Quantifying Potential Fuel Burn Savings from Optimal Cruise Speed and Jonathan Lovegren g R. John Hansman Tom Reynolds Massachusetts Institute of Technology
Motivation Strong interest in operational mitigations to reduce environmental impact of aviation Joint effort between Purdue and MIT to systematically identify, evaluate and prioritize iti potential ti near-term operational changes Improving vertical and speed efficiency in cruise identified as promising area Preliminary effort to identify potential benefits pool This work was funded by the FAA, under FAA Award Nos.: 6-C-NE-MIT, Amendment No. 17 7-C-NE-PU, Amendment No. 24. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the FAA, NASA, or Transport Canada 2
Partial List of Selected Mitigations Mitigation Fuel (F) Climate (C) Air Quality Noise Implementability Potential Impact SURFACE (S) S-1: Queue Management Systems S-1.2: Advanced Systems (optimized strategies) S S P S Medium Strong S-2: Taxi Fuel Minimization S-2.4: Improved surface situational awareness, harvesting ASDE- X data S S P S Easy Mod S: Improved coordination tools S.1: Improvedinformation sharing S S S S Medium Strong S.2: Flight plan change delivery over datalink S S S S Medium Mod DEPARTURE (D) D-1: Departure procedures D-1.1: Operating in best noise configuration /A /A /A P Easy Strong D-2: Increased flexibility in departure routes D-2.1: RNP/RNAV Enabled SIDs S S P S Medium Mod CRUISE (C) C-1: Horizontal Route Efficiency C-1.1: RHSM, multi-laning P P Hard Strong C-1.2: Minimize lateral route inefficiency P P Med Strong C-2: Vertical Routing Efficiency C-2.2: Increased directional airways P P Easy Mod C-2.3: Cruise climb P P Med Strong C-2.4: Step-climb P P Easy Mod C-2.5: Increase priority for giving requested/optimal altitudes P P Easy Mod C-3: Speed Efficiency C-3.1: Individual aircraft fuel-optimized cruise speeds P P Hard Strong C-3.2: Cruise reductions P P Easy Strong C-3.3: More efficient passing options P P Med Strong 3
C-2/3: Cruise Vertical/Speed Efficiency Fuel Climate Air Quality Noise Implementability Pot. Impact P P Medium Moderate/Strong Each aircraft has an ideal minimum fuel burn altitude and speed Air traffic control restrictions and airline preferences often result in off-optimal operations Many mitigations may allow aircraft to fly nearer their optimal altitude and speed, e.g.: Increased directional airways Cruise climb Increased priority for requested altitude/speed Cruise reductions More efficient passing options 4 x RNP
Speed and Analysis: Data Sources ETMS Flight Data for 1 day All domestic flights, 9/21/29 Trajectory data in 1 min steps Latitude/Longitude Groundspeed Filed flight plan information NOAA Atmospheric Data Temperature Wind components Vertically spaced at 3 different pressure levels Laterally spaced at 32-by-32 km gridpoints Sample lateral flight profiles US Surface Temperature Profile profiles
Piano-X Aircraft Performance Primary focus on Standard Air Range (SAR): distance flown per kg of fuel SAR table of speed vs altitude mapped for each aircraft at one weight Fundamental correlation applied to include SAR sensitivity to weight Utilized step climb profiles in Piano-X to match optimum altitude with weight Validated results by checking that weight changed approximately proportionally with air density 42 B737-7 SAR (% Below Max) B757-2 Sensitivity tude (1s ft) Altit Optimal (F FL) 37-2 2 4 6 8-25 85 9 95 1 15 11 115 Aircraft Weight (kg) 6
Standard Air Range Comparison -2 A-2 SAR (% below max) CRJ-2 SAR (% below max) 44 42 B737-7 SAR (% below max) 42 (1s ft) (1s ft) -2 2 4 6 8 2 4 6 8-2 -2 MD82 SAR (% below max) SAR contours represent performance sensitivity to speed and altitude, at a single weight SAR increases approximately linearly as weight decreases 7 28 2 4 6 8 (1s ft) -25-2 2 4 6 8 B757-2 SAR (% below max) -2-2 (1s ft) (1s ft) 2 4 6 8 2
Flight Path Detailed Breakdown Distance(nm) Single Segment 1 2 5 6 7 γ de (1s ft) Altitu 2 1 Distance Cruise Climb Angle 2 4 6 8 1 12 Minutes Distance (nm) 1 2 5 6 7.4 1 Actual speed Speed calculation noisy due to limited ETMS position accuracy.2 2 4 6 8 1 12 Minutes Moving average smoothes data for processing 8
Analyzing the Actual Flight Path Performance Calculations Estimate Initial Aircraft Weight (1s ft) 2 1 Uses initial filed altitude as Segment Info: surrogate for weight estimate Location Assumes start t of cruise occurs Determine at optimum altitude Winds Speed May underestimate weight Climb Angle Loop Distance(nm) Over 1 2 5 6 7 Flight 2 4 6 8 1 12 Minutes Recalculate Aircraft Weight Calculate Fuel Burn Over Segment Lookup SAR Total Fuel Burn 9
Developing The Ideal Flight Path Estimate Initial Aircraft Weight Uses same weight determined in actual fuel burn calculation Cruise Climb Angle Determination Segment Info: Location Speed Climb Angle Performance Calculations Determine Winds Determine Ideal Alt Loop Over Flight Select Speed That Minimizes Wind-Adjusted SAR Recalculate Aircraft Weight Calculate Fuel Burn Over Segment Minimum Fuel Burn Cruise Climb Path e (FL) Altitud 41 39 37 Actual Ideal Best Case 5 1 15 2 1
Sample Flight: B757-2 from BOS to SFO Speed Profile Headwind increases ideal airspeed (FL) 39 37 Profile Actual Ideal (pres) Ideal (dens) Fuel Burn Savings 2.88% Total 7% from altitude-only improvement 2.16% from speed-only improvement 5 1 15 2 25 4 Tailwind Profile 5 1 15 2 25 Instantaneous Standard Air Range (SAR, nm/kg).18 Gspd-Aspd (kts s) 2-2 -4 Headwind -6 5 1 15 2 25 SAR (nm/kg).16.14.12 Spikes correlate with climbs and descents.1 5 1 15 2 25 Persisting operations below the ideal SAR line indicate improvement potential
Selection of Cases for Analysis The relative improvement from actual is calculated for several profiles: Case Speed 1 Best Best 2 Best Actual 3 Best Step 1 ft 4 Best Step 2 ft 5 Actual Best 6 LRC Best Commonly used aircraft spanning a variety of payload and range classes were chosen Routes were selected based on range diversity, frequency, and applicability to the aircraft type Aircraft Route* (and back) Distance (nm) # Flights B737/A LA X SFO 29 29/34 JFK ORD 64 14/3 LA X ORD 151 12/11 JFK LAX 215 6/26 B757 ATL MIA 52 22 LAX ORD 151 18 BOS SFO 2 12 MD82 JFK ORD 64 33 DCA DFW 13 25 CRJ 2 JFK DCA 19 16 LAX SFO 29 17 Dash 8 Q JFK DCA 19 8 JFK PIT 27 15 *Airport codes are representative of the city; other major airports in each metro area are included 12
Secondary Effects Temperate deviations from ISA can be significant ISA + 1C at FL39 increases density altitude by 1 ft Cruise climbs are on the order of 1s feet Optimal altitude is a function of density altitude, but aircraft fly pressure altitude Maintaining correct density altitude can mean unusual profiles B737-7 Los Angeles to Chicago Extra fuel is burned in the cruise climb This is mostly recovered in descent, but must be included A cruise climb, excluding the benefit of descent, can appear worse than level flight Temperature (C C) -48 2 4 6 Actual ISA Non standard temperatures can lead to unusual altitude profiles ) (FL) 44 42 398 396 394 392 39 Actual Ideal (dens) Ideal (pres) Descent must be included to make up for climb energy 8 5 1 15 Distance (nm) 388 2 6 8 1 12 1 13
Long Range Example: B757-2 Boston San Francisco (2, nm) 5 B757-2 Headwind Case 4 Avg Improvement: 3.73% 3 Alone: 1.36% 2 Speed Alone: 2.52% % Improvem ment 6 1 Speed e (FL) Altitud 39 37 Actual Ideal(d) 5 1 15 2 25 (FL) 39 37 5 1 15 2 25 1 2 3 4 (FL) 37 5 1 15 2 25 e (FL) Altitud 37 5 1 15 2 25 1 2 3 4 1 7 5 1 15 2 25 5 1 15 2 25 5 1 15 2 25 5 1 15 2 25 14
Medium Range Example: B737-7 Los Angeles Chicago (1,51 nm) B737-7 4 3.5 3 Speed Tailwind Case 25 2.5 Avg Improvement: 1.53% 2 Alone: 9% 1.5 Speed Alone: 1.29% % Improveme ent 1 39 1 2 3 4 45 39 e (FL) 37 Altitud e (FL) 37 (FL) 39 37 (FL) 395 39 5 1 15 2 6 8 1 12 2 6 8 1 12 385 5 1 15 1 2 3 4 5 1 15 2 6 8 1 12 2 6 8 1 12 5 1 15 15
Short Range Example: MD82 3.5 New York Chicago (64 nm) 3 MD82 Avg Improvement: 1.81% 2.5 Alone:.35% Speed Alone: 1.68% % Improvemen nt 2 1.5 1 Speed 5 1 2 3 4 5 6 7 8 9 1 11 (FL L) 315 35 295 (F L) (FL L) 315 35 295 (FL L) 29 5 1 15 2 25 1 2 29 1 2 29 1 2 1 2 3 6 5 5 5 5 1 15 2 25 1 2 1 2 1 2 16
Short Range Example: B737 B737, New York Chicago (64 nm) Eastbound Avg : 1.37% Alone: 1.1% Speed Alone: 3% Westbound Avg: 3.31% Alone: 1.71% Speed Alone: 2.25% 25% % Improveme ent 4.% 3.5% 3.% 25% 2.5% 2.% 1.5% 1.% %.% Eastbound Speed 1 2 3 4 5 6 7 % Improveme ent 8.% 7.% 6.% 5% 5.% 4.% 3.% 2.% 1.%.% Westbound Speed 1 2 3 4 5 6 7 (FL L) 2 6 (FL L) 39 37 1 2 (FL L) Earlier step-downs 2 6 L) (F Earlier step-downs 1 2 2 5 2 6 2 6 1 2 2 6 1 2 17
Sensitivity Example ) (FL) Washington Dallas (1,3 nm) 4 MD82 3.5 Avg Improvement: 2.3% 3 Alone: 1.4%* 2.5 Speed Alone: 1.35% 2 *Results possibly skewed by weight estimate 1.5 Sensitivity to weight estimate for #3, 5, and 1 9 examined Actual Ideal ) (FL) ) (FL) % Improveme ent 4.5 Speed 1 2 3 4 5 6 7 8 9 1 11 12 13 (FL) 2 6 8 29 2 6 8 2 6 8 29 2 6 8 3 5 7 9 ach M ach M 2 6 8 2 6 8 2 6 8 2 6 8 18
Sensitivity Example ) (FL) Washington Dallas (1,3 nm) Speed 4 MD82 3.5 Avg Improvement: 2.3% 3 Alone: 1.4%** 2.5 Speed Alone: 1.35% 2 improvement potential may be 1.5 exaggerated due to weight estimate 1 Sensitivity to weight estimate for #3, 5, and 9 examined 1 2 3 4 5 6 7 8 9 1 11 12 13 Actual Ideal 2 6 8 ) (FL) nt % Improveme 4.5 29 2 6 8 ) (FL) 2 6 8 ) (FL) 29 2 6 8 3 5 7 9 M ach Examined sensitivity to weight estimate on following slide ach M 2 6 8 2 6 8 2 6 8 2 6 8 19
Performance Sensitivity to Weight Estimate 4.% 3 Flights from Washington to 3.% Dallas 2.% MD82s Examined 1.% sensitivity to initial.% weight estimate Plots show fuel burn reduction 5.% from actual to 4.% improved Varying bar height 3.% indicates volatility 2.% to weight estimate 1.% Shorter bars.% represent cases ion From Actual % Fuel Burn Reducti where given weight estimate 2.5% brings improved 2.% case closer to 1.5% actual 1.% %.% Weight Estimates Initial Alternate Total Speed Only Alt Only Bar labels indicate Initial altitudes resulting from weight estimates t FL) ( Altitud de (FL) Actual Ideal Improved trajectory using weight estimate based on filed altitude 2 6 8 Total Speed Only Alt Only 29 2 6 8 tude (FL) Improved trajectories using alternate weight estimates Total Speed Only Alt Only 29 2 6 8 Altit 2
Very Short Range Flights Short flights often lack significant cruise leg Alternative analysis required to develop optimum profile CRJ-2 LAX SFO (29 nm) Distance(nm) 5 1 15 2 25 Short flights often cannot reach ideal altitude Operators stay low for speed, simplicity Weight estimation unclear Dash 8 Q JFK PIT (27 nm) Distance(nm) 5 1 15 2 25 2 Al ltitude (1s ft) 2 1 Alt titude (1s ft) 15 1 5 Ideal trajectories using alternate weight estimates 1 2 3 4 5 6 Minutes Distance(nm) 5 1 15 2 25 1 2 3 4 5 6 7 Minutes Distance(nm) 5 1 15 2 25 2 Altitud de (1s ft) 2 Altitud de (1s ft) 15 1 5 1 2 3 4 5 6 Minutes 1 2 3 4 5 6 Minutes 21
Speed and Optimization Overview Speed and Optimization Identified as Potential Opportunity Focused on Vertical and Speed Cruise Optimization i for a limited it scope of flights and aircraft type 2% cruise fuel burn reduction appears possible 1-2% from altitude improvements 2-4% from speed improvements Next steps Additional aircraft types and routes Attempt to obtain data set with actual weights Larger time scope (more than 1 day) Include optimal climbs and descents