Reduction of weather-related terminal area delays in the free-flight era

While much of the emphasis of the free-flight movement has been concentrated on reducing enroute delays, airport capacity is a major bottleneck in the current airspace system, particularly during bad weather. According to the Air Transport Association (ATA) Air Carrier Delay Reports, ground delays (gate-hold, taxi-in, and taxi-out) comprise 75 percent of total delays. It is likely that the projected steady growth in traffic will only exacerbate these losses. Preliminary analyses show that implementation of the terminal area technologies and procedures under development in NASA's Terminal Area Productivity program can potentially save the airlines at least $350M annually in weather-related delays by the year 2005 at Boston Logan and Detroit airports alone. This paper briefly describes the Terminal Area Productivity program, outlines the cost/benefit analyses that are being conducted in support of the program, and presents some preliminary analysis results.


INTRODUCTION
Free flight will revolutionize air traffic in the 21st century. The entire notions of 'delays' and 'system capacity' will change. Theorists, system users, and even system operators speak of potential time and resources savings in the billions of dollars per year. Most free-flight analyses focus almost entirely on the en-route environment, where the operational and economic gains are most evident. However, the potentially huge economic benefits of free flight may remain largely unrealized if the system is bottlenecked at terminals. Despite projected steady increases in air traffic, few new airports or runways are likely for the foreseeable future; thus, for the full benefit of free flight to be realized, airport and runway throughput must somehow be increased.

Dr. Peter F. Kostiuk
Dr. David A. Lee Robert V. Hemm Earl R. Wingrove I11 Logistics Management Institute procedures to reduce delays in the terminal area of the future. The majority of significant terminal area delays are caused by weather, and thus weather-related delays are the main focus of the TAP program; however, many of the technologies and procedures can also mitigate delays when weather is not a factor. This paper describes the TAP research program and details ongoing system studies that are being conducted to support the TAP program. Reduced Spacing Operations focuses on developing and demonstrating technologies and procedures to mitigate the reduction in arrival capacities of runways currently experienced under Instrument Flight Rules (IFR) operations. This work will result in decreasing the separation requirements between each pair of landing aircraft as well as allowing the independent operation of parallel runways spaced closer than 3400 feet apart. A significant part of this research effort is devoted to theoretical understanding and modeling of the transport and decay of the wake vortices created by aircraft in flight. The Air Traffic Management subelement is  developing and demonstrating a set of automation   aids for air traffic controllers in the TRACON. These automation aids will allow fuller utilization of the Flight Management Systems and datalinks that are increasingly available on commercial air transports to increase airport capacity, will support more closely spaced parallel runway operations, and will allow more rapid reconfiguration of operational runways and the terminal airspace.

TAP SYSTEM STUDIES
The LVLASO subelement focuses on mitigating delays in runway occupancy, taxiing, and crossing active runways due to low visibility conditions. This element includes situational display aids in the cockpit, such as taxi-map displays, and controller aids in the tower, such as ASDE-3 (Airport Surface Detection Equipment) enhancements to identify taxiing aircraft on controller displays, as well as technologies and procedures to facilitate pilotkontroller communications, such as computer-generated datalink messages.
The fourth TAP subelement, AircrafUAir Traffic Control Systems Integration, supports integration of the other subelements as well as integrated experiments and simulations to demonstrate concepts. The system studies described in this paper are a major part of this subelement.
The goal of the system studies is to enable knowledgeable, well-founded decision-making for managing the TAP program by determining the most significant weather-related delay problems in the terminal area, determining the usefulness of individual TAP technologies in a systems context, and assessing which solutions are most costbeneficial. The TAP research program will culminate in the development and demonstration of a number of technologies and procedures to mitigate weather-related delays in the terminal area. The ultimate decision to implement those solutions in the operational environment will depend on whether the FAA and the airlines (and possibly avionics equipment manufacturers as well) are convinced that it is substantially in their financial interest to invest in these technologies. Characterization of Airport Operations and Delays Before analvsis of the imDacts of the TAP technologiei and procedires could begin, detailed data on the operations of the 10 airports in various weather conditions had to be obtained.

Approach
The first step was examination of existing databases and reports available from the FAA and other government agencies, such as the Official Airline Guide (OAG) and the Consolidated Operations and Delay Analysis System (CODAS). Detailed weather histories of the 10 airports were obtained from the National Climatic Data Center.
Site visits to selected airports were used to understand more fully how airports operate in various weather conditions. Tower counts and detailed descriptions of airport operations and runway configurations used for various weather conditions were obtained from tower personnel.
Since TAP is a long-term research program and the majority of the technologies and procedures resulting from the TAP research will not be operationally deployed until the year 2005 or later, the costhenefits analyses are being conducted for the years 2005 and 2015, rather than for the present. Therefore, the TAP costhenefit analyses are being conducted relative to a baseline that already takes into account the delay reductions expected from the operational deployment of technologies and procedures being developed under other FAA and NASA near-term airporthirspace development programs. The FAA Terminal Area Forecast is being used to determine estimated air traffic for those future years.
Survey of Surface Delays and Causes During the data assessment Dhase of the studies, it was noted that very little data is available on the problems of surface movement in low-visibility conditions. Therefore, a detailed survey is being conducted to collect data on weather-related surface delays [ 11. The survey is being completed by airport managers, traffk management specialists, tower controllers plus airline ramp managers and Operations Center personnel. The focus of the survey is on identifying and prioritizing the root causes of surface delays, rather than on collecting hard numeric data, but this information will be important in determining which areas of LVLASO might have the highest payoff. The survey data will also aid in the later development and validation of fast-time simulation models of surface movement at each airport.
Analytical Modeling of the Approach and Landing Phases The effects of reducing the separation between arriving aircraft on a single runway, reducing runway 6ccupancy time, and enabling independent operation of closely spaced parallel runways in low-visibility conditions are being examined through analytical modeling of the approach and landing phases [2]. No existing analytical models were found that allowed the flexibility to accurately model the effects of the TAP technologies, so the following models were developed: Runway capacity model -A parametric model of the capacity of a single runway or set of runways operated jointly, that accounts for the effects of meteorological conditions and can be adjusted to reflect the presence or absence of various combinations of the TAP technologies. Whole airport capacity model -A model of the capacity of an entire airport as a function of meteorological conditions. This model accounts for the various combinations of runways that can be used in varying wind directions and speeds and in varying visibility and ceiling conditions, with parameters that can be adjusted to reflect the effects of various TAP technologies. Demand model -Hour-by-hour airport demand is determined based on tower counts or other data. Queuing model -A model that generates delay statistics based on a given time series of capacity and demand.
These four models are used together in the following manner. The parameterized runway and airport capacity models are used to generate a time series of airport capacity for the weather-days analyzed. A corresponding time series of demand is generated using airport traffic counts or OAG data, adjusted by reference to the Terminal Area Forecast for the desired year. The capacity and demand series are input to a queuing model, which generates statistics on delay. Economic models then estimate the financial impacts of the delays.  TAAM is a rule-based 3-D simulation tool with a powerful graphical display that facilitates the systematic and interactive evaluation of changes in operations at an airport. For example, TAAM allows a user to define gates, taxiways, runways, etc. and set usage rules and restrictions to assess improvements to ground operations. TAAM simulations of the three New York area airports, JFK, LGA, and EWR, will be used to evaluate ground delay effects of the TAP technologies that improve poor visibility operations. After modeling the airport layout, actual traffic data will be used to recreate traffic patterns for a badweather day. The TAAM model will be calibrated to emulate aircraft performance and taxi, runway, and other airport operations for the baseline scenario. A new scenario will be simulated for each increment of TAP technologies and procedures to be analyzed. For example, to simulate cockpit moving map displays, the new I $46.80 scenario would reflect higher taxi-idout speeds derived from simulation studies performed in the LVLASO element of TAP. A comparison of the delays incurred from the two scenario simulations will provide a measure of the impact of the TAP technologies and procedures.

$96.70
Life-Cycle Cost Estimation To complete the costhenefit analyses, the lifecycle costs of $1 73-40 operational depfoyment, operation, and maintenance of the TAP technologies and procedures will be estimated. Preliminary lifecycle cost estimates will be made early in the research program to aid in programmatic guidance towards the most cost-beneficial research areas. As the research program progresses, these preliminary estimates will be updated to reflect design choices and refined as more data on implementation details becomes available.

PRELIMINARY RESULTS
Some preliminary results from early TAP system studies are presented in this section. These results are from preliminary analytical modeling that has been completed for two of the ten focus airport plus some preliminary results from the surface operations survey forms that have been received to date.

Analytical Modeling
Preliminary results from analytical modeling have shown substantial benefits from implementation of TAP technologies applicable to the approach and landing phases at two airports analyzed thus far, Boston Logan and Detroit, as shown in Table 1. These results must be considered preliminary because they are subject to several limitations, as outlined below. However, they are useful for establishing the rough order-of-magnitude benefits expected from the TAP program. TAP 2 is reduced interarrival spacing plus a 20% reduction in runway occupancy time, and TAP 3 is reduced interarrival spacing, reduced runway occupancy time, plus further reduction in interarrival spacing enabled by using onboard flight management systems to accurately deliver the aircraft over the threshold at the time requested by the controller. 0 $1 6.21 $29.06 $1 9.01 $33.82 These preliminary results illustrate that the impacts of the TAP technologies and procedures will vary significantly from airport to airport. In this case, Boston cannot use key runways in Instrument Flight Rule conditions because of close spacing of parallel runways, hence Boston derives more operational benefit from the TAP delay reductions than does Detroit.
The major limitations to these preliminary analytical results are as follows: Only a subset of the TAP technologies and 0 procedures are considered Accurate modeling of the expected performance of the TAP technologies and procedures is difficult this early in a research program Available data on runway occupancy times is limited and the operating conditions are seldom known. Thus, although this study showed that reduced interarrival separation would make runway occupancy time a significant capacity factor, less conservative assumptions of baseline runway occupancy times show a different result. Because only the approach and landing phases were analyzed, assessment of broader system interactions was not included.
The analytical models produce an estimate of the minutes of delay per year for each combination of TAP technologies and procedures. The minutes of delay must then be translated into a cost savings for the airspace users. Since the various airlines and other entities use several different methods for calculating the costs of delays, and the mix of ground and airborne delay is unknown, an upper and lower bound was calculated for this study. The costs per minute of delay used for Boston and Detroit are shown in Table 2. are a help, but the controllers report that the pilots also need situational awareness displays in the cockpit for safe and efficient active runway crossings in lowvisibility conditions.
The vast majority of surface delays are incurred waiting in the departure queue, and the second highest surface delays at many airports are spent waiting for gates to become available. These queuing delays result from lack of capacity, airline scheduling, air traffic control inefficiencies, en route or destination weather, or airspace congestion. In the New York City area, airspace congestion and restrictions intensify the severity of ground delays. Since there is virtually no flexibility to divert, arrivals are heavily favored over departures when there is airspace congestion, and this increases departure delays and surface congestion.

Completion of Preliminary Analyses
Only a subset of the TAP technologies and procedures have been analyzed to date. A preliminary costhenefit analysis of all of the TAP technologies and procedures will be completed by December 1997. This will include a rough estimate of life-cycle costs, analysis of parts of TAP not yet modeled, including independent operation of closely spaced parallel runways, and analysis of additional increments and combinations of TAP technologies and procedures.

CONCLUDING REMARKS
As described in this paper, preliminary analyses have shown that implementation of the terminal area technologies and procedures under development in NASA's Terminal Area Productivity program can potentially save the airlines at least $350M annually in weather-related delays by the year 2005 at Boston Logan and Detroit airports alone. Advanced technologies and procedures to support free flight, whether en route or in the terminal area, will have to be integrated effectively. The TAP program is addressing potential bottlenecks that could greatly reduce the impact of any free-flight strategy. According to the Air Transport Association (ATA) Air Carrier Delay Reports [4], ground delays (gate-hold, taxi-in, and taxi-out) comprise 75 percent of total delays. It is likely that the projected steady growth in traffic will only exacerbate these losses. Unless terminal area delays can be mitigated, airlines will not be able to take full advantage of the en-route travel time reductions possible with free flight because schedules will still have to be padded to allow for terminal area delays.