Business Intelligence-Driven Analysis of Airline Passenger Satisfaction Using SAS Enterprise Miner

Abstract

Business intelligence is essential for converting corporate data into meaningful insights for strategic decision-making. This research presents an advanced data mining architecture using SAS Enterprise Miner to examine customer satisfaction trends for business intelligence purposes. The experimental assessment used a publicly accessible Kaggle dataset including airline passenger service and demographic data. The dataset comprises essential factors such as Age, Flight Distance, Departure Delay, Arrival Delay, Class, Type of Travel, and several service rating criteria such as Inflight Wi-Fi Service, Seat Comfort, Food and Drink, and On-board Service, assessed on a 1–5 scale. Data preprocessing methods, such as missing value imputation, categorical encoding, and normalization, were implemented before model construction. Classification algorithms were developed to forecast passenger satisfaction levels and identify significant business factors. The model's performance was assessed using accuracy, precision, recall, and AUC metrics. The findings indicate that sophisticated data mining methodologies in SAS Enterprise Miner effectively reveal essential service quality metrics, facilitating data-informed decision-making and improved business intelligence approaches.