Assessing the Impact of Stochastic Capacity Variation on Coordinated Air Traffic Flow Management

For decisions about air traffic flow management (ATFM), an accurate characterization of resource capacities at a lead time useful for planning is paramount. Unfortunately, this description is difficult to develop because of the complex nature of the airspace system and the unpredictable nature of the weather phenomena that often influence capacities. Capacity disruptions may be characterized simply by onset, duration, and severity; each of these parameters has a different effect on planning. Having a better understanding of the sensitivity of the air traffic system to uncertainty in each of these parameters can enhance decision making and improve model building. To help characterize the sensitivity of ATFM models to uncertainty in various capacity parameters, this research applies a modified Monte Carlo framework to a simplified model of capacity to identify output effects. In addition to the variations induced in resource capacities, randomness is included on demand profiles to avoid dependencies on a single demand scenario. The results demonstrate that ATFM decision making is quite sensitive to variations in each of the parameters used to characterize capacity. Typically, the impact of capacity variations is marginally increasing. The results of this type of analysis have several applications. First, the particular sensitivities of this deterministic model suggest that benefits may be realized by reformulating the model to explicitly consider stochasticity in resource capacity. Also, results suggesting greater sensitivity to specific capacity parameters may help to motivate research on mitigating uncertainty. Finally, this analysis presents an interesting application of the interplay between simulation and optimization techniques.