A linear programming approach for capacity estimation and robustness analysis of the European air traffic network

Abstract

The European Air Traffic Network (ATN) consists of airports and Area Control Centres (ACCs), and is a highly complex system. Given that the Air Traffic Management (ATM) in Europe is shifting from a local level to a network-wide level through the implementation of a new trajectory-based Concept of Operations (ConOp), a network capacity estimation method is required to indicate and monitor the performance of the ATN. The current indicator of network capacity, Air Traffic Flow Management (ATFM) delay, is not a direct measure of capacity and is insufficient for planning and management purposes. Existing literature on ATM shows that current capacity estimation methods tend to focus on capacity issues at local levels. Although some research has been undertaken on capacity estimation for large scale ATNs, these methods are neither transferrable nor flexible. In order to fill this gap, this thesis proposes an analytical approach based on Linear Programming (LP) to estimate the capacity of the European ATN. In addition to the network capacity, the factors that influence capacity are identified and quantified by applying statistical methods to the historical data regarding ATFM delays. Based on empirical data on flights and capacities of ACCs and airports in Europe, the network capacity is calculated as the theoretical uppers in terms of traffic volume subject to the constraints of connectivity, demand, capacity, and flight routes. In particular, a static modelling approach is employed to find the maximum flow in the ATN, and a dynamic approach is used to minimize travel time. The air traffic and ATFM delays collected between 15th and 20th July 2014 are used to test and validate the network capacity estimation model. The results show that, in comparison with the ATFM delays, the proposed model is more capable to capture the network-wide impact of the incident of Malaysian Airlines flight MH17. The results also suggest that the negative impacts of the MH17 event on total flying times could be alleviated by assigning the traffic demands to other alternative paths and available slots. Based on the capacity estimation capabilities of the proposed models, this thesis conducts robustness analysis of the European ATN through an investigation of its topological and operational characteristics. By applying the capacity estimation model to a number of scenarios involving local capacity reduction, this research proposes a new robustness index called the Relative Area Index (RAI). The RAI quantifies the importance of an individual node to the performance of the entire network in the presence of local capacity reduction. Air traffic data from three typically busy days in Europe are utilised to shown the advantage of the RAI over Betweenness Centrality (BC), which is a conventional robustness index, in capturing the network-wide impact of local degradation. This index can be used to assess network robustness and provides a valuable tool for airspace managers and planners.