Prescriptive Analytics in Airline Operations: Cost Index Optimization through Improved Arrival Time Prediction

Abstract

Facing enduring cost pressure, the airline industry has turned to machine learning to enhance their operations. In our work, we develop a predictive analytics model, focused one improving aircraft arrival time prediction taking as inputs aircraft, airport and weather information. Furthermore we integrate our arrival time prediction with different speed levels. Through this we are able to determine the optimal aircraft speed by weighing fuel costs against cost of time. The study is based on ensemble model of linear regression and gradient boosting. The optimization uses actual flight data of a European airline.

Biography

Anna Achenbach (born 1990) is a Ph.D. student and research assistant at the WHU – Otto Beisheim School of Management (Chair of Logistics Management) since August 2015. Her research is focused on Advanced analytics in operations, Predictive modelling, and Large‐scale data mining. Anna Achenbach holds a Master of Science in Global Logistics. She studied at Technische Hochschule Ingolstadt (Germany) and Kühne Logistics University (Hamburg, Germany). During her studies she focused on Operations Research and Supply Chain Management and worked as a research assistant. In her master thesis she analyzed the application of Big Data Analytics in airline operations.
Between 2012 and 2013 Anna Achenbach gained practical experience as a business analyst specialized on supply chain management and business process outsourcing while working for Nokia. Besides that she has worked as an intern for the demand planning department of Colgate‐Palmolive and in various humanitarian logistics organizations. As part of her educational and professional career Anna Achenbach had the chance to study and work abroad including India, Jordan and Namibia.