Solving Air Transport Contingencies by Using Genetic Algorithms – RUCCMAN

Abstract

Transportation networks are dynamic systems that suffer continuous unexpected events. People in charge of managing them in an efficient way usually make their decisions based on their personal expertise, relevant data and the accumulated experience gained on the matter. However, the cognitive capacity of the human brain is limited, and therefore the design of the solution cannot take advantage of the extensive data available. Despite the fact that this paper deals with the specific transportation network of Airport management, the proven excellence and virtues of the application of Genetic Algorithms to support the decision-making process can be extended to any transport domain. This paper addresses an airport management scenario, in particular, the management of runway configuration changes. When wind forecasts lead to thoughts of the need to change the use of the existing runway configuration, the person in charge of making that decision has to do his best by stating when and how to execute the change. Concurrently, there are flights approaching the airport, flights taxiing to the runway holding points, while other flights are still on-block at their gates. The better the decision is, the less the related delay will be. Nevertheless, delay is not the only performance indicator that may interest an Airport Manager; other aspects such as fuel consumption, safety, capacity, predictability and so on may be determinant. Consequently the aforementioned aspects introduce a new and wider set of parameters to take into account when making the decision. In order to support this complex decision making process, an automated tool, based on genetic algorithms and called RUCCMAN (Runway Configuration Change MANager), has been designed and validated.