The Complexity and Interconnectivity of the National Airspace System Flows and Airports Mounting air travel delays seem to be a fact of life. On the one hand, air commerce is universally used due to its increasing affordability, motivating the airlines to put more aircraft into service and to pack them full. This is good for the economy. On the other hand, the increased density and complexity of the resulting traffic on an already burdened air traffic control infrastructure is causing severe system meltdowns at an increasing rate. As the Federal Aviation Administration's (FAA) federally funded research and development center, The MITRE Corporation's Center for Advanced Aviation System Development (CAASD) helps the FAA understand the genesis and dynamics of current problems in the air traffic management (ATM) system and proposes solutions. To understand the problems behind the increasing frequency of severe delays, CAASD's analytical team modeled the interconnectivity between traffic flows and airports. Next, we looked at the ripple effect caused by minor perturbations on the local level. What we found can be nicely described by the now famous problem from chaos theory: a butterfly flapping its wings on one side of the globe can theoretically cause major weather disruptions a half a world away. The butterfly flapping is a local phenomenon while the result has a global impact. In one case study presented here, an overdemand problem at Newark International Airport (the "butterfly") resulted in the movement or holding of more than 250 aircraft inside the ATM system (the "global impact"). The Analysis Although delays often occur in the National Airspace System (NAS), it is not always easy to diagnose the causes. CAASD uses several high-fidelity simulations and models of airport and airspace events and flows to link what is occurring in the air to capacity on the ground. The following scenarios illustrate how seemingly small, localized problems can rapidly expand across the country. The scenarios illustrate the interdependence of the resources in the NAS and the impact of loss of flexibility, or buffering capacity, on the overall system performance. The data used in all the scenarios is based on actual events and are part of the analysis used to develop the NAS Operational Evolution Plan. - The first scenario occurs on a clear weather day in which a problem is caused by a demand/capacity imbalance at an airport.
- The second scenario illustrates how severe weather disrupts a major traffic flow, causing a ripple effect across the country, including in areas where weather is not a factor.
- The third scenario shows how severe weather conditions at a specific airport can have a system-wide effect.
Setting the Stage The airspace above the continental United States (CONUS) is divided into 20 control jurisdictions called air route traffic control centers (ARTCCs), or en route centers (See Figure 1). All aircraft routes transit airspace in at least one ARTCC.
Figure 1: Air Route Traffic Control Centers (ARTCCs) over the CONUS | As they have for decades, most aircraft today navigate via a ground-based system of radio beacons. (This is slowly changing as aircraft operators equip with more precise satellite navigation systems.) To navigate using the ground-based system, an aircraft tunes into a beacon's frequency and then flies either towards it or away from it. The most commonly used routes between ground beacons are tested by the FAA to ensure their accuracy. Such routes are called air routes, hence the name of the en route centers. For a major airport, there is a widening fan of airways that meet at the airport and extend to different areas of the country. For example, aircraft travelling to Newark International Airport use the airways illustrated in Figure 2 (for a larger view, click on the images).
| | | Figure 2: Aircraft arrival routes into Newark International Airport going through Cleveland, Indianapolis, and Memphis centers. | By itself, the problem of guiding arrivals through the airspace and to the airport doesn't seem that bad. After all, there's lots of airspace, right? Well, not really. Newark is but one of the busy airports of the Northeast. Adding the arrival routes for other airports complicates the control picture (See Figures 3-6).
 |  | | Figure 3: Aircraft arrival routes to Newark and Kennedy airports. | Figure 4: Arrival routes to Newark, Kennedy, and LaGuardia airports. |  |  | | Figure 5: Arrival routes to Newark, Kennedy, LaGuardia, and Philadelphia airports. | Figure 6: Arrival routes to New York and Washington Metropolitan area airports. | Adding the routes to other underlying airports within and around the "fan," and adding the north-south crossing traffic, forces the picture to "fade to black." There used to be a "big sky" up there, but it very nearly full. The busy airports of the east coast work at or near capacity, which means that these airways are nearly always full. In turn, there is no room to park excess traffic into holding patterns because such patterns can possibly interrupt the flows of traffic along many routes.
|
