Quantifying and Reducing Demand Uncertainty in Ground Delay Programs. Michael O. Ball Thomas Vossen University of Maryland

Transcription

1 Quantifying and Reducing Demand Uncertainty in Ground Delay Programs Michael O. Ball Thomas Vossen University of Maryland

2 Ground Delay Program (GDP) delayed departures delayed departures delayed arrivals/ no airborne holding delayed departures

3 Measuring Quality of GDP Planning and Execution reduced arrival capacity (stochastic) origin airports destination airport airborne queue throughput into airport airborne delay ground delay

4 Uncertainty in Arrival Stream slots w multiple flights flights destination airport airborne queue open slots Sources of uncertainty: cancellations, pop-ups, drift

5 Shift in Distribution of Cancellation Notification Time with CDM ave = 44 w/o CDM ave = - 49 Notification time given in minutes before OETD (Original Estimated Time of Departure) Airport = SFO

6 Effects of Compression Algorithm AAL owns but cannot use flights destination airport airborne queue AAL and UAL swap slots AAL moves flight up Net effect: win-win for airlines slots that may have gone unfilled are used

7 Questions What is cost of various forms of uncertainty and what is value of reducing that uncertainty? What is extent and nature of benefits of compression? Can any changes in approach to planning and controlling GDPs better deal with uncertainty?

8 Planned vs Actual Airport Acceptance Rate AAR: airport acceptance rate rate at which flights land t aairport Planned AAR (PAAR) flights destination airport queue Actual AAR In order to fully utilize arrival capacity, a queue must be maintained to keep the pressure on the airport This is done by regulating the value of the PAAR

9 Models Integer Program considers only flight cancellations variables: PAAR in each time period, Pr{k flights in airborne queue at end of period i} obj fcn: min total exp airborne delay inputs: flight cancellation prob, overall slot utilization, AAR Simulation Considers cancellations, pop-ups, drift

10 Airborne Delay vs Cancellation Prob (IP Model) Per-Flight Air-Borne Delay(in min) Exp Num of Open Slots in Program Cancellation Probability 7

11 Marginal Effects of Uncertainty in Presence of Cancellations, Pop-ups, Drift (Sim Model) Effect of Cancellation Probability (lambda = 20, drift = [-5,15]) Airborne Holding (minutes/flt) Cancellation Probability Utilization = 0.97 Utilization = 0.96 Utilization = 0.95 Utilization = 0.94

12 Marginal Effects of Uncertainty in Presence of Cancellations, Pop-ups Drift (Sim Model) Effect of Drift (p_cnx = 0.1, lambda = 20) Airborne Holding (minutes/flt) [-15,25] [-10,20] [-5,15] [-2.5,7.5] [-1,4] Drift Utilization = 0.97 Utilization = 0.96 Utilization = 0.95 Utilization = 0.98

13 scheduled departure times Prelude to Understanding Compression Impact: Delay vs Arrival Times ground delay departure times airborne delay arrival times Delay(f) = arr_time(f) sched_arr_time(f) Delay(f) = arr_time(f) sched_dep_time(f) dir_en_route_time(f) Tot_delay = Σarr_time(f) Σsched_dep_time(f) Σdir_en_route_time(f) Ł As long as the overall set of arrival times (arrival slots used) remains the same total delay remains the same

14 Compression Benefits Compression has filled in holes in the PAAR which has reduced the amount of assigned ground delay Possible system effects: Holes in AAR have been filled in and total delay has been reduced There were no holes in AAR so reduced ground delay has been replaced by airborne delay Uncertainty in the arrival stream has been reduced enabling a reduction in the PAAR required to achieve a given AAR and a reduction in overall airborne delay

15 Change in PAAR Policy?? PAAR during 1800 Z begin CDM GDP Days at SFO

16 Typical PAARs Used Today PAAR Time Period (hour)

17 PAAR Sample PAARs Recommended by IP Mode l Time Period (hour)