Benefits Assessment Methodology for an Air Traffic Control Tower Advanced Automation System Tom Reynolds, Rich Jordan, Masha Ishutkina, Rob Seater & Jim Kuchar 10 th AIAA Aviation Technology, Integration and Operations (ATIO) Conference Fort Worth, TX -- 13-15 September 2010 MIT Lincoln Laboratory Slide-1 TGR 9/14/2010
Outline • Overview of system • Need for benefits assessment • Methodology • Application/Data analysis • Results: Informing system development priorities • Summary MIT Lincoln Laboratory Slide-2 TGR 9/14/2010
Tower Flight Data Manager (TFDM) • Integrated tower system being considered for development by FAA Enablers External Sources Tower Flight Data Manager Consolidated tower systems Enhanced cross-domain information exchange Decision support tools Terminal and Surface Surveillance Benefits Decision Support Remote / Enhanced Robust operations Tools (DSTs) Visual Awareness Reduced delay, fuel, environmental impact Flight Plan Data Enhanced safety Ability to support remote operations: Staffed NextGen Tower (SNT) Traffic Flow Constraints Operational Users Surveillance Display Flight Data Manager Tower controllers Flight data, Clearance, Ground, Local, Supervisor Flight Operations Data TRACONs, ARTCCs Flight Operations Centers Ramp Tower MIT Lincoln Laboratory Airport Authority Slide-3 Net-centric Infrastructure Weather / Hazards TGR 9/14/2010
Need for Benefits Assessment • Quantifies how well the new system performs relative to baseline New Performance Metric system New system • Needed for Investment benefits Analysis to make business case for Baseline continued development system and/or deployment Increasing demand • Leads to understanding of system inefficiencies Time and causality to help guide capability development MIT Lincoln Laboratory Slide-4 TGR 9/14/2010
FAA Standard Benefits Assessment Methodology • FAA defines 11-step benefits analysis methodology • Distilled version: 1. Understand the program 2. Identify relevant performance metrics 3. Identify current & future “baseline” system performance 4. Identify current & future “new” system performance 5. Define the benefits impact 6. Convert to economic values and compare to costs 7. Report MIT Lincoln Laboratory Slide-5 TGR 9/14/2010
TFDM Benefits Assessment Methodology 1. ConOps 2. Metric Functional Reqts TFDM Identification Metric Operatl Assessment Archived Data Baseline 3a. Current Baseline e.g. ASPM, 4a. TFDM Capability Inefficiency ASDE-X System Metric & causality Development Predictions identification Time Future 4b. Future TFDM 3b. Future Baseline System System Metric System Metric Forecasts Predictions (Alternative 1..n) Predictions e.g. TAF, FACT 1. Understand the new + - system 2. Identify relevant 5a. Benefits Claimed metrics +- by Other Systems 3. Establish baseline metric values 5b. TFDM Benefits 4. Establish new system metric values Monetization +- 5. Define the benefits 6a. TFDM Costs impact 6. Convert to economic 6b. TFDM Cost / Benefit values and compare Analysis to costs 7. Report MIT Lincoln Laboratory 7. Report Slide-6 TGR 9/14/2010
TFDM Benefits Assessment Methodology Application • Step 1: Primary objective of TFDM is to improve efficiency of surface operations • Step 2: Taxi-out delay time & fuel burn performance metrics • Step 3a: Current baseline system performance – ASPM analysis – ASDE-X analysis • Step 3b: Future baseline system performance – Queuing model • Step 4a: Informing TFDM capability development • Step 4b: Future TFDM system performance • Step 5/6/7: TFDM cost/benefit analysis and report MIT Lincoln Laboratory Slide-7 TGR 9/14/2010
TFDM Investment Analysis Airports SEA PDX MSP BOS BDL PVD MKE DTW LGA EWR ORD CLE JFK SLC PHL MDW PIT BWI SFO IAD DEN ADW CVG MCI DCA STL SDF LAS CLT LAX SNA MEM PHX SAN ATL DFW IAH MSY HOU MCO ANC FLL HNL MIA MIT Lincoln Laboratory Slide-8 TGR 9/14/2010
Current Baseline System Performance ASPM Analysis • FAA Aviation System Performance Metrics (ASPM) data extracted for analysis airports • Taxi-out delay time: average versus unimpeded push-back- to-wheels-off time • Taxi-out delay fuel: Delay time x Fleet-mix-weighted fuel flow – Fuel flow for individual aircraft from ICAO ground idle rate – Assumes all delay absorbed with engines on (upper bound) MIT Lincoln Laboratory Slide-9 TGR 9/14/2010
Current Baseline System Performance ASPM Analysis (2008) (100% = Actual Taxi-out Time or Fuel is Double Unimpeded) 280 140% Time Fuel % Increase over unimpeded 260 Taxi-out Delay % Increase Over Unimpeded 240 120% Total Taxi-out Delay Fuel (tonnes/day) Total Taxi-out Delay Time (hours/day) 220 200 100% 180 160 80% 140 120 60% 100 80 40% 60 40 20% 20 0 0% STL ANC ATL BDL BOS BWI CLE CLT CVG DCA DEN DFW DTW EWR FLL HNL HOU IAD IAH JFK LAS LAX LGA MCI MCO MDW MEM MIA MKE MSP MSY ORD PDX PHL PHX PIT PVD SAN SDF SEA SFO SLC SNA MIT Lincoln Laboratory Average total delay: 2533 hrs/day (925 khrs/yr), 1874 tonnes/day (684 ktonnes/yr) Slide-10 TGR 9/14/2010
Current Baseline System Performance ASDE-X Analysis ASDE-X • Airport Surface Detection Equipment-Version X (ASDE-X) surveillance allows identification of location of delay on surface – Gate – Spot – Queue – Runway • At these locations, inefficiencies can be observed & control mechanisms applied • ASDE-X data from Dallas-Fort Worth (DFW) airport analysed – TFDM prototype site MIT Lincoln Laboratory Slide-11 TGR 9/14/2010
Current Baseline System Performance ASDE-X Analysis Runway Runway Wheels-off Spot Gate Queue Enter e.g. Total Taxi-out Time Position Performance Metric: Runway & hold Benefit queue gained from delay Taxi-out Delay “Benefits pool” TFDM Alternative n Remaining Spot “avoidable” delay delay “Unavoidable” delay Queue Runway Ramp Taxi MIT Lincoln Laboratory Slide-12 TGR 9/14/2010
Current Baseline System Performance ASDE-X Analysis 11 • ASDE-X observed delay: 6.1 mins 10 ASDE-X delay wrt 10 th %ile: • 9 4.1 mins Relative to Unimpeded ASPM delay wrt 10 th %ile: • 8 4.3 mins Taxi-out Delay 7 (mins) 6 5 4 3 2 1 0 VMC IMC n ≈ 3000 n ≈ 1500 MIT Lincoln Laboratory Slide-13 TGR 9/14/2010
Future Baseline System Performance Queuing Model • Investment analysis period: 2015-2035 • Queuing model developed to project taxi-out delay time & fuel at analysis airports into future • Assumptions: – Runway is dominant airport constraint – Poisson demand rates – Exponentially-distributed service times • Model inputs: – Demand: FAA Terminal Area Forecast – Capacity: FAA FACT2 Airport Capacities (2007-2025, no increase 2025-2030) – Average delay capped at 15 mins in VMC and 45 mins in IMC (consistent with system evolving when delays increase) MIT Lincoln Laboratory Slide-14 TGR 9/14/2010
Future Baseline System Performance Queuing Model 7 Delay Time Across 43 Airports 6 Annual Runway Queueing (Relative to 2008) 5 4 3 2 1 0 2005 2010 2015 2020 2025 2030 2035 2040 Year MIT Lincoln Laboratory Slide-15 TGR 9/14/2010
Future Baseline System Performance Queuing Model 100 • TFDM capabilities 90 Delay Time Across 43 Airports Cumulative Runway Queueing should be designed to 80 deliver benefits against (Relative to 2008) 70 this portion 60 50 Unavoidable Delay 40 (notional) Claimed by 30 other Unique savings systems 20 available to (notional) TFDM 10 (notional) 0 0 20 40 60 80 100 % Cumulative Runway Queuing Delay Reduction MIT Lincoln Laboratory Slide-16 TGR 9/14/2010
Informing TFDM Capability Development • Mapping delay location to possible causality Location Identified Causes TFDM Opportunities of Delay Aircraft not ready Situational awareness Ground crew not ready Situational awareness Ramp Ramp blocked Situational awareness Forgotten at spot Efficiency improvement Back propagation of delay Indirect impact Runway crossings required Situational awareness Taxi Long taxi route Efficiency improvement Taxiway capacity limit Efficiency improvement Runway crossings by others Situational awareness No airborne route available Efficiency improvement Queue Runway capacity limit Efficiency improvement Inefficient departure Efficiency improvement sequence Aircraft not ready Situational awareness Runway crossings by others Situational awareness Runway Aircraft performance Situational awareness No airborne route available Efficiency improvement MIT Lincoln Laboratory Slide-17 TGR 9/14/2010
Recommend
More recommend
