To the best of our knowledge, although there is a substantial literature focusing on factors affecting freight mode and route choice, limited scientific interest is presented regarding to the factors that affect the containers’ Dwell Time (DT). The innovation of this paper is to study these aspects of the logistic chains within a container terminal and to create a methodological framework able enough to predict the container dwell time and to provide yard planners and terminal operators with a forecasting tool that will assist them when making daily decisions regarding stacking policies, optimal equipment and human resources allocation.

This paper is structured as followed: After introduction the existing literature on DT prediction is presented. In section 3 the input data and the methodological framework are extensively described. Section 4, presents the data input for the model development. Section 5 discusses the model estimation results. The paper concludes by identifying the implications of the research on terminal operations when the exact DT of an import container is to be predicted and suggests more areas for further research.

2. Literature review

Before being picked-up and transported to a terminal’s mainland or being loaded onto a ship containers are stacked inside the terminal yard. Dwell Time (DT) is defined as “the total time a container spends in one or more terminal stacks”, (Ottjes et al., 2007). Container DT may be influenced by several factors such as gate operations, availability and efficiency of hinterland connections and customs regulations. Consignee, namely the receiver of the goods can be identified as one of the key stakeholders who determine DT since he decides when to pick-up import containers or when to deliver export containers. In addition, it has been found that the stacking area needed is linearly proportional to the average container time in a container terminal (Little, 1961).

Attempts have been made to estimate the influence of DT in terminal capacity. Specifically, Hoffman (1985) developed an equation that estimates the necessary size of the yard as a function of dwell time, the height of the stacked containers, the peak-hour and the total number of containers handled each year.

where:CY= Container yard area (m2)

C= Expected container volume (TEU) A= Area (m2) per container  (TEU)  T= Average container dwell time

F= Peaking factor (~20%) ensuring the yard’s efficiency

Furthermore, Dally (1983) developed a formula that estimates the number of containers a yard can accommodate. This equation uses the number of container ground slots, the mean stacking height and the container dwell time to estimate the annual yard capacity. He also applies a peak factor that usually varies from 1.2 to 1.3. It is expected that the new generation vessels will have an impact on the peak factor that has not yet being determined (Ottjes et al., 2007).

where:C= the annual yard capacity (TEU/year) Cs= Number of ground slots

H= mean profile height

W= Working slots (TEU’s) expressed as a proportion (~0.8) K’= Number of days a year

T= Mean container dwell time in the yard

F= Peaking factor (~20%) ensuring the yard’s efficiency

Hence, the average DT plays a crucial role in determining the overall terminal capacity (Chu et al., 2005). Nowadays, the increased container volumes in combination with the new massive container vessels are demanding bigger terminal capacities. One solution could be the increase of terminal size which, apart from being a very expensive investment, may be not feasible due to space limitations. Consequently, terminal operators are trying to decrease average DT. In order to do so, they have to determine the main factors that influence the number of days a container stays in the terminal. Nowadays, limited research focusing on quantifying the determinants of DT exists.

One  of  the  first  researchers  of  the  impact  of  DT  on  terminal  capacity  is  Merckx  (2005)  who    designed a framework that assists terminal operators to optimize terminal capacity, by imposing a number of pricing mechanisms based on different dwell time charging schemes. In addition Rodrigue (2008) discussed how logistic companies that use sea port terminals for their shipments and have limited distribution centres and storage areas fully utilize their free-of-charge time on the terminal’s yard. On the other hand, terminal operators react on this