Elsevier

Transport Policy

Volume 42, August 2015, Pages 75-85
Transport Policy

Evaluating the efficiency performance of airports using an integrated AHP/DEA-AR technique

https://doi.org/10.1016/j.tranpol.2015.04.008Get rights and content

Abstract

Airport efficiency is an area of increasing interest to academics, policy makers and practitioners. This has resulted in a body of literature applying various econometric techniques to compare efficiency between different samples of airports. This paper uses the multi-criteria decision making method Analytic Hierarchy Process (AHP) to incorporate the weightings of input and output variables into Data Envelopment Analysis (DEA) and Assurance Region DEA (DEA-AR) models, with 24 major international airports in the empirical analysis. The paper concludes the discriminatory power in the proposed AHP/DEA-AR model is greater than in the basic DEA model when measuring the efficiency of airports. By applying this approach, policy makers and practitioners can effectively compare operational efficiency between airports, and therefore generate more informed decisions.

Introduction

Airport efficiency evaluation has been a burgeoning area of research in recent years. These analyses are important for a variety of stakeholders, including airports, regulatory bodies, governments, passengers and airlines (Humphreys and Francis, 2002). Motivations for examining airport efficiency include assessing financial and operational efficiency, evaluating alternative investment strategies and monitoring airport activity (Doganis, 1992).

Lai et al. (2012) pointed out that after the year 2000, more than 50 papers related to airport efficiency have been published, but before this, only four papers were published. From a methodological perspective, one of the dominant approaches taken has been the application of econometric tools, featuring in 80% of all published papers in this area (Lai et al., 2012). In terms of the specific techniques adopted, Data Envelopment Analysis (DEA) featured in one of the first papers published (Gillen and Lall, 1997), and has become a popular tool since then, being employed in around 50% of these papers. In doing so, developments to improve accuracy with the employment of DEA, such as bootstrapping, have been incorporated (Curi et al., 2011).

In a DEA model, the preference weights of input and output variables are automatically calculated, but the importance of these variables relative to each other is not included in the calculation (Coelli et al., 2005). Therefore, it is considered that each variable has an equal level of importance. In reality, preference will be given towards certain variables and these preferences may change depending upon the considered stakeholders. For example, Humphreys and Francis (2002) discuss how airport managers, airport owners/shareholders, governments, airlines and passengers have varying motivations for performance measurement, and therefore use different measures.

To resolve this problem, applying a multi-criteria decision making (MCDM) method to derive the weight of importance of each variable, before undertaking DEA analysis, is a way of overcoming this issue. A popular method of MCDM is the Analytic Hierarchy Process (AHP). AHP develops from a linear additive model, respectively, on pairwise comparisons between criteria and between variables (Saaty, 1980). In the context of airport management, AHP has seen use in the context of evaluating the risk factors for a logistics hub development (Tsai and Su, 2002), evaluating the competitiveness of Asia-Pacific air cargo hubs (Chao and Yu, 2013) and choosing a simulation software package to support airport operations (Otamendi et al., 2008).

In this paper, AHP and DEA are combined to evaluate airport efficiency, an approach which has not previously been undertaken within published research in this area. Through this, the effectiveness of the DEA Assurance Region (DEA-AR) model as a means for increasing discriminatory power is highlighted, as an alternative to other approaches, such as changing the ratio of decision making units to variables (Charnes et al., 1985) or referral clustering (Zhu, 2009). A secondary aim of the paper is to show the value of combining AHP and DEA in reflecting the opinions of different stakeholder groups. While this combined approach has been adopted elsewhere, it has not been applied to the air transport sector. As identified earlier, the stakeholder groups may have different views on the importance of particular variables, which then influence their perception of efficiency (in airport terms).

The paper proceeds as follows. Section 2 examines the airport efficiency evaluation literature, while Section 3 reviews AHP/DEA models. In Section 4, elucidation of the AHP and DEA model employed in this analysis is given. The next section describes the collection of the data required in our analysis. In Section 6, the AHP specific results are exposited, followed in Section 7 with the DEA results (post use of AHP findings) being presented in terms of the two efficiency models considered. In Section 8, discussion and conclusions are given, including thoughts towards future research.

Section snippets

Airport efficiency evaluation

In the mid-1990s, the literature on efficiency evaluation, which had already been applied to numerous industries, was introduced to the airport sector (Gillen and Lall, 1997). Since then, a number of papers have been published on airport efficiency, although the depth of coverage is perhaps less than in other transport industries such as seaports (Woo et al., 2011).

One approach adopted is the use of partial measures, which calculate the ratios of one input to one output to assess efficiency in

Integrated AHP/DEA models

Within the literature, Ho (2008) undertook an extensive review of integrated AHP and its applications, and has reported only a limited number of publications with combined AHP and DEA models. Several studies have indicated that AHP can be applied to form an AHP/DEA ranking model for the purpose of improving DEA usability (Feng et al., 2004, Friedman and Sinuany-Stern, 1998, Lee and Tseng, 2006, Sinuany-Stern et al., 2000). The advantage of the AHP/DEA ranking model is that the comparative

Model elucidation

In this section, the model used within this paper is elucidated, starting with AHP before developing the DEA-AR and integrated models.

Data collection

The data collection process for this research followed two main stages. Firstly, variables for the model were selected, along with the sample of airports. Second, in order to enable the AHP analysis, a questionnaire survey was carried out. Finally, DEA models were applied to airport data for 2010 (Air Transport Research Society, 2011)

AHP results

Table 5 gives an overview of the AHP results generated from the survey, for inputs and outputs and Levels 2 and 3 (see Fig. 1). Considering the inputs first, at Level 2, the experts considered financial inputs to be more important than capacity inputs. At Level 3, there were some variations between the capacity variables, with number of gates, size of terminal area and number of runways being considered the most important. These variables particularly reflect the number of aircraft an airport

AHP/DEA-AR results

The results of the two efficiency models are summarized in Table 8. With the DEA-BCC model, 19 of the 24 airports are considered relatively efficient but with AHP/DEA-AR this reduces to just 5 airports. This is reflected in the average efficiency scores for the sample as a whole, which is 0.9719 for DEA-BCC and 0.7231 for AHP/DEA-AR. Consequently, there is greater discriminatory power within the latter model, as evidenced further by the standard deviation of the results increasing from 0.0786

Discussion and conclusion

The main aim of this paper was to evaluate the impact of integrating AHP into DEA analysis in the context of airport efficiency. As noted earlier, while this has been undertaken in other sectors, it has yet to transfer into the transportation industry. Such a move is important as it enables the efficiency analysis to more accurately reflect the perceptions of stakeholders in relation to the selected variables. The study shows that, through using an AHP/DEA-AR model, that the subjectivity does

References (80)

  • C. Curi et al.

    New evidence on the efficiency of Italian airports: a bootstrapped DEA analysis

    Socio-Econ. Plan. Sci.

    (2011)
  • T. Ertay et al.

    Integrating data envelopment analysis and analytic hierarchy for the facility layout design in manufacturing systems

    Inf. Sci.

    (2006)
  • M.T. Escobar et al.

    A note on AHP group consistency for the row geometric mean priorization procedure

    Eur. J. Oper. Res.

    (2004)
  • E.H. Forman

    Random indices for incomplete pairwise comparison matrices

    Eur. J. Oper. Res.

    (1990)
  • G. Francis et al.

    The benchmarking of airport performance

    J. Air Transp. Manag.

    (2002)
  • L. Friedman et al.

    Combining ranking scales and selecting variables in the DEA context: the case of industrial branches

    Comput. Oper. Res.

    (1998)
  • D. Gillen

    The evolution of airport ownership and governance

    J. Air Transp. Manag.

    (2011)
  • D. Gillen et al.

    Developing measures of airport productivity and performance: an application of Data Envelopment Analysis

    Transp. Res. Part E: Logist. Transp. Rev.

    (1997)
  • S.X.H. Gong et al.

    The impact of airport and seaport privatization on efficiency and performance: a review of the international evidence and implications for developing countries

    Transp. Policy

    (2012)
  • A. Graham

    The objectives and outcomes of airport privatisation

    Res. Transp. Bus. Manag.

    (2011)
  • W. Ho

    Integrated analytical hierarchy process and its applications: a literature review

    Eur. J. Oper. Res.

    (2008)
  • P.G. Hooper et al.

    Measuring total factor productivity of airports: an index number approach

    Transp. Res. Part E: Logist. Transp. Rev.

    (1997)
  • I. Humphreys et al.

    Performance measurement: A review of airports

    Int. J. Transp. Manag.

    (2002)
  • S. Ison et al.

    UK regional airport commercialisation and privatisation: 25 years on

    J. Transp. Geogr.

    (2011)
  • W. Kong et al.

    Assessing the performance of business colleges in Taiwan using data envelopment analysis and student based value-added performance indictors

    Omega

    (2012)
  • J. Korpela et al.

    Warehouse operator selection by combining AHP and DEA methodologies

    Int. J. Prod. Econ.

    (2007)
  • M. Kwiesielewicz et al.

    Inconsistent and Contradictory Judgements in Pairwise Comparison Method in AHP

    Comput. Oper. Res.

    (2004)
  • L.C. Lin et al.

    Operational performance evaluation of international major airports: an application of data envelopment analysis

    J. Air Transp. Manag.

    (2006)
  • J. Otamendi et al.

    Selection of the simulation software for the management of the operations at an international airport

    Simul. Model. Pract. Theory

    (2008)
  • T.H. Oum et al.

    Privatization, corporatization, ownership forms and their effect on the performance of the world's major airports

    J. Air Transp. Manag.

    (2006)
  • T.H. Oum et al.

    Ownership forms matter for airport efficiency: a stochastic frontier investigation of worldwide airports

    J. Urban Econ.

    (2008)
  • T.H. Oum et al.

    Measuring airports' operating efficiency: a summary of the 2003 ATRS Global Airport Benchmarking Report

    Transp. Res. Part E: Logist. Transp. Rev.

    (2004)
  • E. Pels et al.

    Relative efficiency of European airports

    Transp. Policy

    (2001)
  • R.F. Saen et al.

    Determining relative efficiency of slightly non-homogeneous decision making units by data envelopment analysis: a case study in IROST

    Appl. Math. Comput.

    (2005)
  • J. Sarkis et al.

    Performance based clustering for benchmarking of US airports

    Transp. Res. Part A: Policy Pract.

    (2004)
  • L.M. Seifert et al.

    Identifying excesses and deficits in Chinese industrial productivity (1953–1990): a weighted data envelopment analysis approach

    Omega

    (1998)
  • Z. Sinuany-Stern et al.

    An AHP/DEA methodology for ranking decision making units

    Int. Trans. Oper. Res.

    (2000)
  • Y. Takamura et al.

    Comparative site evaluation study for relocating Japanese government agencies out of Tokyo

    Socio-Econ. Plan. Sci.

    (2003)
  • B. Tovar et al.

    Technical efficiency and productivity changes in Spanish airports: A parametric distance function approach

    Transp. Res. Part E: Logist. Transp. Rev.

    (2010)
  • M.C. Tsai et al.

    Political risk assessment on air logistics hub developments in Taiwan

    J. Air Transp. Manag.

    (2002)
  • Cited by (93)

    • Measuring Spanish airport performance: A bootstrap data envelopment analysis of efficiency

      2023, Utilities Policy
      Citation Excerpt :

      Along with the financial variables, most works employ physical or infrastructure-related variables such as the number of gates as inputs (Gillen and Lall, 2001; Gutiérrez and Lozano, 2016; Keskin and Köksal, 2019). The length of the runway is used by Wanke (2012), Lai et al. (2015) and Yang et al. (2019); the runway area by Ahn and Min (2014), Ülkü (2015) and Keskin and Köksal (2019); and the number of runways by Lin and Hong (2006) and Stichhauerova and Pelloneova (2019). Other variables include the number of baggage belts (Gillen and Lall, 2001), terminal area (Yuen and Zhang, 2009; Yang et al., 2019), and apron stands or apron area (Yu, 2004; Gutiérrez and Lozano, 2016).

    • Technical efficiency and productivity change analysis: A case study of the regional and local airports in Thailand

      2022, Case Studies on Transport Policy
      Citation Excerpt :

      Assaf (2010) revealed that the scale efficiency of UK airports operated under decreasing and increasing returns to scale. According to our review, several empirical studies affirm that the airport industry should operate under variable returns to scale assumption (Chen et al., 2017; Lai et al., 2015; Liu et al., 2021; Olariaga & Moreno, 2019; Örkcü et al., 2016; Ripoll-Zarraga & Mar-Molinero, 2020; Tsui et al., 2014; Wanke et al., 2016). Computing the TES requires a specific set of input and output variables, which is part of either a financial or physical dataset.

    View all citing articles on Scopus
    View full text