Computation Offloading Optimization in Cloud-Assisted Edge Computing Based on an Improved-IGEA Intelligent Optimization Method

Abstract

With the deep application of Internet of Things (IoT) technologies in 3C manufacturing, workshop-level intelligent production systems generate a large number of complex computing tasks. Based on the IGEA algorithm reported in [1] and the corresponding MATLAB source-code modeling logic, this paper jointly optimizes production scheduling and computation offloading. An Improved-IGEA algorithm is proposed by integrating elite preservation, elite crossover, Lévy flight, and chaotic mapping. The proposed method is compared with six algorithms, including GWO, IVYA, IGEA(Paper), PSO, SAEO, and E-GIGEA. Experimental results under seed=42 and 300 iterations show that Improved-IGEA obtains the best objective value, R = 699.23, reducing the objective value by 10.4% compared with IGEA(Paper) (R = 780.36) and by 41.5% compared with SAEO (R =1195.47). Parameter scanning and robustness experiments further demonstrate the stable superiority of the proposed algorithm under different problem scales.