Peter W. Merlin, Gregg A. Bendrick, and Dwight A. Holland Library of Congress Cataloging-in-Publication Data Merlin, Peter W., 1964- Breaking the mishap chain : human factors lessons learned from aerospace accidents and incidents in research, fight test, and development / Peter W. Merlin, Gregg A. Bendrick, Dwight A. Holland. p. cm. “NASA SP-2011-594”--T.p. verso. Includes bibliographical references and index. ISBN 978-0-16-090414-1 1. Aircraft accidents--Human factors--Case studies. 2. Research aircraft-- Accidents--Case studies. 3. Space vehicle accidents--Case studies. 4. Research aircraft--Design and construction--Case studies. 5. Space vehicles--Design and construction--Case studies. 6. Aeronautics--Research--Case studies. 7. Flight testing--Case studies. I. Bendrick, Gregg A. II. Holland, Dwight A. III. Title. TL553.5.M39 2012 363.12’414--dc23 2012008429 I SBN 978-0-16-090414-1 F o srale b ythe S uperintendent of Document U .s SG,.overnment Printi n Ogffice Internet: bookstore .gp oP. hgoo nve: toll free ( 86 65)12-1800; D C are a( 20 25)12-1800 9 0 0 0 0 Fax: ( 20 25)12-210 M4 ail: Sto IpDCC W,ashington D,C 20402-0001 ISBN 978-0-16-090414-1 9 7 8 0 1 6 0 9 0 4 1 4 1 I SBN 978-0-16-090414-1 For slea b yhte u Sperintendent oDf ocuments, U. oSv. eGrnment Printing fOfice Internet: bookstore.gpo.gov : Ptohlol nfreee (866) 512-1800 ; C Dare a2(02) 25-11800 9 0 0 0 0 Fax: (202) 512-2104 il:M Satop DICC, Washington, DC 20402-0001 ISBN 978-0-16-090414-1 9 7 8 0 1 6 0 9 0 4 1 4 1 Peter W. Merlin is an aerospace historian with extensive knowledge of the various factors involved in aerospace mishaps. Under contract to the National Aeronautics and Space Administration (NASA) at Dryden Flight Research Center, Edwards, CA, since 1997, he has authored a variety of books, including several NASA Special Publications on aeronautical research projects. He served as coauthor of research pilot Donald Mallick’s autobiography, Te Smell of Kerosene: A Test Pilot’s Odyssey, and X-Plane Crashes: Exploring Experimental, Rocket Plane, and Spycraft Incidents, Accidents and Crash Sites, with Tony Moore. He has also authored several technical papers for the American Institute of Aeronautics and Astronautics, as well as numerous journal articles on aerospace history and technology. In addition, he serves as contributing editor for historical publications at Dryden and has appeared in more than a dozen docu- mentary television programs for the Discovery Channel, the History Channel, National Geographic Channel, and others. He holds a bachelor of science degree in aviation management from Embry-Riddle Aeronautical University. Gregg A. Bendrick, M.S., M.D., M.P.H., is the chief medical ofcer at NASA Dryden Flight Research Center. He previously served 9 years in the U.S. Air Force as a fight surgeon before join- ing the Ochsner Medical Center in New Orleans, LA, where he practiced occupational medicine for nearly 3 years. At Dryden he oversees all aspects of aerospace medicine, occupational medicine, and ftness center operations. Additionally, he is the medical review ofcer for workplace drug test- ing and served as onsite medical coordinator for Space Shuttle landings at Edwards Air Force Base (AFB). He is board-certifed in aerospace medicine and designated by the Federal Aviation Administration (FAA) as a senior aviation medical examiner. He is a fellow of the Aerospace iii Breaking the Mishap Chain Medical Association, an instructor in the University of Southern California’s Aviation Safety and Security Program, and an afliate faculty member with Embry-Riddle Aeronautical University. He has authored several technical papers on various aspects of aerospace medicine as well as a novel. Dwight A. Holland, M.S., M.S.E., Ph.D., M.D., is a principal with Human Factors Associates, a fellow in the Aerospace Medical Association, and an academician in the International Academy of Aviation and Space Medicine. He has served as the president of the International Association of Military Flight Surgeon Pilots and of the Space Medicine Association. In 2003, he served as the technical cochair of the largest International Systems Engineering Conference held to date. He has over 100 academic presentations, abstracts, book chapters, journal special editions, and papers to his credit. As a U.S. Air Force Reserve ofcer, he has taught and codeveloped the Human Factors in Flight Test courses at the Air Force and Navy test pilot schools. He holds com- mercial and jet ratings with over 2,000 hours’ fying time in more than 40 civil and military aircraft. He also participated as a geophysicist on a remote feld expedition to the Antarctic, later writing a study comparing the human factors issues of polar exploration to those of long-duration space fight operations. iv Table of Contents Acknowledgments vii Introduction ix Part 1: Design Factors Chapter 1: “It May Not Be Hooked Up”: Automation Bias, Poor Communication, and Crew Resource Management Factors in the X-31 Mishap ............................................................................ 3 Chapter 2: Habit Pattern Transfer During the First Flight of the M2-F2 Lifting Body .............................................................................. 23 Chapter 3: Pilot-Induced Oscillation During Space Shuttle Approach and Landing Tests .......................................................................... 33 Part 2: Physiological Factors Chapter 4: Screening Versus Design: The X-15 Reentry Mishap ................. 55 Chapter 5: Six Million Dollar Man: The M2-F2 Task Saturation Mishap ........ 81 Chapter 6: Almost-Loss of Consciousness in the F-22A Raptor .................. 101 Part 3: Organizational Factors Chapter 7: Decision Chain Leading to the XB-70/F-104 Midair Collision ............................................................................................... 127 Chapter 8: Mission Management and Cockpit Resource Management in the B-1A Mishap .................................................................. 143 Chapter 9: Collision in Space: Human Factors in the Mir-Progress Mishap ...................................................................................... 159 Conclusions 189 Bibliography 191 Index 207 v Te authors would like to thank the many people who helped make this book possible. First of all, thanks to Tony Springer, NASA Aeronautics Research Mission Directorate, for sponsoring this project. We are grateful for the eforts of many people at NASA Dryden Flight Research Center including, but not limited to, Dr. Christian Gelzer, Tom Tschida, Karl Bender, and espe- cially to Sarah Merlin for copyediting the fnal manuscript. Special thanks to Dr. Tomas Hofmann and Dr. Philip J. Scarpa at the John F. Kennedy Space Center, Dr. William J. Tarver at the Lyndon B. Johnson Space Center, and Dr. James W. Butler, who reviewed our material for technical accuracy. Tanks to the staf of the Air Force Flight Test Center, particularly Dr. Craig Luther, for providing valuable source material and images. Apologies to anyone we missed. Any factual errors are the authors’ responsibility. We made an attempt in good faith to get the facts straight by using the best available source material. vii The popular image of test pilots as cowboys is misleading. Actual research pilots and test pilots like Joe Walker, seen here exiting the cockpit of the X-1A, were and are highly trained, educated professionals. (NASA) viii When a team of military and civilian researchers conquered the sound barrier with the rocket-powered Bell X-1 piloted by Capt. Charles E. “Chuck” Yeager in October 1947, fight performance was considered to be primarily a func- tion of the airframe/powerplant confguration, fight controls, and pilot skill. Design focus in the 1940s and 1950s, sometimes called the golden age of fight test, concentrated on vehicle confguration; aerodynamic control problems; and performance during low-speed, transonic, and supersonic fight. Human- machine interaction was not considered a key issue, except for the narrow 1 dimension of fying handling qualities. Within the fight-test community, a good pilot needed not only fying skills, but also engineering knowledge. An optimal pilot embodied a combination of the requisite skills, training, courage, and experience (i.e., the “right stuf,” 2 as described by author Tom Wolfe in his brilliant book of the same name). Such pilots few dozens of experimental craft, known collectively as X-planes, to evaluate a wide variety of cutting-edge confgurations and capabilities. Tey few in regimes of fight where new rules had to be written for aerodynamics, 3 propulsion, navigation, and thermal efects. Aviators were frequently exposed to extreme and unpredictable fight conditions that provided both physiologi- cal and cognitive challenges. In the early years of aviation, mishaps were often unfairly attributed to “pilot error,” a catchall term too often used to describe a variety of issues related to human factors. As early as the 1920s, however, a trio of Army Air Corps ofcers began to think about the efects of the human neurovestibular system on spatial orienta- tion in fight. Bill Ocker, David Myers, and Carl Crane often went against the prevailing wisdom of the era, which was that instruments were not essential to fying safely in what are now termed “instrument fying conditions” (when visual references are obscured by darkness or weather). When Ocker and Crane 1. Richard P. Hallion, Test Pilots: The Frontiersmen of Flight (Garden City, NY: Doubleday, 1981); Peter W. Merlin and Tony Moore, X-Plane Crashes—Exploring Secret, Experimental, and Rocket Plane Crash Sites (North Branch, MN: Specialty Press, 2008). 2. Tom Wolfe, The Right Stuff (New York: Farrar, Straus and Giroux, 1979). 3. Dennis R. Jenkins, X-15: Extending the Frontiers of Flight (Washington, DC: NASA SP-2007-562, 2007). ix Breaking the Mishap Chain wrote the very frst book on instrument fight in 1932, Blind Flight Guidance, it became the basis for most aviation instrument-training programs of the day. Te book was quickly adopted by the civilian aviation community and was used 4 throughout the world but was only reluctantly accepted by the U.S. military. When James H. “Jimmy” Doolittle made the frst successful airplane fight without the use of outside visual references—using only fight instruments— in 1929, those involved in aviation began to recognize the signifcance of the interaction between a pilot and an aircraft’s controls and display systems. In the 1930s, as Sir Frederick Bartlett of Cambridge University began to research pilot error in a simulator known as the Cambridge Cockpit, there was an increasing, though limited, realization that the design of the hardware inter- face could afect human performance. During World War II, technological advances resulted in the production of faster, higher-fying, and more efcient aircraft. At the same time, researchers began to notice an increasing number of errors in human performance in the cockpit. As a scientifc approach to fight operations replaced earlier, more intuitive methods, psychologists such as Paul Fitts began to write about the birth of modern human factors engineering. By the 1950s and 1960s, it appeared as though aircraft designs were reach- ing the limit of human performance in terms of pilot workload and task satura- tion. Such aircraft as the F-4 Phantom Jimmy Doolittle piloted the frst airplane and B-52 Stratofortress featured classic fight made without outside visual refer- examples of steam gauge–type cockpits, ences—using only fight instruments—in with their confusing array of dials and 1929, demonstrating the signifcance of the switches. Pilots often found instrument interaction between a pilot and an aircraft’s controls and display systems. (U.S. Air Force faces too small to see clearly (espe- via Hill Air Force Base Museum) cially during violent maneuvers), or 4. Fred H. Previc and William R. Ercoline, eds., “Spatial Disorientation in Aviation,” in Progress in Astronautics and Aeronautics, vol. 203 (Reston, VA: American Institute of Astronautics and Aeronautics, 2004). x