At congested airports outside of the US, capacity is managed through the allocation of arrival and departure slots. Many optimization models for this have been proposed, capturing in detail the regulations for allocating slots. However, a common weakness of many of these models is that they do not incorporate the inherent uncertainty in when aircraft arrive and depart and the length of time the corresponding runway movements take.

Ignoring these uncertainties can lead to queueing on the ground and in the air if runway throughput is assumed to be too high, and overly conservative scheduling if this is assumed to be too low. In this work we propose a new two-stage stochastic linear programming model for slot allocation, which incorporates the described uncertainty and presents some preliminary numerical results.