Insight from scientific study in logistics using text mining

Abstract

Big text data show trends from past logistics research and define freight flow and socio-economic relationships in the global logistics network. This relationship plays an important role in predicting future logistics trends and determining the direction of research. The purpose of this study was to collect logistics and freight related papers published in Transportation Research Record: Journal of the Transportation Research Board, since 1996 and to derive the main topics of the logistics studies that have been performed via topic modeling, using the Latent Dirichlet Allocation (LDA) approach. From the results, 20 main topics with keywords and phrases were extracted from the logistics research papers, which suggests that topics such as trip generation model, urban freight, and logistics hub have been emerging for scholars in the fields of road, air, and shipping logistics and have been examined for some time. In addition, big data, the Internet of Things (IoT), and information and communications technology have recently been applied to the logistics field. Research on data collection technology and route optimization algorithms that incorporate the technologies have, therefore, attracted a great deal of interest from current researchers. Through the framework of this study, it is expected that future trends in the field of logistics will be predicted, and that appropriate planning and strategies can be established.