Agent-Based Modelling of Air Transport Demand

摘要

Constraints such as opening hours or passenger capacities influence travel options thatcan be offered by an airport and by the connecting airlines. If infrastructure;;policy ortechnological measures modify transport options;;then the benefits do not only depend onthe technology;;but also on possibly heterogeneous user preferences such as desired arrivaltimes or on the availability of alternative travel modes.This paper proposes an agent-based;;iterative assignment procedure to model Europeanair traffic and German passenger demand on a microscopic level;;capturing individual passengerpreferences. Air transport technology is simulated microscopically;;i.e. each aircraftis represented as single unit with attached attributes such as departure time;;flight durationor seat availability. Trip-chaining and delay propagation can be added. Microsimulation isused to verify and assess passengers’ choices of travel alternatives;;where those choicesimprove over iterations until an agent-based stochastic user equilibrium is reached. Thisrequires fast simulation models;;thus;;similar to other approaches in air traffic modellinga queue model is used. In contrast to those approaches;;the queue model in this work issolved algorithmically. Overall;;the approach is suited to analyze;;forecast and evaluate theconsequences of mid-distance transport measures.