The boarding process of airlines is an important component of air travel efficiency and passenger satisfaction (Kierzkowski and Kisiel in Proc Eng 187:348–355, 2017; Cotfas et al. in Symmetry 12:1087, 2020; Memika and Polat in Intell Autom Soft Comput 35:2687–2702, 2023). The trivial boarding processes, such as sequential or zone-based boarding, often lead to incompetency, congestion, and pace anomalies. To deal with these problems, we have leveraged AI/ML predictive modeling methods to enhance the boarding process by real-time prediction of boarding times accurately. The AI/ML predictive model technique offers the potential to revolutionize the boarding process by providing airlines with valuable insights into boarding dynamics and passenger behavior. The innovative boarding platform developed by TCS provides: 1. Analysis of flight information, and other relevant factors using historical boarding data by means of Statistical Process Control (SPC) charts, 2. Prediction of an overall boarding time taken by all the passengers to board a particular fleet, 3. Prediction of the congestion and pace anomalies occur while passengers stay inside the jetbridge configured to the fleet, 4. Estimation of the boarding time is taken by remaining passengers scheduled to board the particular fleet after each time slot of the previous boarded passenger, and 5. Developing AI models to predict execution time of boarding process and its subprocesses by means of AI predictive process mining, allowing airlines to better handle ground operations, gate assignments, and staffing levels. Deploying prediction models and process mining for boarding time optimization can obtain significant benefits for airlines and passengers alike. By minimizing boarding-related delays and congestion, airlines can improve on-time performance and operational efficiency. The passengers also stand to benefit from reduced waiting times, faster onboarding with dual experiences and increased satisfaction with their air-journey.