 The common root of almost all accidents is human error. By the operator, the maintenance technician, the manufacturer, or the designer. Somewhere along the line, someone did not do his job right, and the error was never caught. Human errors occur in three broad areas. In communications, when man's interpersonal relations with his environment, with other men and machines, break down. Accidents also occur when the demands of a situation exceed man's physical, physiological, and psychological limits. Or when the integrating function, which is supervision, fails. The key to the entire process of decreasing the demands upon the human capability is not only a clear awareness and understanding of the mission, but also controlled supervision at all levels of selection, training, and use. The supervisor has a unique role to play in safety and a unique responsibility. There is a good reason why 95% of the recommendations in accidents are directed toward the supervisor. In the final analysis, no matter how good the equipment or how reliable, no matter how well selected and trained the man, no matter how well aware they may be of their own limitations, no matter how clear-cut the objective of the mission, and no matter how well this objective may be communicated, unless they are all welded together, accidents happen. Supervision is this welding together of man, machine, and mission, this integrating effort that gives strength to the whole structure. It is the tool man can best employ to combat human error. And you are a supervisor, whether you are an executive officer with authority ranging over a great and complex organization or a technician cracking the whip on the line during an operation or even when you work alone, for perhaps the most vital facet of supervision is self-supervision. The same basic principles apply at every step of an operation and at every level of command. A supervisor must first know the mission, the plan and its goals, the importance of success, and the results of failure. In every instance, he should know as much about the plan as is required, his place in it, and the place of every machine and of every man working for him. Then he must be sure that the details of the plan can be communicated properly, that the channels up and down the chain of command are functioning well. He must have some system for monitoring the mission, for getting feedback, and for following through himself to be sure the job is getting done safely and well. Once he has the program or mission firmly in mind, he must be able to assign the right task to the right man. To do this, he must know his own strengths and weaknesses, the strengths and weaknesses of his machines, and perhaps most important of all, he must know his man, what they can do and what they can't. He must be sure when he makes the assignments that each man has the authority, commensurate with his responsibility, and accepts these limits, so there will be no passing of the buck. A man can delegate authority, but he can never abdicate his responsibility. A supervisor must back up his man with effective supporting procedures for the maintenance and operation of the machines with technical orders, checklists, and in special areas with special systems like nuclear weapons with access control and with challenge and response checks. Here it must be absolutely clear that there is no margin for error. In every system, he must always remember that he is dealing with men as well as machines. And men have basic needs for attention to be a part of the program, to be able to make suggestions and have them evaluated and accepted when they are valid. He must be able to see the job from their point of view, what problems they may have. Finally, he must be certain that he provides the security of a safe working environment. When all this is accomplished and the plan is in operation, he still must be prepared for change. The original plan should be all inclusive but flexible, changing times constantly require new or modified equipment, new skills, and possibly different people. And he must ensure that as changes are introduced, his people are prepared for them with new information. He sees to it that every echelon is aware of and enforces these new procedures. But he must not expect a change to be accepted simply because it is for the better. He must recognize that any change is associated with emotional impact for the individuals involved. He must help resolve these emotional disturbances by providing a sympathetic ear and, where possible, an opportunity for more training. But he must never allow emotional problems involved in a change to compromise the mission. Finally, he must never allow his own supervision to become an end in itself. It must always be the means to accomplish the mission. His men have the knowledge to get the job done. He has the task of getting them to use that knowledge to function at their best. If he succeeds, safety through good supervision will be integrated into every stage of the mission. If he fails, catastrophe. This is the wreckage of a T-33 at Rattlesnake Gulch. It tells a sad story of a special kind of supervisory failure. The wreckage was spread over an area a mile wide and three miles long. The experts from the directorate of aerospace safety diagrammed the location of the parts and gathered the bits and pieces together and sent them back to the base. There began the tedious process of putting the parts into some semblance of the original plane and the minute inspection of every piece to try to determine the cause of the accident. As a result of this accident, one of the pilots died. The other pilot survived. This is what he had to say. Well, it was a real nice spring day. Clear, balmy, but not hot. The mission had been laid on the day before. After a local clearance was filed, we began pre-flighting the aircraft. I'd never met the other pilot until that morning. We were both highly qualified. He had 2,200 hours, 1,400 in a single-engine jet aircraft. I remember him saying, you take the bottom and I'll take the top. We took off at 0920, and I was flying the aircraft from the rear seat. I made a standard VFR departure. At 16,000 feet, making about 330 knots, I reduced throttle to 90% and started a gentle turn to the left. I believed that I'd turned through about 30 degrees and started to advance the throttle. Without warning, our aircraft suddenly and violently entered an uncontrollable role to the left. There was a short blank space in my memory, and I had the sensation I was upside down and being thrown violently against my shoulder harness. Then blindness and the wind striking my eyes. I released a seat belt, reached for the D-ring, and pulled it. The shoot opened immediately. I could see nothing. My entire field of vision was bright red. I landed in the brush but could determine nothing because of my blindness. Apparently, the day before the accident, maintenance personnel did not completely secure all of the airlock fasteners on the left forward portion of the door. This error also went unnoticed by maintenance personnel during the pre-flight on the morning of the accident, both the result of poor supervisory procedures. When the airspeed built up, the doors were subjected to distortion and bending, which in turn caused their entering the airstream at an angle that created aerodynamic loads powerful enough to overstress the aircraft, and it disintegrated. But the most important supervisory error was that of the pilot himself. He too missed the access doors. The most vital aspect of supervision is self-supervision. It's your life that is at stake, and there's no way around it. The greatest responsibility for it lies with you, on duty or off duty, whether you are behind the wheel of your car or handling a dangerous weapon. And if you happen to be the high-ranking man, other serious aspects of supervision enter the picture. Emotional immaturity, trying to be too nice a guy. I was in charge of the group. The airman was young and very fast in the draw, and he kept wanting to prove it. I didn't want to get nasty about it, so I let him show off a little, get it out of his system. First, he'd borrowed the other man's pistol and outdrew me. Then he borrowed mine and went after the other man. When she allows something like this to get started, there's not much chance of stopping it. But I finally put an end to it. I have to admit that around 2 a.m., things can get a little boring. Even the headlines seem pretty dull before I realized what was happening. Most men reflect the quality of their leader. A good supervisor doesn't have to be a good guy, but he must be emotionally stable and set a good example. A supervisor, however, is a man. And as such, he is subject to all the limitations of any other man. It was a clear Sunday morning. My engines were out saturating the runway for a new aircraft gear and brake test. Job was going along pretty good. 15 to 20 minutes would have seen it finished. The man in the number two engine was having little trouble seeing where he was going. I thought I'd better caution him to be sure not to overrun his marks. After that, I looked off and sure enough, there was George, my assistant, early as usual. I really appreciated that. I thought I'd let him take over and I'd head home to get the family up. George was glad to help. While I sat there explaining the details to him, I heard the crash. And there were the two engines in a head-on collision. I never made it home that morning. Following the prescribed procedures is always my thing. But in the field of nuclear safety, the yardstick is, has been, and must continue to be, zero accidents. There is no other acceptable statistic. Supervision must always be a minute by minute, second by second awareness. And supervision under unusual weather conditions calls for added alertness. This crew was unloading a nuclear weapon in accordance with authorized checklist procedures using the challenge and response system. Disconnect P101 from J1. Disconnected. Disconnect D-ring, mission MJ1. One of the items set forth in the checklist, disconnect D-ring, wasn't done. By not disconnecting the D-ring as the bomb was lowered, it actuated this S1 switch, canceling one of the many safety features on the bomb. Fortunately, the other safeties prevented an explosion. But in nuclear safety, a supervisor must consider every safety check as if it were the very last. It is not enough to follow procedures. A supervisor must also know his men. Apparently, before the flight, some repairs were needed on the brakes of this F-100. A young, inexperienced airman was at work. And he had scrambled the parts a little. And during installation, he put the discs and carriers back out of phase. The maintenance inspector was deeply involved in conference and allowed the work to continue without adequate supervision. Then, when the job was done, he accepted it without the required inspection, even though this man was inexperienced. A good supervisor not only knows his men, but he knows that some men need supervision more than others. The F-100 came in for a full stop landing. Touchdown was normal and after the nose gear touched down, the pilot lightly applied the brakes to test their operation. The first operation appeared normal. At approximately 80 knots, the pilot again applied the brakes. The left brake failed and the plane veered to the right. The metal-to-metal contact and the badly assembled brakes caused enough friction at the actuation cylinder head to wear through the metal. Hydraulic fluid leaked out of the cylinder and was ignited by the friction-generated heat. Fortunately, the pilot managed to stop the aircraft. Perhaps one of the most vital aspects of supervision is what is called follow-through. Five out of the seven crew members in this accident were fatally injured when power was lost on a go-around. For a long time, the accident investigators were puzzled. Then one man took a good look at the fuel sprainer. There was something about it that bothered him. So he thought he would discuss it with the others, working on the investigation. And sure enough, another man remembered something. This was the wrong fuel sprainer. They decided to check one of the planes on the line. They pulled the fuel sprainer up and examined it. This was different from the one in the wrecked plane. This one had a bypass filter. The other one didn't. The cause of the accident was then established as fuel system icing. It had blocked the fuel sprainer screen and in turn caused the flame out as a result of fuel starvation. But where did the old type sprainer come from? A check with maintenance supply provided the answer to the mystery. It was buried four years ago in a technical order change. A fuel sprainer without a bypass filter had been replaced with a new one that had a bypass filter. But someone failed to follow through on every level. The old ones had been removed from the aircraft. But they remained in supply. Four years later, they had made their insidious way back into the aircraft. Later, it was learned that 17 other aircraft also had the old filter reinstalled. One of the most important words in the safety language is survey. Through it, a supervisor can ensure that safety is integrated into his program at every stage. As this safety survey officer will readily testify. This story is almost incredible. A couple of years ago, we undertook a safety survey of an Air Force base. A survey may appear to be dull, tedious, supervisory business. It's true much of its routine. Checking to see the procedures already documented are in fact being executed correctly. The survey lasts a few days. Usually, there is no positive way of knowing that it's going to do any good at all. But no amount of effort can be considered too great. We were driving along checking the condition of the runway and discussing some of our findings when I noticed that just beyond the runway, a section of the land seemed to be extremely rough. So I suggested we have a look over there. Well, not only was the land dug up, but we also found there was a large open hole. It was pretty obvious that this land was thought to be leveled and the hole covered. This was only one of 1,000 routine checks we made on that survey. And we followed through. The land had been leveled and the hole covered. I happened to be in the tower when a B-58 came in for a landing. The first B-58s in operation. And the pilot was having a rough time with the brakes. In an anxious moment of decision, he decided to help slow the aircraft down by going off the side of the runway. This is the actual film taken when the plane went off the runway and its wheel passed directly over the covered hole. Supervision set up the survey program. Supervision located the problem. Supervision corrected the problem. Supervision prevented the accident. The mission is successful only when the integrating function of supervision welds together all facets of the man-machine environment. And whenever the demands of the situation are kept from exceeding man's physical, physiological, and psychological limits, accidents are not inevitable. They can be prevented.