Multi-objective Integer Programming Model and Algorithm of the Crew Pairing Problem in a Stochastic Environment

Crew scheduling is an important production planning of airlines. Being optimized crew scheduling could make full use of human resources, and reduce operating costs. The traditional airline crew scheduling model is deterministic and does not include potential disruptions due to weather, air traffic control, etc. To take into account of effects of random factors such as weather, air traffic control, passenger demand, etc., we develop a stochastic chance-constrained programming model (SCCPM) for minimizing the crew cost and maximizing the passenger satisfaction. Based on Monte Carlo method, Back Propagation (BP) neural network and genetic algorithm, we develop a hybrid intelligent algorithm to solve the model. To evaluate the robustness of the model, the signal to noise ratio (SNR) method is included in this paper. We present computational results which show the effectiveness of our SCCPM and the hybrid intelligent algorithm. Key–Words: airline operations, crew scheduling, passenger satisfaction, stochastic chance-constrained programming, hybrid intelligent algorithm.