Mining the Sources of Delay for Dray Trucks at Container Terminals

Abstract

To isolate the causes of abnormally high truck turn time, this paper develops a methodology for examining the sources of delay for dray trucks at container terminals. It is motivated by the need of port authorities and terminal operators to develop specialized solutions to reduce turn time based on terminal-specific causes. Although many ports have taken steps to improve the general level of service for trucks, such as establishing chassis pools and extending gate hours, fewer have performed the transaction-level analysis required to determine why a certain subset of operations is significantly higher than the average, thereby hindering the overall level of service. After problematic steps in the truck transaction process are isolated, terminals can select and deploy a range of technological or organizational countermeasures to address the problem. This study draws on a database of truck activity from the port of Houston, Texas. Because of the large number of gate transactions and potential factors that could contribute to high truck turn time, a data mining technique is used. Specifically, a decision tree technique is explored and described in this paper. Results indicate that import transactions that require chassis tend to have high truck turn time because truckers need to find a matching chassis. This paper demonstrates how decision trees can be used by port authorities and terminal operators to gain insight into their operations without the need to perform exhaustive data analysis.