Travelling by air is a commonly used mode of transportation, but there are certain unavoidable difficulties, like baggage problems, delays, cancellations and so on. These issues affect the traveller experience and airline operations. External factors that contribute to these issues include weather conditions, mechanical problems etc. Baggage issues incur significant costs for airlines, especially when they have to deliver lost bags. To better understand and address these challenges, airlines regularly report operational data, including cancellations, delay, overbookings of flights along with baggage complaints and other operating issues. This paper, utilizing the Facebook Prophet model, analyzes time series data related to airline baggage complaints for the first time, as per the best knowledge available till date. The aim is to forecast future baggage complaint trends, identify patterns, and provide insights that can aid the airline industry in improving overall customer satisfaction and operational efficiency. In this work, forecasting baggage complaints is carried out using the Facebook Prophet model. Added with that, a comprehensive comparison conducted between the forecasting performance of Prophet and Seasonal Autoregressive Integrated Moving Average with Exogenous Factors (SARIMAX) models on airline baggage complaint data. The evaluation metrics, including MSE, RMSE, MAE, and MAPE, consistently demonstrated superior results for Prophet. Prophet demonstrates better performance, showcasing a 7.60% lower Mean Squared Error, 3.88% lower Root Mean Squared Error, 9.69% lower Mean Absolute Error, and 2.10% lower Mean Absolute Percentage Error than SARIMAX. This signifies that Prophet, with its unique forecasting capabilities, outperformed SARIMAX in accurately predicting baggage complaints and seems to be more effective in analyzing trends over time, giving useful information to improve baggage complaints predictions in the airline industry. Also, forecasting of upcoming year’s baggage complaints using Prophet model is performed.
Predictive Modelling of Airline Baggage Complaints Using Facebook Prophet: A Time Series Analysis
Communic.Comp.Inf.Science
Analytics Global Conference ; 2024 ; Kolkata, India March 07, 2024 - March 08, 2024
2024-11-21
12 pages
Article/Chapter (Book)
Electronic Resource
English
Baggage handling -- Airline baggage handling systems
Engineering Index Backfile | 1963
|