Outlier detection is a key component of several machine learning approaches. However, many existing techniques, especially for multi-dimensional signals, are not interpretable and do not explain why a specific classification was assigned to a particular data point. Another limitation is that most methods only consider the magnitude or intensity of the signal, and not its spatial distribution. We present a spectral approach to identify outliers based on the spatial distribution of a signal across the nodes of a graph without any explicit assumptions on the underlying probability distribution of the signal. By applying these techniques to airport delays, we not only identify outliers in the spatial distribution of delays, but also gain insights into the delay dynamics. Specifically, we compare spatial delay distributions in the US and China during the period 2012-17, and identify several interesting characteristics pertaining to critical airports for outlier detection. We characterize typical variabilities in the delay distributions, and the frequency of occurrence of outliers. Our results highlight the differences between the operational dynamics of the US and Chinese air transportation systems, and contribute to performance benchmarking between different airspace systems. © 2020 IEEE.

Network science has been a promising tool for characterizing and understanding complex systems. A challenging problem in network science is to uncover the community structure of the network. Community structure generally presents the partition of the nodes in the network into several groups based on various structural properties or dynamic behavior. In this paper, we analyze the community structure of Chinese airport network based on Stochastic Block Models (SBM). Different from exisiting studies, the Chinese Airport Delay Correlation Network (CADCN) is constructed with airports as nodes and the correlations between hourly delay time series of airport pairs as edges. To analyze the temporal patterns of community structures, we employ spectral clustering method and classify Chinese airports into 6 different communities. Airports within each community have closer relationships to each other on the delay propagation. A similar investigation to the traditional Chinese airport network (CAN) is carried out based on SBM as well. By comparing the results of two networks, we find that the CADCN has the advantage in revealing the implicit delay correlation than the directed flights connection between airports. Our findings have potential meanings to understand the spread of flight delays and to develop relevant management and control strategies. © 2020 IEEE.

An Airport-Sector Network Delays model is developed in this paper for flight delay estimation within air transport network. This model takes both airports and airspace capacities into account by iterating among its three main components: a queuing engine, which treats each airport in the network as a queuing system and is used to compute delays at individual airport, a Link Transmission Model, which computes delays at individual sector and transmits all air delays into ground delays, and a delay propagation algorithm that updates flight itineraries and demand rates at each airport on the basis of the local delays computed by the queuing engine and flow control delays computed by the Link Transmission Model. The model has been implemented to a network consisting of the 21 busiest airports in China and 2962 links that represent to 151 enroute control sectors in mainland China, and its performance is evaluated by comparing with the actual delay data and results of Airport Network Delays model. It is found that the proposed model is well-suited for simulating delays in air transport system where either airports or airspace could be the bottleneck of the system. © 2018 IEEE.