A centralised DEA approach to resource reallocation in Spanish airports

Abstract

A centralised data envelopment analysis (DEA) approach optimises the resource usage for all the different units in an organization rather than for each unit separately (conventional DEA). This is particularly relevant for the Spanish airports controlled centrally by the Spanish airport authority rather than by individual airport managers. In this study, a non-oriented Slack-based inefficiency (SBI) DEA model is used in order to reallocate two transferrable inputs (namely, labour costs and operating costs) between the different airports. Firstly, we apply a conventional (i.e. non-centralised) non-oriented SBI model to identify the inefficient airports. Then, we apply the corresponding centralised DEA model to the inefficient units to maximise the potential improvements (slacks) obtained by reducing the total consumption of the inputs (allowing resource reallocation) and increasing total outputs. The results show how it is feasible to increase the total amount of passengers and cargo as well as the number of aircraft movements without increasing the total amount of inputs, just by reallocating them in an efficient way. Several progressively more relaxed scenarios have been considered, leading to larger total potential efficiency improvements. The model identifies airports that suffer from over-capacity and provide individual targets in each input and output to become more efficient. Sensitivity analysis to outliers or mavericks (i.e. airports with a strong specialisation) has been carried out. The size efficiency of individual airports as well as the overall company has also been studied. The potential efficiency gains and the optimal number of airports in a radical system restructuring have been computed and the optimal size operating point has been determined and compared with the actual inputs and outputs of the existing airports.