A machine learning model to predict runway exit at Vienna airport. (November 2019)

• Record Type: Journal Article
• Authors: Herrema, Floris; Curran, Ricky; Hartjes, Sander; Ellejmi, Mohamed; Bancroft, Steven; Schultz, Michael
• Is Part Of: Transportation research. Volume 131(2019)
• Page Start: 329
• Page End: 342
• Publication Date: 2019-11
• Subjects: Component -- Runway utilisation -- Runway capacity -- Runway occupancy time -- Gradient boosting; AIP Aeronautical Information Publication -- AMAN Arrival Manager -- APSS Statistical Package for the Social Science -- AROT Arrival Runway Occupancy Time -- ATCO Air Traffic Controller -- AUC Area Under the Curve -- BD Big Data -- CAST Comprehensive Airport Simulation Technology -- DMAN Departure Manager -- dmax maximum tree depth -- FL Flight Level -- FN False Negatives -- FP False Positives -- GB Gradient Boosting -- HIRO High Intensity Runway Operations -- ICAO International Civil Aviation Organisation -- ILS Instrument Landing System -- lmin minimum leaf size -- MAD Median Absolute Deviation -- ML Machine Learning -- NREX procedural or non-procedural runway exit taken -- RU Runway Utilisation -- RWY34 runway 34 -- TP True Positives -- TR True Negatives -- VIE Vienna; Logistics -- Periodicals; Transportation -- Periodicals; 388.011
• DOI: 10.1016/j.tre.2019.10.002
• Languages: English
• ISSNs: 1366-5545

Abstract

Highlights: The runway exit utilised is key in generating predictions with respect to the predicting runway throughput. The ML technique Random Forest that we use to build the model is fast, intuitive and interpretable. The model can help the airport managers to understand the driving features of the runway exit to be used. The model can update the predictions in real-time. Abstract: Runway utilisation is a function of actual yearly runway throughput and annual capacity. The aim of the analysis in this project is to find data driven prediction models based on the features and relevant scenarios that might impact runway utilisation. The Gradient Boosting machine learning method will be assessed on their forecast performance and computational time for predicting the procedural and non-procedural runway exit to be utilised after the landing rollout. The Gradient Boosting method obtained an accuracy of 79% and was used to observe key related precursors of unique data patterns. Tests were conducted using runway and final approach data consisting of 54, 679 arrival flights at Vienna airport.