![Sample our Engineering & Technology journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5933/TOC-Subject-Sample-2021-Engineering-Tech.jpg"})
Abstract
This paper aims to provide new measures of airline network configuration with a view to analyse effectively the complexity of modern carriers’ network design. It studies network configurations in the airline sector by taking into account both spatial and temporal dimensions. The spatial dimension is measured by using both the Gini index and the Freeman index, which originate from social science research. The temporal dimension is measured by the connectivity ratio, i.e. the share of indirect connections over the total number of connections. According to these indicators, the configuration of the largest full‐service carriers and the largest low‐cost carriers in Europe is investigated. The results show that the temporal dimension provides a clear distinction between full‐service carriers and low‐cost carriers; while the spatial dimension appears useful when identifying the peculiarities within groups.
Acknowledgements
This paper has benefited from suggestions by Aura Reggiani, Lanfranco Senn and two anonymous referees. The authors would like to thank KLM Royal Dutch Airlines. The authors’ opinions do not necessarily reflect the official KLM viewpoints.
Notes
1. This reorganization took place between 1978 and 1985, according to Reynolds‐Feighan (Citation2001). Many authors (e.g. Borenstein, Citation1989; Berry, Citation1994; Oum et al., Citation1995; Button et al., Citation2000; Burghouwt et al., Citation2003) put much effort into explaining the reasons for the change and the advantages of carriers. Above all, it was emphasized that both trunk and regional carriers adopted the HS structure to exploit the dominant position of the hub and the cost advantages of a centralized network, such as economies of density and scale.
2. The authors thank an anonymous referee for suggesting this clarification.
3. For a theoretical and empirical investigation of hub connectivity, see Bootsma (Citation1997), Button et al. (Citation1998), Dennis (Citation1998), Rietveld and Brons (Citation2001), Veldhuis and Kroes (Citation2002), and Burghouwt and De Wit (Citation2003).
4. Intra‐European flights are considered to be within the EU 25 Member States plus Romania, Bulgaria, Switzerland, Turkey and the Canaries Islands.
5. Comparing the LCC networks (which are focused purely on the intra‐European market) with the European segments of the FSC might seem somewhat problematic. The FSCs jointly optimize the intra‐European market and the intercontinental market, and overall the intra‐European segments are also functioning as feeders to intercontinental flights. However, the intra‐European market still represents a considerable part of their revenue and deserves a dedicated strategy (integrated with the intercontinental market strategy). On the other hand, a comparison of the whole flag carriers network with that of the LCC makes even less sense. Note that the present aim is also to detect if the FSCs are changing their network strategy as a reaction to LCCs in a liberalized market.
6. Ryanair data are present in the OAG database only until 2000; therefore, they the data for 2004 are missing in the present analysis. Virgin Express is not the fourth largest LCC but is seventh in terms of network seats supplied. It is included in the analysis as representing a known case of a different LCC model philosophy. This will also emerge in the results.
7. A traveller who needs to fly from one point (origin) to another (destination) first prefers direct flights, then connected flights with one stop, etc. If an airport is on the shortest path between an origin and a destination, this means the traveller will probably choose to pass through that airport.
8. The Freeman index was calculated with UCINET 6 for Windows (Borgatti et al., Citation2002).
9. The original version of the Bonacich (Citation1972) index was produced using the adjacency matrix. The present paper uses an extension of the Bonacich index using the value matrix (Bonacich, Citation1987). The adjacency matrix that represents the network contains a ‘1’ when there is a link between the nodes; and zero otherwise. In the value matrix the ‘1’ is replaced by a measure of the strength of the link. In the present case the value matrix is constructed with the number of flights per week between the airports.
10. A complete overview of factor analysis output and the first eigenvector values are available upon request to the authors.
11. It is remarkable that the most decreasing airport in terms of centrality is Berlin, where the LCC Air Berlin developed one of the biggest bases.
12. The analysis refers to intra‐European flights and data cover 1996–2004 for the summer season schedule (a representative week in August).
13. The present paper does not aim to identify the characteristics of the wave‐system structure as many studies have done. It aims to find a simple temporal concentration measure to make comparisons between carriers during the time periods.
14. The analysis can be performed by using the Gini index. As the results are similar, the analysis is omitted.
15. Primary airport operations can be considered as an additional service as they often reduce passenger travel costs from the city to the airport.
