This dissertation is concerned with airport ground network operations and runway design problems. An aircraft landing simulation model is proposed to predict landing aircraft operations in airfield networks considering gate location and taxiway network information. By using the integrated simulation results, an improved dynamic programming model is established to solve the runway exit location problem.

The simulation model uses aircraft kinematic functions coupled with individual parameters to describe the landing process. A multiobjective optimization and a shortest path algorithm are used to predict the aircraft exit choice and expected taxiway path in a runway-taxiway network. By recognizing pilot motivation during the landing process, several influencing factors such as terminal location, traffic density, runway and weather conditions are considered in the aircraft landing simulation. Random variables such as aircraft approach speed, deceleration rate and runway exit speed are created to represent the stochastic aircraft landing behavior by using a Monte Carlo sampling technique. The model results, with some simplifications, could be used to solve runway exit location problem by providing the expected distribution of aircraft landing distances and aircraft runway occupancy times. With real-time input data, the model could also provide information on aircraft exit choice, runway occupancy times and shortest taxiway path to an assigned terminal location for both the pilot and the air traffic controller in a ground traffic automatic control system.

The dynamic programming model is based on previous research results with substantial improvements. Instead of only focusing on the runway configuration, the proposed model recognizes airline terminal location as an influence factor in the runway exit location problem and introduces this influence in the optimization procedure.

An interactive computer program has been developed in C language on a RISC 6000 workstation to perform all these tasks. The choice of a workstation platform improves the performance of the dynamic programming algorithm when compared to personal computer solutions.