This paper explores how artificial intelligence and robotics can transform airport processes by optimising functions such as baggage handling and check-in using AI models and robotics. Nine machine-learning algorithms were applied. Key findings include a 25% reduction in flight delays using linear regression, a 15% decrease in lost revenue from no-shows with logistic regression, and a 20% reduction in baggage mishandling through decision trees. K-means clustering insights resulted in a 10% increase in ancillary revenue via targeted marketing. Principal component analysis (PCA) accounted for 85% of operational data variance, improving decision-making and reducing predictive maintenance costs by 18% with 92% accuracy and an F1 score of 0.91. Gradient-boosting reduced passenger check-in wait times by 30%. Convolutional neural networks (CNNs) improved security efficiency by 15% with 94% accuracy. Recurrent neural networks enhanced passenger flow forecasting, reducing congestion with a MAPE of 4.2% and an R-squared of 0.79. Overall, AI and robotics show potential.
Jindal et al. (Wed,) studied this question.