Abstract: During the Covid-19 pandemic, the e-commerce service industry has played a significant role in ensuring the development of e-commerce and promoting the innovation of new business models. Grasping the trend of the future development of the e-commerce service industry leads to pushing the government to formulate industrial development strategies and to ensure the rapid recovery of the postepidemic economy. This paper proposes a Multi-Model Grey Fusion (MGF) procedure to solve the reliability and robustness problems of predicting the future trend of the e-commerce service industry. The experimental results show that MGF performs well in handling the small data prediction problem and that it is more reliable and robust than the four single-base models, which are the grey model, linear regression, back-propagation neural network, and support vector regression. The forecast by combining the MGF with the rolling framework shows that China’s e-commerce service industry will keep an excellent development momentum. Moreover, the market size is expected to reach RMB 8.78 trillion by 2025, with an average annual growth rate of 8.8%.

Key words: Grey Theory, Fusion Procedure, Forecasting, E-commerce Service Industry