Machine Learning: An Airport Operations AllyBy
Airports are non-stop infrastructures. They open every day of the year – a continuous activity that takes place in a complex ecosystem, in infrastructures of limited capacity from which maximal performance must be obtained.
One of the critical processes is planning the allocation of platform stands, where do the planes park while they are in the airport. This planning is done based on multiple variables such as flight plans, airline preferences, types of services, ground crew services…
These days, the use of algorithms has largely automated this process. However, despite the advanced scheduling of aircraft departures and arrivals, there are still circumstances hard to accommodate that still are being processed manually.
Ikusi is developing an algorithm, based on machine learning and big data, capable of responding as well to these cases more difficult to automate with regard to the allocation of stands, thus optimizing airport capacity.
To do this, the data processed by the algorithm is being enriched with historical operations data reflecting multiple circumstances which, when added to the basic rules used in these processes, will endow Beluga – the Ikusi solution to optimize airport operations and resource allocation – with greater capacity to automate the stand allocation process.
This all takes place in a dynamic context, where the solution monitors the stand allocation process in real time and proposes optimal solutions to airport managers for the efficient use of airport capacity.
This way, in addition to availing of a complete allocation plan for an operation day, there are also available real-time responses to the usual incidents and events in the daily routine of airports.
This represents one more step in incorporating machine learning and big data techniques to the technological solutions for the analytical capacity required to build a smart airport ecosystem.