By reviewing literature related to logistics demand forecasting retrieved from two academic databases of Scopus and CNKI and using the literature visualization software CiteSpace, this paper analyzes the cooccurrence, emergence, and time zone maps of key words in these articles. The results show that in terms of research hotspots, Chinese researchers mostly focus on the meso-micro level, from classical statistical models and optimization algorithms to neural network prediction models and their improvements for predicting the logistics in specific industries, while global researchers pay more attention to the application of intelligent network models and other methods in macro-regional logistics. In terms of research frontiers, in the past three years, there has been a lot of research on the combination of intelligent optimization algorithms and multiple regression models in China, while global researchers pay more attention to the application of artificial intelligence methods and their combinations under the development of big data, and pay more attention to the optimization of the resource allocation of enterprise inventory and spare parts in the logistics supply chain. Finally, this paper puts forward some possible research lines.