Ground Handling Process Optimization Model Linked to Flight Delay Prediction Results

To address the persistent issue of flight delays, this research presents a novel approach that integrates delay prediction with the optimization of ground service operations. A Random Forest model is employed to forecast flight delay durations, achieving an impressive 100% accuracy rate under the 15-minute delay threshold. Based on the prediction results, individual flights are assigned unique delay coefficients, which are then used to construct an optimization model for ground services. This model leverages a Genetic Algorithm, enhanced through improved gene encoding in the initial population using a segmented encoding strategy. The application of this optimized framework to schedule and manage ground service vehicles effectively eliminates delays across all flights in the dataset.