Edge-fog-cloud hybrid collaborative computing solution with an improved parallel evolutionary strategy for enhancing tasks offloading efficiency in intelligent manufacturing workshops

In intelligent manufacturing workshops, the lack of an efficient collaborative mechanism among the various computational resources leads to higher latency, increased costs, and uneven computational load distribution, compromising the response efficacy of intelligent manufacturing services. To address these challenges, this paper introduces an edge-fog-cloud hybrid collaborative computing architecture (EFCHC) that enhances the interaction among multi-layer computational resources. Furthermore, the computational tasks offloading model under EFCHC is formulated to minimize objectives such as latency and cost. To refine the offloading solution, a novel multi-group parallel evolutionary strategy is proposed, which includes a two-stage pre-allocation scheme and a hyper-heuristic evolutionary operator for effective solution identification. In multi-objective benchmark testing experiments, the proposed algorithm substantially outperforms other comparative algorithms in terms of accuracy, convergence, and stability. In simulated workshop scenarios, the proposed offloading strategy reduces the total computational latency and cost by 17.81% and 21.89%, and enhances the load balancing efficiency by up to 52.50%, compared to six typical benchmark algorithms and architectures.