Jit-Tat Chen

Airport Capacity Prediction with Explicit Consideration of Weather Forecast Uncertainty

This paper describes a stochastic analytical model for predicting airport capacity with a look-ahead horizon suitable for strategic traffic flow management. The model extends previous research on airport capacity estimation by explicitly integrating the impact of terminal weather and its uncertainty. Different types of weather forecast inputs are explored, including deterministic forecasts, deterministic forecasts with forecast error models, and ensemble forecasts, to produce distributions of predicted arrival and departure capacity for each runway configuration at an airport. The paper introduces a mathematical capacity prediction model and weather data sources supported by a proof-of-concept prototype implementation, including results of validation studies at Hartsfield–Jackson Atlanta International Airport. These results are compared with standard airport benchmark capacities and actual observed throughputs. Results show that our analytical airport capacity model accurately predicts maximum available a...

Integrated arrival and departure weather avoidance routing within extended terminal airspace

Analysis of historical trajectories showed that when weather is impacting terminal operations, it is common for flights to deviate from published routes, even to the extent of using departure airspace for arrivals. This demonstrates that more flexible routing is needed in terminal airspace, especially when there are weather constraints. In this paper, we present an algorithm for the integrated design of dynamic arrival and departure weather avoidance routing within extended terminal airspace. This algorithm can serve as a strategic planning tool to aid air traffic controllers in managing terminal operations during weather events. Due to safety and efficiency considerations, arrival route structures are first designed and then modeled as constraints when designing the departure routes. The algorithm combines trajectory optimization with heuristics and takes into account human factors constraints, such as distance between merge points and number of merging flows per merge point. We demonstrate through fast-time simulation that dynamic weather avoidance routing can improve terminal operation efficiencies.

Weather avoidance optimal routing for extended terminal airspace in support of Dynamic Airspace Configuration

This paper describes an algorithmic approach for designing dynamic route structures for extended terminal airspace in the presence of convective weather. The algorithm combines trajectory optimization with heuristics to design efficient weather avoidance dynamic route structures. Optimal routing is performed for as far as 150 nautical miles from major airports. In support of Dynamic Airspace Configuration, factors such as distance between merge points, number of merging flows per merge point, and robustness of routes with respect to uncertainties in weather forecasts are explicitly considered. We present analysis of historical terminal operations that was conducted to identify suitable scenarios and also to gain insights and guidance for the design of dynamic route structure. We also present the algorithm details and experimental results.

Flexible Tube-Based Network Control

One of the most restrictive aspects in today’s Air Traffic Management (ATM) system is that aircraft are still largely limited to flying routes defined by fixed, ground-based navaids. With current air travel experiencing increasing amount of delay, it is evident that this inflexible jet route system will not be able to handle future demand, especially when jet routes get closed by hazardous weather constraints. However, not having any route structure in the AS might not be a viable approach either as it could lead to high complexity and low traffic pattern predictability. For use in the ext Generation ATM system ( extGen), the etwork Flow Organizer ( FO) has been designed as a Traffic Management Initiative (TMI) to control the topology of the air traffic flow network and its flow properties. The FO alleviates en route congestion by creating airspace capacity through dense, automatically monitored traffic flows. This is in contrast to other TMIs that restrict capacity, such as Airspace Flow Program, miles-in-trail, ground delays and ground stops. The FO forms and dissolves flows based on demand and the necessity to resolve demand-capacity imbalances, thus providing structure only when needed without unnecessarily constraining traffic. Such flows allow parallel lanes of traffic, creating high traffic throughput with low complexity and workload. The flows are dynamically relocated throughout the day to be safe from weather hazards and to avoid Special Use Airspace. A concept for protected slot templates is established to allow conflict free intersection of flows and turning at intersections.

Modeling off-nominal events and mitigation strategies for Super Density Operations

This paper describes mathematical models of off-nominal events and the corresponding mitigation strategies in Super Density Operations. Research in mitigation strategies focuses on procedural and algorithmic solutions to mitigating the off-nominal condition and returning the system back to the nominal state when it is feasible to do so. Mitigation strategies investigated in this paper include use of flexible airspace to perform path stretching and reroutes, placement of holding patterns, as well as speed control to open gaps in streams of aircraft. We also describe a computation framework developed to conduct analysis of off-nominal conditions and report results of initial simulation studies demonstrating the feasibility of the developed approach.

Routing Flexible Traffic into Metroplex

This research introduces the concepts of metroplex and flexible flight. The dependency metric between the airports in a metroplex is illustrated. Moreover, this paper focuses on metroplex routing algorithm. The authors establish a network model and develop the routing algorithm for flexible flights which fly from their original airports to destination metroplexes based on the model. The routing algorithm is performed under the sector congestion constraint. A metroplex consists of several airports around a metropolitan area instead of one single airport. As a flexible flight follows the metroplex routing instructions and approaches the decision boundary of the metroplex, a scheduler in the Multi-center Traffic Management Advisor (McTMA) will decide which runway of which airport the flight will land in. Together the routing and the scheduling make a complete flexible flight operation.

Probabilistic Airport Capacity Prediction Incorporating Weather Forecast Uncertainty

This paper introduces a stochastic analytical model for generating probabilistic airport capacity predictions for strategic traffic flow planning. The model extends previous research on airport capacity estimation by explicitly integrating the impact of terminal weather and its uncertainty. In particular, it ingests different types of weather forecast inputs, including deterministic forecasts, deterministic forecasts with forecast error models, and ensemble forecasts, to produce probabilistic distributions of predicted arrival and departure capacity for each runway configuration at the airport. The paper briefly introduces the formulation of a mathematical model and weather data sources supported by a proof-of-concept prototype implementation. It also introduces a methodology for validating probabilistic airport capacity predictions and results of validation studies at Hartsfield-Jackson Atlanta International Airport. These results are compared with standard airport benchmark capacities and actual observed throughputs.

Co-evolutionary Approach to Improve Robustness of Routing Algorithms against Disruptive Events on the Airport Surface

One of the shortcomings of today’s airport operations is the lack of optimization-based routing and guidance support incorporating various types of factors that can adversely impact routing of aircraft on the airport surface. Even though many types of routing algorithms have been developed over the years to provide efficient routing options for taxiing aircraft at an airport, very few of them, if any, have been thoroughly tested for their robustness in response to disruptive events that occur on the airport surface. The paper describes a novel concept for identifying efficient routing algorithms that are robust enough to respond to disruptive events that often occur during taxiing operations. The concept uses Co-evolutionary Red Teaming methodology which employs a class of co-evolutionary algorithms to automatically search for robust routing algorithms subjected to disruptive events. In the concurrent search process conducted by co-evolutionary algorithms, disruptive events, or their combinations, that cause major impact for airport surface operations are also discovered.

Generating Topologically Different Weather Avoidance Routes Using Mincut as Constraints

This paper introduces an algorithm for generating a set of dynamic weather avoidance routes that are topologically different. The algorithm consists of two main components: pathfinding and constraints formulation. Pathfinding can be accomplished with any standard pathfinding algorithms such as Dijkstra’s or A* search. The main contribution of this paper is the constraints formulation component which uses mincut as constraints to steer the pathfinding algorithm towards generating topologically different routes. In this application, the mincut of an airspace segment effectively models the bottleneck of flows due to weather constraints for a prescribed flow direction (e.g., east to west flow). Thus, by sequentially incorporating mincuts as constraints into the pathfinding algorithm, a set of topologically different routes can be found.

Conflict Resolution Algorithms for Super Density Operations Impacted by Off-Nominal Events

In this paper we present an approach for resolving conflicts in Super Density Operations (SDO) impacted by off-nominal events. We describe a dynamic routing algorithm in conjunction with a dynamic stochastic route blockage model to predict and efficiently resolve trajectory conflicts in SDO airspace. The proposed algorithms define new flight trajectories that avoid conflicts with off nominal events caused by irregular aircraft behavior or unexpected weather conditions. The paper describes a mathematical model and a set of algorithms implemented in a prototype simulation system. This prototype was used to conduct a series of simulation studies whose results are described and analyzed.