An integrated MCDM framework for trust-aware and fair task offloading in heterogeneous multi-provider Edge-Fog-Cloud systems

The rapid growth of Internet of Things (IoT) devices, artificial intelligence (AI) applications, and real-time computing workloads has significantly intensified the need for efficient resource management in heterogeneous Edge–Fog–Cloud environments. Task offloading plays a critical role in such environments by dynamically distributing computational workloads to appropriate resources, thereby optimizing system performance and responsiveness. These environments, often operated by multiple independent service providers, face complex challenges in dynamically allocating tasks across diverse computational layers, while maintaining low latency, energy efficiency, security, and fairness. This paper proposes an integrated Multi-Criteria Decision Making (MCDM) framework for trust-aware and fairness-aware task offloading in multi-provider heterogeneous systems. The framework introduces comprehensive performance formulations that account for resource heterogeneity, Quality of Service (QoS) demands, and provider diversity, while evaluating criteria such as response time, execution cost, power consumption, energy dissipation, and security cost. To validate the proposed approach, extensive simulations are conducted, combined with rigorous evaluation methodologies including statistical analysis of results and dominance relationship assessment to ensure reliable performance comparisons across different methods. Simulation results show that the proposed MCDM-based approach achieves improvements of up to 67.6% in task response time, 79.1% in execution cost, 48.5% in effective power consumption, 41.2% in energy dissipation, and 27.6% in security cost. These improvements highlight the effectiveness of integrating fairness and trust considerations into multi-criteria task offloading decisions, leading to enhanced security, resource efficiency, and system performance. Overall, the proposed framework offers a robust, scalable, and practically applicable solution for optimizing task offloading in next-generation heterogeneous distributed systems, particularly within multi-provider environments. • Addressed the critical problem of multi-criteria task offloading in highly heterogeneous and resource-constrained Edge-Fog-Cloud environments. • Recognized the challenges posed by fairness imbalance, trust issues, and resource monopolization in multi-provider edge ecosystems. • Proposed a novel integrated MCDM framework for trust- and fairness-aware task offloading in heterogeneous multi-provider Edge-Fog-Cloud systems. • Simultaneously optimized multiple objectives including response time, energy dissipation, power consumption, execution cost, security, and fairness using a flexible decision-making model. • Incorporated various MCDM techniques and weighting methods (TOPSIS, VIKOR, PROMETHEE, Entropy, CRITIC, Hybrid) to enhance robustness and adaptability in offloading decisions. • Addressed critical gaps in prior research by introducing fairness and trust mechanisms to prevent resource monopolization and improve system security. • Conducted rigorous statistical and Pareto dominance analyses to validate the effectiveness and significance of the proposed approach under real-world data traces. • Demonstrated significant improvements in performance metrics compared to conventional methods, offering a scalable and equitable solution for next-generation IoT-driven computing environments.