

The proposed methodology introduces a hierarchical two-layer optimization model in which exploration-oriented swarm dynamics are combined with exploitation-focused adaptive learning strategies. Extensive simulation analysis under dynamic cloud conditions demonstrates significant improvements in resource utilization stability, scheduling robustness, and convergence reliability compared to standalone metaheuristics. The results highlight that hybridization enhances global search diversity while preserving convergence precision in highly nonlinear solution spaces. The study contributes both theoretically-by bridging manufacturing and cloud scheduling paradigms-and practically-by providing a scalable optimization architecture suitable for Industry 4.0 and distributed cloud ecosystems.