University of Twente Student Theses


Forecasting Airport Passenger Flow to Improve Cleanliness Perception in Restrooms

Weijers, D.S. (2019) Forecasting Airport Passenger Flow to Improve Cleanliness Perception in Restrooms.

[img] PDF
Abstract:Passenger flow forecasting is essential for identifying bottlenecks, dynamic planning and maximizing customer satisfaction. Even though this essence is clear, it remains unknown whether crowdedness truly results in lower satisfaction and whether forecasting techniques are feasible in an airport restroom context. These research topics are addressed by analysing the performance of three statistical techniques on the correlation between crowdedness and cleanliness perception, and by evaluating six regression models to forecast the number of restroom visitors at Schiphol airport respectively. The real dataset from Schiphol and Asito includes 887.822 FeedbackNow votes cast in 87 restrooms over a period of 7 months, flight information of 3.018 airplanes that arrived at Pier E, and the corresponding 97.521 passengers who visited one of the two restrooms on the arrival floor. Two of the three statistical models confirm the hypothesis that crowdedness results in a negative perception. Moreover, Ridge regression is able to predict the number of restroom visitors quite successfully (R-Squared = 0.83). It is concluded that while the forecasting method is almost advanced enough to be used in practice, the correlation hypothesis needs further analysis before complete confirmation.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Business & IT BSc (56066)
Keywords:Customer satisfaction, Cleanliness perception, Forecasting, Machine learning, Prediction, Regression, Passenger flow
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page