The growth of e-commerce in the last decade has increased warehouse demand and promoted the development of various types of warehouses. Accor- dingly, in South Korea, warehouse development patterns in the Seoul metropolitan area (SMA) and the rest of ...
The growth of e-commerce in the last decade has increased warehouse demand and promoted the development of various types of warehouses. Accor- dingly, in South Korea, warehouse development patterns in the Seoul metropolitan area (SMA) and the rest of the country (RC) have substantially changed. On the one hand, as the concentration of e-commerce sales in the SMA has been far exceeding that of the RC, this research assumes that the warehouses in the SMA were developed to accommodate the region’s e-commerce demand. On the other, this research further assumes that the warehouses developed in the RC were accommodating the conventional B2B logistics demand. Namely, this research compares the location factors for warehouses of different regions (SMA vs. RC) to have implications for the location factors for warehouses of different uses (E-commerce vs. conventional B2B logistics).
This research first quantifies the sprawling and re-urbanizing patterns of warehouse development and tests various location factors using count data models. Four regression models are formulated by region (SMA vs. RC) and over time (sprawl vs. re-urbanization). As the count data model, this research uses negative binomial models, accounting for the over-dispersed nature of the distribution of warehouse development. Lastly, the results of the spatial and regression analyses are compared and verified to have a comprehensive understanding of the phenomena.
Results show that accessibility to industrial complexes, accessibility to logistics parks, the shortest distance to highway exits, and land price (population density) are important factors for warehouse location choice in the Seoul metropolitan area. Initially, consumer accessibility was not a statistically significant factor in the regression model. Later, a parallel development pattern is confirmed, verifying the importance of consumer accessibility: large-scale hub-fulfillment centers on the urban periphery and micro-fulfillment centers near urban areas. In the rest of the country, accessibility to industrial complexes, traffic volume, access to logistics parks, and land price (population density) are important factors for warehouse location choice. It implies that, in the RC, warehouses are located in places close to industrial complexes with low traffic congestion and away from high land prices. In the SMA, warehouse development patterns changed substantially, suggesting skyrocketing e-commerce expansion, which contrasts the insignificant changes in development patterns of those for conventional logistics demand in the RC. Lastly, analysis results suggest potential trade-offs between logistics cost and service level.