A Privacy-Preserving Target Time Management System for Air Traffic Flow Management

In Air Traffic Flow Management (ATFM), flights are often assigned new arrival times in case of unexpected events such as poor weather conditions. The assignment of arrival times results in flight delays and thus additional costs for airspace users (AUs) and reduced efficiency for the airport. Since the impact of delays differs across flights, the airport and AUs would prefer to prioritize important flights. For this purpose, a Target Time Management System (TTMS) has been proposed that enables AUs and the airport to collaboratively prioritize flights and optimize flight lists. In this paper, we evaluate an implemen-tation of a privacy-preserving TTMS that combines evolutionary algorithms and secure multi-party computation (MPC) to protect the confidentiality of AUs’ preferences. We use data from 51 real-world regulation shared by Zurich Airport for the experimental evaluation. The results indicate that a privacy-preserving TTMS can find solutions almost as good as the solutions found by a deterministic algorithm, while protecting the preferences of AUs. In addition, runtime measurements are reported to demonstrate that the TTMS is capable of finding solutions within practical time constraints. The privacy-preserving TTMS is a promising alternative for settings in which AUs do not fully trust the platform provider.

Keywords: flight prioritization, privacy-preserving opti-mization, multi-party computation, evolutionary algorithm