Procedures for Off-Nominal Cases: Three Closely Spaced Parallel Runway Operations Savita Verma, Sandra Lozito, Deborah Ballinger Thomas Kozon, Herbert Resnick, Gordon Hardy, Ramesh Panda, Darrell Wooten & Diane Carpenter, NASA Ames Research Center Perot Systems / NASA Ames Raytheon Corporation SAIC/ NASA Ames ATM 2009 Conference June 29- July 2 2009 1
Agenda Background Approach Focus of the study Results Conclusions Future Research 2
Background Motivation Potential capacity gains from third runway 750 ft apart especially because it could be added within the existing foot print of the airport No previous research on procedures for three parallel runways 750 ft and their impact on workload and situation awareness 3
Background Previous Research Concept No. of runways Runway Separation Researcher / Research area Triple at Hartsfield, Atlanta Multiple Parallel Approach Program- (MPAP) 3 1000ft/ 4000ft 3 4000 ft and 5300 ft Variations on simultaneous approaches Gladstone (2000) Mitre Breakout maneuvers on three runways Magyarits & Ozmore, 2002 FAA SOIA/ MPAP 2/3 >=750 ft Non-Transgression Zone (for breakout maneuvers) 4
Focus Focus of Study Investigate procedures, roles and responsibilities for offnominal situations that require a breakout maneuver for simultaneous approaches to three parallel runways 750 ft apart. Feasibility of the procedures Impact of procedures on workload and/or situation awareness levels 5
Independent Variables Approach Nominal vs. breakout Cause of breakout: Aircraft deviation (blunder) Crosswind blowing wake towards the trailing aircraft Location of breakout: Above 500 ft Between 200 ft and 500 ft Relative position of the simulator Center Trailing 6
Approach Experimental Procedure Advanced Concepts Flight Simulator: digital, audio, and video data collection Six degree motion Performance similar to Boeing 757 Side stick control 8 days of data collection - 8 participants Total runs = 24 per participant. Each run was 10-15 min long followed with questionnaires and debriefs. All runs had visibility of 2 nm at 400 ft. 7
Approach Our Approach Runways 750 ft apart Dynamic S- trajectories adaptive and displayed in real time DGPS*, ADS-B**, winds on a/c assumed S trajectory parallels 1.5 nmi for center a/c and at 3 nmi for the trailing aircraft *Differential Global Positioning System **Automatic Dependent Surveillance System 8
Approach Procedure / Scenario Navigational Display Ownship Coupling Point 9
Displays for Breakout Procedure Normal conditions Approach Breakout trajectory wake Primary Flight Display LSI wake Navigation Display LSI Predictor Dots 10
Approach Breakout Procedures and Displays Wake drifting towards follower Leader s wake drifted by one wing span Follower s center of gravity in leader s wake 11
Approach Breakout Procedures & Displays Aircraft Blunder Leader deviates 60 ft towards follower Leader deviates 120 ft towards follower 12
Approach Breakout Maneuvers Altitude Bank Angle > 500 ft 30 deg 200-500 ft 10 deg Relative Position Center Trailing Heading change 20 deg 40 deg 13
Approach Dependent Variables Aircraft separation during breakout Accuracy of breakout maneuver Subjective data: Workload Situation awareness Acceptability of procedures 14
Results Aircraft Separation During Breakout Three-way repeated measures analysis of variance Dependent variables: - Slant range between 1. leader / center aircraft 2. center / trailing aircraft Independent variables: - Cause of breakout - Location of breakout (low / high altitude) - Relative position of the simulator (center / trailing) 15
Results Aircraft Separation During Breakout 4500 leader/center 4500 center/trailing 4000 4000 3500 3500 3000 3000 2500 2500 2000 Breakout Breakout+15s Breakout+30s 2000 Breakout Breakout+15s Breakout+30s Slant range separation (ft) No single instance of slant range separation less than 2400 ft Breakout aircraft separation always well above the threshold value specified in an FAA study (Magyaritis & Ozmore, 2002) 16
Results Aircraft Separation During Breakout Leader / Center Separation 15s past breakout Mean aircraft separation (ft) 3400 3300 3200 3100 3000 2900 2800 2700 2600 2500 Breakout cause F=89.87 df=1,7 p<0.0001 Aircraft blunder Wake Breakout location F=20.45 df=1,7 p<0.01 Higher altitude Lower altitude Pilot participants indicated that the uncertainty of wake characteristics prompted faster responses Higher altitude breakout reflect different post breakout geometries between aircraft 17
Results Aircraft Separation During Breakout Center / Trailing Separation 15s past breakout Mean aircraft separation (ft) 3400 3300 3200 3100 3000 2900 2800 2700 2600 2500 Breakout Location F=44.73 df=1,7 p<0.001 Higher altitude Lower altitude Sim ulator Position F=40.39 df=1,7 p<0.001 Center Trailing Pilots indicated that the center aircraft needs to maintain separation with two other aircraft 18
Results Pilot Workload: NASA Task Load Index 7 6 nominal breakout Workload 1=low 7=high 5 4 3 2 1 mental physical * temporal * performance * effort * frustration * composite * p<=0.05 Higher overall workload in breakout runs Off-nominal conditions require the safe maneuvering of the aircraft following breakout, rather than implementing normal approach procedures 19
Results Pilot Situation Awareness Dependent variables from Situation Awareness Rating Technique (SART) - Ten subscales combined Into 3 broader categories Demands of the situation - instability, variability, complexity Personal resources - alertness, spare mental capacity,concentration,division of attention Situation provision - information quantity, information quality, familiarity 20
Pilot Situation Awareness (SART) Results 26 nominal breakout 21 level of situation awareness 16 11 6 1 * situation demands (range= 3 to21) personal resources (range = 4 to 28) situation provision (range = 3 to 21) Higher overall situation demands in breakout runs (F=25.46, df=2,6, p<=0.01) Situation instability, variability, complexity: consistent with workload results Non-significant results for situation provision indicate that information and procedures provided was adequate for both nominal and off-nominal cases 21
Results Pilot Discussion / Feedback Feedback on Procedures Some maneuvers were aggressive Trust in automation could be an issue Automate breakout maneuvers User Interface Feedback Pilots wanted the system to automatically set the field of view of the navigation displays so that all three aircraft were always visible Information provided via the displays was considered simple, clear and unambiguous 22
Conclusions Summary Feasibility of breakout procedures was tested using three very closely spaced parallel runways Aircraft separation was maintained well above 2400 ft during breakout across all conditions Similar levels of situation awareness between nominal and breakout runs indicate adequacy of information and procedures While workload was higher in breakout flights, it still remained manageable across all conditions 23
Summary Future Work Auto pilot versus manual breakout procedures Evaluation of the role of air traffic control Evaluation of post-breakout maneuvers back into the arrival flow Evaluation of the integration of our concept with terminal and surface operations 24
Savita.A.Verma@nasa.gov Thomas.E.Kozon@nasa.gov Photo courtesy of the FAA 25