Are output disaggregation and energy variables key when measuring container terminal efficiency?

ABSTRACT

This paper addresses the question whether energy consumption variables and the disaggregation of output matter in the context of efficiency analysis of container terminals. While it is obvious that the energy consumption of refrigerated cargo is higher than the energy consumption of non-refrigerated cargo, this work investigates whether those differences show in an overall efficiency analysis of terminals. This would point to a potentially important input for efficiency measures, to be considered in future productivity and efficiency analysis of terminals. Starting with a discussion on theoretical concepts and variable selection for measuring the energy dimension of terminal efficiency, this is the first paper that applies data envelopment analysis (DEA) comparing results with and without energy consumption, as well as differentiating productive outputs (dry and reefer container handling). The results reveal how the output disaggregation leads to substantially different efficiency scores and are a first step to show the relevance of output disaggregation and the inclusion of the energy variables as inputs in container terminal efficiency studies.

KEYWORDS:

• Terminal efficiency
• output disaggregation
• energy consumption
• Data envelopment analysis
• dry container
• reefer container

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. As Cullinane et al. (2004) recognized: ‘This study initially intended to investigate individual container terminals. However, data sources often reported the required data at the aggregate level of the whole port, … In these cases, the input and output of a port are defined as the aggregation of the input and output of individual terminals within the port. It is important to recognise, however, that such aggregation may prove problematic in reflecting the true production efficiency of the individual terminals within the same port.

2. The efficiency of terminals with multipurpose facilities (those handling also non-container cargoes) is out of the scope of the present paper but the interested reader could be found some example in Chang and Tovar (2014a, 2014b and 2017a, 2017b).

3. All calculated p values are greater than 0.05, meaning that the null hypothesis cannot be rejected on those bases. Given that the approximation of U by the normal distribution is best when both populations are equal or greater than 10, it is recommended to work with the tabulated value for respective sample sizes. In this respect, the statistic U is never below the tabulated value, indicating against this background the null hypothesis cannot be rejected, either. This indicates that there is no difference in the computed efficiency scores whether a terminal is a transshipment/hybrid terminal or an import/export terminal.