Berghouwt, G., Lieshout, R.  and Veldhuis, J. (2008) Estimation of the welfare impact of airline flight schedules changes with netscan, Airneth research memorandum 005-007, Paper presented at the Annual Conference of the Air Transport Research Society (ATRS), Berkeley, USA, 22-24 June 2007


The performance of airline networks may be assessed using distinct types of indicators. Most benchmark analyses between airports or airlines use relative simple, easy to communicate indicators, such as number of destinations and frequencies of direct connections to particular destinations. But also indirect connections contribute significantly to the accessibility by air between airports and regions.

Moreover, the large hubs and their airlines focus particularly to indirect connections, as these are indispensable to retain particular market shares. The performance and quality of networks, including indirect connections, are less easy to communicate, but models have been developed to monitor this. The various models have their own advantages and disadvantages. The advantage of NetScan is its accuracy. It takes into account the departure and arrival times of each individual connection.
In case of indirect connections, also the connecting time at specific hubs can be assessed. This enables the user to determine elapsed (and perceived) travel time of each individual connection. This detailed analysis, including the account of departure and arrival times, makes the NetScan model however inappropriate for use in forecasting studies. Determining departure and arrival times of individual connections in distinct future scenarios is not only a practically impossible task, but may also be undesirable. This disadvantage is overcome by the NetCost model. This model functions at a slightly higher aggregation level and ignores departure and arrival times. It is less accurate for that reason, but it can be used in forecasting, if network scenarios at higher aggregation level are specified4 5. This makes the NetCost model suitable for using it in strategic evaluations of distinct network policies, not only for today, but also for the future.

This paper is an attempt to combine the advantages of the two models: it is an analysis of current networks, taking into account departure and arrival moments of individual connections, assessing the actual connecting times at hubs (if any), determining its connectivity value (as is done in NetScan) ánd determining the generalised travel costs of that particular connection (as is done in NetCost).