Abstract

Despite its long existence and international acceptance, network theory and analysis is a practically unknown approach in Documentation, both theoretically and methodologically speaking. Fortunately, this trend is changing, inasmuch as network theory and analysis may mean a quantitative and qualitative leap forward in the representation and analysis of the structure of all types of scientific domains, whether geographic, thematic or institutional. The extraordinary advances that have taken place in recent years in the study and analysis of complex networks have been made possible by a number of parallel developments. First of all, with computerized data acquisition and handling, large databases can be managed, leading to the emergence of different real network topologies. Secondly, the increase in computing power has made it possible to explore networks with millions of nodes. Thirdly, there is the slow but sure breakdown of boundaries between disciplines. This can be seen by researchers because of their ability to access and use databases that facilitate an understanding of the generic properties of complex networks.

Abstract

Due to a soaring air travel growth in the last decades, air traffic management has become increasingly challenging. As a consequence, planning tools are being devised to help human decision-makers achieve a better management of air traffic. Planning tools are divided into two categories, strategic and tactical. Strategic planning generally addresses a larger planning domain and is performed days to hours in advance. Tactical planning is more localized and is performed hours to minutes in advance. An aggregate route model for strategic air traffic flow management is presented. It is an Eulerian model, describing the flow between cells of unidirectional point-to-point routes. Aggregate routes are created from flight trajectory data based on similarity measures. Spatial similarity is determined using the Frechet distance. The aggregate routes approximate actual well-traveled traffic patterns. By specifying the model resolution, an appropriate balance between model accuracy and model dimension can be achieved. For a particular planning horizon, during which weather is expected to restrict the flow, a procedure for designing airborne reroutes and augmenting the traffic flow model is developed. The dynamics of the traffic flow on the resulting network take the form of a discrete-time, linear time-invariant system. The traffic flow controls are ground holding, pre-departure rerouting and airborne rerouting. Strategic planning - determining how the controls should be used to modify the future traffic flow when local capacity violations are anticipated - is posed as an integer programming problem of minimizing a weighted sum of flight delays subject to control and capacity constraints. Several tests indicate the effectiveness of the modeling and strategic planning approach. In the final, most challenging, test, strategic planning is demonstrated for the six western-most Centers of the 22-Center national airspace. The planning time horizon is four hours long, and there is weather predicted that causes significant delays to the scheduled flights. Airborne reroute options are computed and added to the route model, and it is shown that the predicted delays can be significantly reduced. The test results also indicate the computational feasibility of the approach for a planning problem of this size.