Approach to Optimize the Airplane Boarding Process by using Image Recognition and Stochastic Simulation to Predict Overhead Bin Fill Levels

In single-aisle aircraft, the available storage space for carry-on baggage is inherently limited. When the aircraft is fully booked, this often results in insufficient overhead bin space, necessitating last-minute gate-checking of carry-on items, which delays the boarding process and reduces operational efficiency. A promising approach to mitigating this issue involves the integration of computer vision technologies with an appropriate data storage system and stochastic simulation to enable accurate and supportive predictions that enhance planning. In this work the YOLOv8 image recognition algorithm is used to identify and classify each passenger’s carry-on baggage into categories such as handbags, backpacks and suitcases. This data is then linked to passenger information stored in the NoSQL database MongoDB, which included seat assignments and the number of carry-on items per passenger. Stochastic simulation is applied to predict the occupancy levels of overhead storage bins across different seat rows during the boarding of the passengers. This allowed real-time assessment of whether the remaining storage capacity is sufficient to accommodate additional baggage. The results of the stochastic simulation reveal potential bottlenecks in baggage storage even before the boarding process is completed. By identifying these critical points in real time, the system can alert gate agents to proactively manage baggage distribution and mitigate overcrowding in aircraft overhead bins. This approach has the potential to streamline the boarding process, thereby reducing aircraft turnaround times and improving overall efficiency within commercial aviation.