Training for high reliability: COMPLEXITY and Tight Coupling in a helicopter fleet replacement squadron.

Submitted to

Western Academy of Management

2001 Annual Meeting

Elkhorn, Idaho

James C. Spee

University of Redlands

1200 E Colton Ave.

Redlands, CA 92374

909-748-6265

spee@uor.edu

Training for high reliability: COMPLEXITY and Tight Coupling in a helicopter fleet replacement squadron.

Abstract: Technological complexity and the tight coupling of events that can lead to catastrophe are two key elements which organizations must manage if they are to move from high levels of risk to high levels or reliability (Perrow, 1984). While the literature of high reliability organizations (HROs) mentions the importance of training in maintaining a culture of high reliability, none of the previous studies have examined the training organizations that provide HROs skilled operators. This study examines a helicopter squadron that trains replacement pilots and air crews for duty on aircraft carriers. The study confirms that the training organization meets the definition of an HRO and shows how the entire staff of the squadron, not just the pilots and crews learn to deal with complexity and tight coupling.

Training for high reliability: THE helicopter fleet replacement squadron.

Submitted to the Western Academy of Management 2001 Annual Meeting.

The growing literature on high reliability organizations (HROs) has focused on the factors that contribute to their ability to achieve their missions while avoiding catastrophic errors (Rochlin, Roberts & LaPorte, 1987; Weick, 1987; Roberts, 1990a; Roberts, 1990b; LaPorte & Consolini, 1991; Schulman, 1993; Creed, Stout & Roberts, 1993; Roberts, Rousseau & La Porte, 1994; Roberts, Stout, & Halpern, 1994; Brierly & Spender, 1995; Klein, Bigley & Roberts, 1995; Grabowski & Roberts, 1997). Roberts (1994) lists several issues raised by existing research including:

• Whether organizations that manage potentially dangerous technologies can manage without breakdowns.
• How HROs are organized.
• What steps HROs take to mitigate error.

Research into these issues has focused on disasters (Perrow, 1984), organizational culture (Schulman, 1993), and decision-making (Roberts, Stout, & Halpern, 1994). Research on linkages between subsystems that make up a larger HRO have not received as much attention (Roberts, 1994).

Previous research also gives minimal attention to the task of training operators to operate in the high reliability environment. In most cases, the systems are so complex that most of the training must take place on the job (Weick, 1987).

During a workup period, the crew of an aircraft carrier learns to work together to launch and recover aircraft (Rochlin, et al, 1987). For the pilots and aircrews on the carrier, training begins long before they are assigned to the operational squadron. Pilots and crews move from primary training to an advanced training organization called a Fleet Replacement Squadron or FRS (These were formerly known as Replacement Air Groups or RAGs.) The FRS conducts advanced pilot and aircrew training that goes beyond the "basic instruction" suggested by Rochlin, et al (1987).

The fleet replacement squadron may have a characteristic unique among training organizations for the HROs that have been studied to date. An FRS operates the same aircraft (albeit an older model) that trainees will eventually operate when they join an operational squadron. Airplanes and helicopters are high-risk technologies that can risk the careers and lives of the trainees, other members of the squadron and a "significant number of outsiders" (Schulman, 1994, p. 34).

The existing literature gives only brief descriptions of the role of training in creating HROs. The Fleet Replacement Squadron follows a rigorous program of qualifications for pilots and crews to prepare them for work on an aircraft carrier. These include qualifications in a larger, more complex helicopter, night flying over land, night flying over water, search and rescue operations, sonar, and anti-submarine warfare.

Roberts (1990a) describes potential dysfunctionalities that could plague high-risk technologies as described by Perrow (1984) and Shrivastava (1986). In her study, she found that continuous training mitigated the potential for unexpected consequences due to complexity, the danger of complex technology, the dangers of poor training, the danger of poor motivation, and the dangers of defective, substandard or malfunctioning equipment. Constant training also keeps risky technologies such as jet fuel, munitions, and steam catapults separated from each other to minimize the risk of unanticipated interactions (Roberts, 1990b, 1990a). These characteristics of HROs will be examined in greater detail below.

Wieck (1987) warns that training can sometimes create accidents rather than prevent them. Training for high reliability organizations may rely on verbal abuse or other tough and demanding methods designed to weed out the faint of heart and the incompetent. Often, he notes, the training settings have only modest validity once trainees move to their actual jobs. Weick argues that when people are trained for high reliability, the initial reactions they learn are the most important because those are the ones most likely to reappear under pressure. Even successful training has drawbacks because errors seldom occur in any discernable pattern in the real system (Weick, 1987)

In this paper, we will argue that a helicopter training squadron (or Fleet Replacement Squadron, FRS) based in San Diego, California, meets the definition of an HRO set out in the literature. Data for this study was collected through interviews with 22 squadron personnel conducted in March 2000 by the research and an assistant. Subjects were chosen from various levels in the organization from the commander down to a recent recruit across all of the major departments. A goal of the study is to explain how training enhances the reliability of HROs such as aircraft carriers in spite of Weick’s warnings.

Roberts (1990) identifies high reliability organizations as the subset of hazardous organizations that enjoy a record of high safety over long periods of time.

At the time of the study, the squadron had flown over 34,000 flight hours without a Class A incident, one that involves damage of $1 million or more. The last Class A incident at the squadron took place in 1994, six years before the study took place. The squadron operates in the center of a major metropolitan area, San Diego, California. Although many of its flights are over the mountains, the desert, or the ocean, an accident anywhere near its home base could result in substantial loss of life to the base’s neighbors. This would be especially true if the aircraft was carrying weapons or a heavy load of fuel.

The squadron’s record is less surprising in light of overall Navy and Marine aviation safety statistics, shown in Figure 1. Replacement Air Groups (RAGs) became operational in the mid-1950s. They are cited in the figure as a contributor to the decrease in the mishap rate to 1.28 incidents per 100,000 hours in 1996.

(Insert Figure 1 here)

Incident rates for helicopters have not matched the overall rate for the Navy. Between 1986 and 1997, incident rates for Navy helicopters varied between 1.28 incidents per 100,000 hours in 1995 and 2.97 incidents per 100,000 hours in 1990. The linear trend line shows a decrease for helicopters as a whole.

(Insert Figure 2 here)

The incident rate for the H-60 helicopter is even less predicable than that of the helicopter fleet as a whole. Although it reported no incidents in 1987 or 1992, the Navy H-60 helicopter fleet had 4.47 incidents per 100,000 hours in 1997 and 4.08 incidents per 100,000 hours in 1990, more than in the Gulf War year, 1991. These results are not totally unexpected since large aggregates such as the overall Navy figures are made up of multiple observations, all varying around the mean. One year, a particular aircraft may end up well above the overall average while the next year it ends up well below. The trend line for the HS-60 shows an increase for the ten-year period, mainly because of poor performance in the last year shown on the chart. Based on this data, the FRS in this study appears to be more reliable than other Navy H-60 squadrons.

Class A incidents are always followed by investigations to determine what caused the incidents. Figure 3 depicts the incident rates for 1987 to 1997 by cause for all Navy helicopters and for Navy H-60s. The overall incident rate for the ten-year period was 2.25 per 100,000 hours for all Navy helicopters. The rate for H-60s was 1.93 per 100,000 hours.

(Insert Figure 3 here)

Both for the helicopter fleet as a whole and for the H-60 platform, the rate of incidents caused by human error was the highest, followed by the rate for incidents caused by supervisory personnel, and the rate for incidents caused by aircrew factors. As Perrow (1984) points out, these are typical reasons given for multiple failure accidents. Perrow argues that the potential for accidents and incidents is built into high-risk systems so what appears to be operator error may in fact be systemic in nature. Roberts (1994) goes even further, arguing that accidents are embedded in systems of organizations not in single organizations or individuals.

To determine whether the helicopter fleet replacement squadron meets the definition of a high reliability organization the researcher analyzed two key characteristics of HROs: (1) complexity and (2) tight coupling (Perrow, 1984; Roberts, 1990a). High reliability organizations, therefore, are those that find ways to deal with the risks of complexity and tight coupling. Complexity means that many parts of the system interact at once. Tight coupling means that a failure in one part of the system can quickly lead to other failures.

A helicopter is both complex and tight coupled. Its rotor systems have multiple linkages and hinges that must function properly together or the pilot will lose control. The entire rotor head is suspended from a mast and held on by a single large nut. A second rotor system, usually smaller and in the rear of the helicopter, is required to keep the helicopter on a straight heading when power is applied to the main rotor. Extreme control movements during hover or at other critical times can result in dynamic instability and rapid destruction of the aircraft. In spite of the complexity and tight coupling, pilots are able to fly helicopters. They are only one part of the system. The question of interest in the study reported in this paper is how to create a reliable training organization that teaches people to operate within a high reliability organization The sections below describe the squadron in terms of how it manages complexity and tight coupling.

Complexity often results in unexpected sequences of events (Perrow, 1984). For the FRS, unexpected events are most likely to occur while pilot and aircrew trainees are in the air with their instructors, but catastrophic events can also occur on the ground if the maintenance crews are not vigilant. Elements of complexity include:

• Complex technologies
• Subsystems serving incompatible functions
• Indirect information sources
• Baffling interactions (Roberts, 1990a)

Complex technologies are one of the defining features of most High Reliability Organizations. Roberts (1990a) notes that HROs do not necessarily use state-of-the-art hardware but they are sufficiently technologically advanced that errors can have far-reaching negative consequences (p. 106). They often use older hardware for which reliability is more certain. They phase in new technology gradually, keeping old technology as a backup until its reliability is well known.

Fleet replacement squadrons operate the same complex aircraft as operational squadrons, although they may operate earlier models because operational squadrons receive new equipment first. The squadron in this study operated the Sikorsky H-60F SEAHAWK helicopter equipped for anti-submarine warfare. According to the manufacturer, the H-60F has a takeoff mission gross weight of 21,800 pounds, a mission endurance 4.20 hours, and a dash speed of 133 knots. It carries two Mark 50 torpedoes and a crew of four. Powered by twin 1,700 shafthorsepower T700-GE-401C turboshafts, the SEAHAWK can operate from aircraft carriers, cruisers, destroyers, and frigates. The aircraft has radar and sonar systems that allow it to collect information from beyond the horizon. (www.sikorsky.com/programs/seahawk/index.html)

(Insert Figure 4 here)

The H-60 has a single four-bladed main rotor and a four-bladed tail rotor. With many more moving components than the equivalent airplane, helicopters are especially susceptible to failures due to vibration and metal fatigue. Radar, sonar and electronic warfare equipment add to the complexity. The technology is not all new, however. The H-60 helicopter was designed in the early 1970s and has been in service for over 20 years, which gives the squadron a strong base of existing knowledge on which to base its operations. The H-60F fits Roberts (1990) definition of technologically advanced but not state-of-the-art. The reliability limits for the H-60’s major components are well known and stated in Navy operating manuals. The personal operating limits for student pilots are less well known. In a way, they represent "new technology" whose reliability is unknown at the start of training. Instructors, then, serve as a backup, based on the expectation that their experience as pilots increases their reliability. Training is both a way of dealing with the complexity and a source of complexity in the FRS.

Like the aircraft carriers where its pilot and aircrew trainees will serve after completing their training, the squadron deals with complex technology by pushing responsibility and accountability to very low level employees (Roberts, 1990a). At this squadron, trainees are not just pilots and aircrews. Recent enlisted recruits who join the squadron are often assigned to the flight line to train as "Plane Captains." A plane captain conducts a pre-flight inspection on every aircraft scheduled to fly before the pilots come near it. If the plane captain finds anything that is a no-go item, the helicopter is grounded. This emphasizes the importance of spotting problems on the ground before they become emergencies in the air. Other inspections take place before and after the plane captains do theirs. Higher-ranking personnel conduct all of the other inspections. This builds in redundancy, and trains the plane captains for future success if they transfer to the carrier environment because they learn that their inspections hold as much weight as anyone else’s does even though they are low in rank.

Another method of dealing with complex technology is through job differentiation (Lawrence and Lorsch, 1967). Roberts (1990a) notes that one characteristic of HROs is that they operate advanced technology requiring specialist understanding. The squadron divides responsibilities up into manageable chunks so that technicians can develop expertise in one area without needing to know how everything else works. For example, the no one else may know what the ordnance technicians do to maintain the torpedoes on the helicopter. All the others need to know is that the equipment is ready and safe. At the same time, an ordnance expert does not deal with corrosion on the fuselage of the helicopter, because that responsibility lies with the Aircraft Structures technicians. The ordnance technician is responsible for halting work if conditions make it unsafe to work on weapons, such as weather that could include the possibility of lightning strikes during a thunderstorm. The delegation of authority to the lowest parts of the organization is also differentiated by different kinds of expertise. An expert in one field may halt unsafe operations even when other parts of the organization decide it is safe to operate from their knowledge base.

The reinforcement of authority to stop unsafe operations must come from the leader of the squadron. The commander must not overemphasize the mission because that might push lower level personnel to operate outside the threshold of safety.

HROs must keep dangerous functions from interacting and causing accidents (Roberts, 1990a). On an aircraft carrier, for example, fueling and munitions loading must be kept separate to minimize the risk of on board fires. In a nuclear plant, power generating systems can interact unfavorably with the control systems for the reactor causing unforeseen system failures (Perrow, 1984).

The FRS has some of the same incompatible subsystems as the aircraft carrier, including munitions and fuel. Mechanics must be aware of dangerous interactions even from safety devices. One onboard electronic system has an explosive charge that must be disabled before it can be serviced. Cleaning solvents, paints, and fuel are all flammable and must be kept away from sparks or open flames.

In the squadron, incompatible tasks such as munitions handling and refueling are conducted by separate parts of the organization, all of which receive constant training on their tasks. The difference between the land-based training squadron and the carrier-based operation is the amount of space available to keep activities separate. The training squadron has plenty of room as long as everyone stays alert. Unlike the carrier, however, the training squadron must operate year round at a high level of activity. It does not distinguish between "work up" and operations. The squadron must operate at the same level of safety and awareness year round, although it is not as high a level of activity as a carrier squadron experiences once it is on station.

At the FRS, the physical layout of the shops helps to maintain distance between various operations. The squadron hanger and apron area has designated spots for various activities such as fueling, washing and loading. As with air traffic controllers (Roberts, 1990a), pilots and maintenance staff are trained to use specialized language that reduces ambiguity in interactions. Part of the training process is learning "platform specific knowledge" that is unique to the H-60 helicopter.

Indirect information sources are ways for information to flow around the organization other than the chain of command (Roberts, 1990a). Machinery using advanced technology nearly always requires instruments, lights, and signals that communicate indirectly and must be scanned to determine if unexpected patterns are emerging. This increases risk because no single, direct source may be pointing out a dangerous situation, even though multiple indirect sources are.

The helicopter’s instrument panel is a major source of indirect information about the operation of the helicopter. Only the pilot and instructor have access to that information while in flight. Maintenance staff use specialized instruments for tuning and testing equipment on the helicopter as part of their repair procedures.

At the squadron level, information flows mainly concern the flying schedule and maintenance activities. Pilots speak to the control tower using their communication radios. These are the same radios they will use when they get to the carrier. Pilots use their navigation equipment and radar to keep track of their location. Department heads email each other and use face to face communication whenever necessary. Computer tracking systems allow supervisors to check on the status of work underway. The entire maintenance system is built around quality checks after every procedure.

One way that HROs deal with indirect information sources is by increasing direct information linkages to many parts of the organization.

One example of a direct information source is the "squawk book" that the squadron maintains on each helicopter. Any technician, plane captain, or pilot can look in the book and see what problems the aircraft has had and what was done to correct them. They can look for recurring problems or issues that have occurred across several aircraft. As work is completed on "squawks" technicians and quality inspectors file reports that flow to decision centers, such as the maintenance officer, the head quality control, and especially, the schedulers who can now plan to use that aircraft in the daily schedule. If another problem arises, the schedulers need to know that the aircraft is no longer available and make alternate plans.

One of the most dangerous aspect of complex technology is the baffling interactions between various subsystems (Perrow, 1984). When a failure in one subsystem can cause unexpected failures in other subsystems, operators will have trouble seeing the meaning of events that up until that time were considered independent (Weick, 1987).

Helicopter flight is made up of a continuous series of baffling interactions that must be mastered to become safe in the air. Not all of the interactions can be duplicated in training. Trainees will not experience some until they reach the carrier. In the maintenance departments interactions between electronics, weapons systems, petrochemicals, and fast moving rotor blades may be hard to predict. A problem may appear to be fixed under ground testing but recur in the air. A quick, unreported fix on one component could cause other components to fail unexpectedly and unpredictably.

Roberts (1990a) suggests that training to understand the complexities of the technology is one way reliable organizations manage interactions between hardware and humans. Pilot trainees arrive at the squadron with basic training in airplanes and helicopters, but they are still relative novices. Their instructors have completed a tour of duty with an operational squadron. The instructors job is to help the trainees comprehend the systems and subsystems of the H-60 which is about seven times heavier and twice as complex as the helicopter used in their initial training. Instructors drill student pilots on the recommendations made by the Navy manuals, and how to make decisions when the situation does not match anything in the manual.

Newcomers in the maintenance departments must learn to diagnose complex systems based on limited information and seemingly unrelated facts. The researchers originally assumed that the only mission of the squadron was to train pilots and aircrews. Several interview subjects pointed out that the mission of the squadron was to train everyone regardless of the job he or she does. Nearly every task in the squadron involves complex interactions that may not be clear to a newcomer. Structure mechanics must understand the interactions between salt, water, and aluminum. Ordnance technicians must know the interactions between weapons systems and other components of the aircraft. Electronics technicians must know how the avionics systems interact with controls and mechanical components. Engine and transmission specialists must know how their systems connect to others, and so on. Everyone must understand the connections between doing their job correctly and having a safe flight. Mechanics have a big advantage over pilots, they have more time to diagnose problems and discover possible interactions while still on the ground. Problems could arise if they repair one part of the system and fail to realize that another part of the system is also defective. Sometimes the interactions cannot be detected until the aircraft is powered up and ready for take-off or already airborne.

The squadron’s response to manage baffling interactions is to pair experienced people with less experienced ones. Pilot trainees always fly with instructors who have operational experience and have more knowledge about what to expect. Instructors are less likely to continue flight into dangerous conditions because they see the interactions sooner and are not baffled by them. In the same way, experienced technicians work alongside newcomers and can resolve uncertainty more quickly. Given the lower pressure environment of the FRS, technicians are not as pushed to solve problems as quickly as they would in wartime or when deployed. They can keep trying until they find a fix. The danger would come if they misunderstood the interactions between systems. This was more of an issue for electronic systems than for mechanical systems.

Another method of dealing with baffling interactions is through Operational Risk Management (ORM). Several of the squadron members interviewed mentioned the importance of ORM in maintaining high levels of safety. ORM involves brainstorming possible dangers before taking action and work to find ways to reduce the dangers. Using ORM, squadron members think through potential risks and unexpected interactions before an event to prevent unsafe operations. Throughout the squadron, our research found a strong culture of ORM, even though it wasn’t always called by that name. Pilots use ORM before flights to discuss possible dangers with their crew. Technicians think through difficult maintenance tasks and try to anticipate mistakes that could cause serious damage or harm to the team. They review the maintenance manuals reminding themselves of key steps that make the operation safe.

In order to maintain high levels of reliability, HROs must develop ways to manage tight coupling, in which processes are continuous and cannot be easily stopped or started. Roberts (1990a) notes that tightly coupled HROs exhibit high degrees of interdependence that require generalist understanding to see the system as a whole.

Elements of tight coupling include:

• Time dependent processes

• Invariant sequence of operations
• One way to reach goal
• Little slack (Perrow, 1984; Roberts, 1990a).

Time dependent processes are actions which, once started, must be completed within a limited time to maintain safe operation (Perrow, 1984). For example, once a carrier launches its aircraft, it must continue recovery until every aircraft is accounted for. Any delay in the recovery sequence will result in some aircraft running out of fuel and ditching (Roberts, 1990a, 1990b). After every landing, the crew must clear the flight deck and reset the arresting gear before the next aircraft can land.

Operations of the training squadron are less time dependent than those on an aircraft carrier. Trainees would like to finish their flight training within the scheduled time, but no catastrophes result if they finish a few days later than planned. At the squadron level of analysis time dependence appears to be quite low. Activity for the crew has peaks and valleys at take-off, conducting the mission and returning for landing, but the intensity is much more relaxed than on the carrier.

Redundancy is one way of dealing with time dependent processes (Roberts, 1990a). The helicopters carry two pilots, one instructor and one trainee so that redundancy is maintained. The H-60 helicopter has two engines and can usually return safely with one engine out. Communication radios are also duplicated, as are some instruments and navigation systems.

In the maintenance operation, a second pair of eyes inspects every task at least once after completion before the aircraft is signed off for the flight line. Quality assurance inspectors may make additional checks. As noted above, plane captains and pilots also make inspections. The result of this redundancy can be a reduction in the squadron’s capability to conduct the mission. The squadron has fifteen helicopters but some days it had only two or three ready for flight. The current commander raised the goal to six or seven ready aircraft per day. While performance was improving, some squadron members felt that reliability may have been decreasing because of the push for higher output.

When operations must be conducted in a particular sequence, risk of failure increases (Perrow, 1984; Roberts, 1990a, Roberts, 1990b). On the carrier, the landing sequence must be followed in the proper sequence or an aircraft may miss the arresting equipment or cause some other failure.

At the squadron level, the FRS has more latitude in the sequence of its operations that the aircraft carrier. The squadron has much more flexibility in scheduling than seagoing squadrons because it does not have to support the work of the rest of a carrier by being in the air whenever others are flying. The FRS can constantly update its operating schedule as needed because of changing weather conditions or aircraft availability. The training schedule has some flexibility, but some parts of the curriculum cannot be completed until trainees learn simpler tasks. In the shop, procedures are much less flexible. Maintenance manuals depict approved sequences for every maintenance operation and must be followed exactly to maintain safe operations. For pilots, operating manuals and checklists provide a consistent sequence for each key operation of the helicopter.

On the carrier, jobs are clearly defined, but the overall system is very flexible (Roberts, 1990b). The FRS has a similar structure to ensure that operating sequences are followed closely. Each job has a clear set of expectations. Each maintenance task is laid out in a technical manual. Inexperienced mechanics may take longer to complete tasks because they are unfamiliar with the assignment. Experienced workers can run through tasks in their heads while they skim the instructions and visualize each step and the dangers that it might entail. Maintenance schedulers have the option of choosing which area of the aircraft to work on first, depending on the availability of mechanics. Once a procedure is underway, however, the sequence should not be changed without formal approval. For example, hoisting the main rotor assembly atop the helicopter and then connecting the controls could be very dangerous if it is not done in a meticulous step by step manner.

For pilots, standard procedures for instrument flight, for rescue operations, for preflight inspections, for take off and landing are all included in their operating manuals. This still leaves an incredible range of latitude for operating that requires judgement that can only be learned through training and experience.

Coupling is closer when the system has only one path to complete its mission successfully. For example, aircraft carriers have defined a specific procedure for launching and retrieving aircraft based on decades of experience. Deviations from these procedures, such as not resetting the arresting gear for each type of aircraft landing, can result in serious risks (Roberts, 1990b).

For this factor, the assessment depends on what unit we choose to analyze. For pilots and crews, the training can be accomplished along many different paths, but once the flight begins, they must follow standard procedures to minimize the danger to themselves or others. This is especially true in poor weather or when flying in instrument conditions. For the maintenance staff, reliability is actually increased by this factor. Mechanics are expected to follow the procedures outlined in their maintenance manuals. Each procedure is outlined in a step-by-step fashion that helps to reduce uncertainty for less experienced technicians.

Training and socialization appear to be the keys to dealing with the risks of a single path. Throughout the squadron, the phrase "written in blood" was used to explain why procedures had to be followed precisely. The assumption was that every procedure had been tested and modified based on a previous accident that cost someone dearly. The best way to avoid a similar mishap was by following the procedures.

Close coupling also means that operators do not have much slack or margin for error to maintain a high level of safety. In a nuclear power plant, the margin for error can shrink rapidly in emergency situations as one system failure feeds into another (Perrow, 1984).

In the FRS, pilots and flight crews have the least slack when they are airborne. Flying at night over water wearing night vision goggles can be very disorienting. Add an emergency due to engine or instrument failure and what little slack remains can be quickly used up.

On the shop floor, things appear more relaxed, but this is not always the case. The squadron has a daily goal to have six helicopters ready to fly. Although it has a total of 15 aircraft on hand, the squadron has a lower staffing level than an operational squadron. It also has a lower priority for parts delivery, making it more difficult to get as many aircraft ready for flight. Maintenance staff was proud of their record, which was steadily improving at the time of the study. Pressure to complete work in a fixed time reduced slack for the maintenance staff.

Pilots and crews create slack by learning to follow standard procedures and to deal with the unexpected. Instructors have recently returned from a deployed squadron. They are chosen from among the top pilots on the carrier. They use their experience to instill in students a healthy respect for the dangers inherent in completing the mission. Instructors noted that new pilots are not always aware of how much trouble they could get in if they pursue a particular course of action. It is their job to help student pilots develop the kind of reasoning skills needed to maintain flight safety.

Maintenance staff manages the reduced slack through strong quality assurance procedures and through the kind of authority inversions mentioned above. Any mechanic or inspector can ground an aircraft if he or she finds a discrepancy. No one can sign off until they take responsibility for fixing the problem or determining that it is not critical to flight safety. "What are they going to do to me if I keep an aircraft on the ground when someone wants to fly?" the master chief in charge of maintenance told the researchers. "They can’t fire me. I’ve been the Navy for over 20 years. Besides, how would they reward me if I sign an aircraft off just to make the numbers? I can’t be promoted any higher. I have all the authority I need to do my job."

Up until now, the study of high reliability organizations has focused on HROs in action, ignoring the support systems that allow them to function. The purpose of this study was two-fold: first to show that the training squadron is also an HRO based on the way it deals with complexity and tight coupling; and second to investigate how it teaches everyone in the squadron, not just the pilots and aircrews, how to operate in the HRO environment.

Future research should investigate how the decision-making system works at the FRS and how people learn to work within a larger system of organizations on board an aircraft carrier. It should also compare this squadron with other Fleet Replacement Squadrons in the Navy to look for similarities and differences. In addition data from the interviews will be analyzed to look for sources of continuous improvement for the squadron.

Bierly, P.E., III; Spencer, J. -C. (1995). Culture and high reliability organizations: The case of the nuclear submarine. Journal of management, 21 (4) 639-657

Creed, W.E.D.; Stout, S.K.; Roberts, K.H. (1993). Organizational effectiveness as a theoretical foundation for research on reliability enhancing organizations. In K.H. Roberts, ed. (1993). New challenges to understanding organizations. New York: Macmillan, 55-74.

Grabowski, M. and Roberts, K.H. (1997) Risk mitigation in large-scale systems: Lessons from high reliability organizations. California Management Review, 39 (4), 152-162.

Grabowski, M. and Roberts, K.H. (1999). Risk mitigation in virtual organizations. Organization science, 10 (6) 704-721.

Klein, R. L.; Bigley, G.A.; & Roberts, K.H. (1995).Organizational culture in high reliability organizations: An extension. Human relations, 48 (7) 771-793.

Lawrence, P.R. and Lorsch, J.W. (1967) Organization and environment : Managing differentiation and integration. Soldiers Field, MA: Harvard Business School Press

LaPorte, T. & Consolini, P.M. (1991). Working in practice but not in theory: Theoretical challenges of "high reliability organizations." Journal of public administration research and theory, 1 (winter) 19-47.

Perrow, C. (1984). Normal accidents. New York: Basic Books.

Roberts, K.H. (1990a). Managing high reliability organizations. California management review, 32 (4) 101-114.

Roberts, K.H. (1990b). Some characteristics of one type of high reliability organization. Organization science, 1 (2) 160-176.

Roberts, K.H. (1994). Functional and dysfunctional organizational linkages. In C.L. Cooper & D.M. Rousseau, eds. (1994) Trends in organizational behavior, Volume 1. Chichester, West Sussex, England: Wiley, pp. 1-11.

Roberts, K.H.; Rousseau, D.M.; and La Porte, T.R. (1994). The culture of high reliability: Quantitative and qualitative assessment aboard nuclear-powered aircraft carriers.

Roberts, K.H.; Stout, S.K.; and Halpern, J.J. (1994). Decision dynamics in two high reliability military organizations. Management science, 40 (5) 614-624.

Rochlin, G.I.; La Porte, T.R.; and Roberts, K.H. (1987). The self-designing of high reliability organization: Aircraft carrier flight operations at sea. Naval War College review, 40 (4) 76-90.

Shrivastava, P. (1986) Bhopal: The anatomy of a crisis. Cambridge, MA: Ballinger.

Schulman, P.R. (1993). The analysis of high reliability organizations: A comparative framework. In K.H. Roberts, Ed. (1993). New challenges to understanding organizations. New York: Macmillan, 33-54.

Weick, K.E. (1987). Organizational culture as a source of high reliability. California management review, 29 (2) 112-127.

Figure 4 Sikorsky SH-60F. Source: http://www.sikorsky.com