Dynamics of Air Transport Networks: A Review from a Complex Systems Perspective

Air transport systems are highly dynamic at temporal scales from minutes to years. This dynamic behavior not only characterizes the evolution of the system but also affect the system’s functioning. Understanding the evolutionary mechanisms is thus fundamental in order to better design optimal air transport networks that benefits companies, passengers and the environment. In this review, we briefly present and discuss the state-of-art on time-evolving air transport networks. We distinguish the structural analysis of sequences of network snapshots, ideal for long-term network evolution (e.g. annual evolution), and temporal paths, preferred for short-term dynamics (e.g. hourly evolution). We emphasize that most previous research focused on the first modeling approach (i.e. long-term) whereas only a few studies look at high-resolution temporal paths. We conclude the review highlighting that much research remains to be done, both to apply already available methods and to develop new measures for temporal paths on air transport networks. In particular, we identify that the study of delays, network resilience and optimization of resources (aircraft and crew) are critical topics that can benefit of temporal network analysis.

💡 Research Summary

The paper provides a comprehensive review of research on air transport networks from the perspective of complex systems, emphasizing the importance of temporal dynamics that span from minutes to years. It begins by outlining how air transport systems are intrinsically dynamic: flight schedules, passenger and cargo flows, aircraft and crew assignments, weather conditions, and airport capacities all evolve on multiple time scales. Because of this, a static representation of the network—where nodes are airports and edges are routes—captures only a snapshot of a constantly shifting system.

Two principal methodological families are distinguished. The first, network‑snapshot analysis, constructs a series of static graphs at regular intervals (typically yearly or monthly). Researchers then compare structural metrics such as node degree, betweenness centrality, clustering coefficient, network density, average path length, and modularity across snapshots. This approach has been valuable for identifying long‑term trends, such as the rise of low‑cost carriers, the gradual decentralization of hub‑spoke structures, and the impact of regulatory changes on network topology. However, snapshot analysis discards the ordering of events and cannot capture phenomena that depend on the precise timing of flights, such as delay propagation, cascading cancellations, or short‑term congestion spikes.

The second family, temporal‑path (or time‑ordered) analysis, treats each flight as a time‑stamped directed edge that exists only during its scheduled departure‑arrival interval. In this representation, a passenger’s or cargo’s actual itinerary becomes a sequence of temporally feasible edges, respecting both spatial connectivity and temporal constraints (e.g., minimum layover times). Temporal‑path methods enable the study of:

1. Delay propagation – how a delay on one leg influences subsequent legs in real time, allowing the definition of time‑dependent connectivity and resilience metrics.
2. Network resilience – evaluation of the network’s ability to maintain functional paths when specific nodes or edges fail at particular moments, a concept that differs from traditional static robustness measures.
3. Resource optimization – aircraft routing and crew scheduling can be framed as time‑constrained path‑finding problems, facilitating joint minimization of fuel consumption, crew duty time, and passenger waiting time.

The review finds that the majority of existing literature focuses on the snapshot paradigm, while only a handful of studies have applied high‑resolution temporal‑path techniques. This imbalance is partly due to data accessibility: high‑frequency ADS‑B feeds, air traffic control logs, and detailed delay records are often proprietary or subject to privacy restrictions. Moreover, algorithmic tools for temporal networks—such as time‑weighted shortest‑path algorithms, dynamic community detection, and real‑time visualization—are still under development for the aviation domain.

Key gaps and future research directions identified include:

• Data integration – establishing standardized, open repositories of minute‑level flight schedules, actual departure/arrival times, and delay information.
• Methodological advances – extending classic network measures (e.g., centrality, assortativity) to incorporate temporal dimensions, and creating scalable algorithms capable of processing millions of time‑stamped edges.
• Multi‑scale modeling – linking long‑term structural evolution (captured by snapshots) with short‑term operational dynamics (captured by temporal paths) within a unified framework, enabling scenario analysis for policy interventions such as new hub creation or slot reallocation.
• Application domains – beyond academic curiosity, temporal‑network analysis can improve delay mitigation strategies, enhance the design of resilient air traffic networks, support environmentally sustainable scheduling (e.g., minimizing emissions through optimal routing), and aid emergency response planning when airports or airspace segments become temporarily unavailable.

In conclusion, the authors argue that embracing temporal‑path analysis is essential for a deeper understanding of air transport systems. By integrating high‑resolution temporal data with complex‑systems theory, researchers can develop more accurate models of delay cascades, assess resilience under realistic time‑dependent disruptions, and devise optimization algorithms that respect both spatial and temporal constraints. The paper calls for a concerted effort to bridge the methodological gap, develop new metrics tailored to temporal air transport networks, and apply these tools to pressing operational challenges faced by airlines, airports, and regulators.