In April 2010, after a base maintenance check at Exeter, England, a Bombardier DHC-8-102 was flown uneventfully to East Midlands Airport to be repainted. During the return ferry flight to Exeter, the right engine developed a significant oil leak and lost oil pressure, so the flight crew shut the engine down. Subsequently, the crew noticed the left engine also leaking oil, with a fluctuating oil pressure, so they diverted to Bristol, where they landed safely.
The oil leaks were traced to damaged O-ring seals within the oil cooler fittings on both engines. Both oil coolers had been removed and refitted during the maintenance at Exeter. It was probably during re-installation that the O-ring seals were damaged.1
During the investigation carried out by the U.K. Air Accidents Investigation Branch (AAIB), it was found that over the 17-week period leading up to the end of a C-check, one of the maintenance technicians working on the aircraft had worked an average of 57 hours per week, which was nine hours per week more than allowed by the European Working Time Directive (WTD). The investigation also found that during the 10 days prior to the aircraft’s arrival at Exeter, the same technician had averaged 15.7 work hours per day, resulting in the 11-hour post-shift rest entitlement in the WTD being significantly curtailed. The technician reported that he did not consider himself fatigued during this period. However, he also said that during the leak checks on the incident aircraft he felt tired and had a lot on his mind, trying to get the aircraft ready for its scheduled painting slot, although it was not an unusual level of tiredness.2
The AAIB said that a technician tasked to work a 10-day period, with just one day off in the middle, averaging 15.7 hours per day, is a safety concern, particularly if not monitored by the approved maintenance organization (AMO). The AAIB also noted that the AMO involved in this occurrence had no policy on the maximum hours for a technician to work in any 24-hour period and relied on every technician and manager, after undergoing human factors/performance training, to communicate to their supervisors the risk of fatigue. The AAIB said that individuals who have undergone this training probably will be very responsible and will request time off when they feel that they need it. However, for some individuals this may not be the case, particularly when they have a strong desire to complete the job they have started and when there is a financial incentive to work longer hours, the report said.3
According to the AAIB, there is also evidence from fatigue research that people are not very good at detecting their degradation in performance as they become fatigued. Therefore, the responsibility for managing fatigue should belong to the AMO and not just the individual. The AAIB, however, notes that in Europe Part 145 (Annex II to Commission Regulation [EC] No 2042/2003) states that the AMO needs to take human performance limitations into account when planning maintenance tasks and, although not specifically stated, this should include maintenance technician fatigue. However, the acceptable means of compliance (AMC) and guidance material (GM) to Part 145 currently do not explain how this should be accomplished.4
In response to a safety recommendation issued by the AAIB following this serious incident, EASA in early 2013 published a Notice of Proposed Amendment (NPA) to EU Regulation 2042/2003 on continuing airworthiness to add AMC and GM in Part 145 on how approved maintenance organizations should manage and monitor the risk of maintenance technician fatigue as part of their requirement to take human performance limitations into account.
Working Time Directive
“Unlike aircraft crewmembers whose duty time is regulated by ad hoc and specific regulation, aircraft maintenance engineers’ duty time is standardized and regulated by the generic and rather vague Council Directive 2003/88/EC (the WTD), which leaves the door open to many interpretations and exceptions,” says Marco Giovannoli, an aircraft systems engineer at Etihad Airways and a fatigue and safety specialist.
The directive contains many opportunities for derogation, or exception, which maintenance organizations can use to circumvent the limits. The directive states that derogations may be made from the rest periods in the case of activities involving the need for continuity of service or production, particularly in industries where work cannot be interrupted on technical grounds. Also, derogations may be made by collective workforce agreements.
Even without the derogations, the WTD has been interpreted by some to mean that the minimum daily rest period of 11 hours means providing 11 hours of rest after a working period that could be up to 24 hours. Therefore, maintenance organizations need to have clear fatigue management plans that monitor their staff working hours and working patterns to reduce the risk of fatigue-related maintenance errors. The published NPA (2013-01) is actually attempting to standardize across Europe the interpretation of the working time directives for aircraft maintenance operations.
With NPA 2013-01, EASA is filling a regulatory gap compared with other aviation regulatory environments such as Canada and the United States. According to the AAIB, Transport Canada has published two NPAs (2004-047 and 2004-049) which propose requirements for an AMO to manage fatigue-related risks through a safety management system (SMS).
To support these proposed regulations, Transport Canada has published guidelines for a fatigue risk management system (FRMS) that provides a method for quantifying fatigue risk on a numerical scale using knowledge of working hours and rest periods. The AAIB also found that the U.S. Federal Aviation Administration (FAA) has set up a maintenance fatigue working group to review the need for regulatory limits on working hours for maintenance technicians.5
Fatigue Risk Management
EASA’s new requirements for fatigue risk management are part of a larger rulemaking effort related to SMS in maintenance and continuing airworthiness management organizations. Approved maintenance organizations are going to be required by EASA to implement an FRMS.
An FRMS needs to be carefully planned. To some extent, maintenance organizations can draw from already produced research in relation to flight crews’ fatigue, including the “FRMS Implementation Guide for Operators” jointly released by the International Air Transport Association, the International Civil Aviation Organization and the International Federation of Air Line Pilots’ Associations in 2011.
However, there are inherent differences between aircraft maintenance technicians’ work, working environment and associated fatigue issues and those of flight crews. It becomes necessary for a maintenance organization planning an FRMS to properly consider the maintenance-specific fatigue issues.
Fatigue should be a joint concern of technicians, organizations and national aviation authorities (NAAs). Without proper and effective oversight by aviation authorities, maintenance organizations may not have a strong enough incentive to monitor and mitigate fatigue issues. In this situation, the NPA is going to play an important role, as it requires a rule-enforcing effort from European NAAs for fatigue management. More important, however, is going to be the actual cultural transformation within maintenance organizations themselves.
“Aircraft maintenance engineers should be encouraged to report to supervisors their degraded performance caused by fatigue, and they should not face veiled threats of repercussions from their managers,” says Giovannoli. “Nowadays if a pilot reports fatigue, [the reporting] is considered professional, conversely an engineer ‘calling in fatigued’ may be demonized.” For fatigue to be consistently reported, technicians need a better understanding of their fatigue performance limitations. “There is a strong subjective component concerning fatigue, and therefore it is important to clearly illustrate the various indicators and symptoms of fatigue and the possible methods of offsetting it, such as caffeine, which normally is the common counteracting method used by aircraft maintenance engineers,” says Giovannoli.
Technicians should be advised about correct sleep practices and good health. Recommended levels of physical activity should be regularly practiced and a balanced way of life should be followed. “Moreover, managers and supervisors should be properly trained to perceive fatigue symptoms and be empowered to take corrective actions, even if these could be perceived as detrimental to organizational productivity,” he said.
Vigilance decrement is a form of short-term fatigue to which aircraft maintenance technicians are highly susceptible, especially during inspection tasks. This form of fatigue should be particularly monitored by AMOs.
Alan Hobbs, in a report issued by the Australian Transport Safety Bureau, discussed vigilance issues. “During the Second World War it was found that after about 20 minutes at their posts, radar operators became much less likely to detect obvious targets,” he said. “This problem applies to many monitoring tasks where the search targets are relatively rare. Aircraft inspection, the checking of medical X-rays and quality control inspection in factories are areas where vigilance decrements may occur”6 (Figure 1).
“Vigilance decrement applies particularly to detection tasks where the person is required to passively monitor a situation that is boring and monotonous, such as inspecting large numbers of turbine blades. The limiting factor is the ability to keep attention on the task. For example, during the visual inspection of an aircraft, a maintenance worker may look directly at a defect, yet if their attention is occupied with other demands, the defect may not be recognised. In general, inspection tasks that involve variety and regular breaks are less likely to suffer from vigilance decrement.”7
Bio-Mathematical Fatigue Models
Several software models are available from vendors as practical tools for estimating work-related fatigue associated with shift workers’ duty schedules. Some of the models can be used with any duty schedule, in which hours of work (i.e., start/end times of work periods) are the sole input.
The main advantage of bio-mathematical fatigue models is that they allow the generation of qualitative and quantitative forecasts of human fatigue based on a set of equations.8
Giovannoli, however, recognizes such models’ intrinsic limitations, as “they do not consider the worker’s individual variables and the conditions around him, such as task difficulties, working conditions and perception of fatigue, which may drastically increase the individual’s actual level of fatigue.
“It is recommended to integrate a bio-mathematical model in a wider-ranging FRMS,” he said. “However, available models should be carefully considered and validated before implementation.” One example of guidance for the selection of bio-mathematical models suitable for fatigue risk management in aviation maintenance has been provided by the Australian Civil Aviation Safety Authority (CASA).
In its publication, Bio-Mathematical Fatigue Modelling in Civil Aviation Fatigue Risk Management, CASA identifies opportunities for integrating models into a holistic FRMS while developing management systems and a corporate culture that understand the uses and limitations of qualitative/quantitative model predictions, use their outputs with caution and in the context of other operational opportunities and constraints, and adopt complementary multi-layered strategies to proactively identify and manage fatigue risk.9
CASA also discusses key factors to consider when selecting and applying a bio-mathematical fatigue model. These include the type of data to be used as inputs, the physiological factors described by the model components, types of output predictions and their relevance to task risks or other outcome variables, data used for validation and their level of equivalence to the operational environment and subject population, and the interpretation of predictions for use in decision making. CASA states that all these factors must be considered, relative to the specific operational environment for their intended use.10
In the application guidance material, CASA provides an overview of six bio-mathematical models from commercial and academic organizations together with their related products and services. CASA also provides a feature comparison table and a discussion of features in the context of commercial aviation applications.
A software model that has been developed explicitly for flight crew fatigue monitoring should not be selected, as it may not allow for proper collection and elaboration of relevant information in a maintenance environment.
Shift Work and Maintenance Errors
According to a study by Dawson and Reid, “Recent research has shown that moderate sleep deprivation of the kind experienced by shift workers can produce effects very similar to those produced by alcohol. After 18 hours of being awake, mental and physical performance on many tasks is affected as though the person had a blood alcohol concentration of 0.05 percent. Boring tasks that require a person to detect a rare problem, like some inspection jobs, are most susceptible to fatigue effects.”11
Studies have shown, Hobbs said, that 24-hour circadian rhythms influence human error, with many aspects of human performance at particularly low levels in the early morning. Memory and reaction time are at their worst at around 0400, and the chance of error is increased. There appears to be an increased risk of maintenance errors on night shifts.12
Hobbs said, “It has been found that when maintenance technicians are experiencing sleepiness, they are at increased likelihood of errors involving failures to carry out intentions, such as memory lapses and perceptual errors.
Sleepiness, however, seems to be less likely to lead to mistakes of thinking such as procedural misunderstandings.”13
“Twelve-hour maintenance shifts are becoming increasingly common,” Hobbs said. “In some cases, a company’s move to 12-hour shifts is driven by employee preference rather than management pressure. When compared with eight-hour shifts, 12-hour shifts offer certain advantages, such as less commuting time over the course of a week, more days off, and the opportunity to complete more work in each shift, with fewer handovers of tasks between shifts. Although workers tend to be more fatigued at the end of a 12-hour shift than at the end of an eight-hour shift, they sometimes report fewer health problems and better sleep on a 12-hour shift pattern than when on an eight-hour pattern.14
“At present, there is no conclusive evidence to indicate that extending the duration of shifts from eight to 12 hours will increase the probability of accidents or injuries. Nevertheless, 12-hour shifts may not be appropriate in all cases. Whenever a change is being made to 12-hour shifts, it is essential to evaluate the effects of the change on worker well-being and work quality. Quite possibly, the most significant effects of 12-hour shifts would show themselves on the journey home rather than at work.”15
“Rotating-shift workers may never entirely become accustomed to a work schedule, because the timing of shifts is constantly changing,” Giovannoli said. “Direction and speed of rotation can affect the adaptation to rotating shifts.”
Direction of rotation means the order of changes to shifts. Two common types of rotation are forward, when rotation is day to evening to night; and backward, when rotation is day to night to evening. Speed of rotation means the number of consecutive working days before a shift change.
The current fatigue science literature reports that a rapid (two to five days) forward-rotating shift system reduces mental and physical performance degradation and enables faster recovery of both sleep and social activities. In contrast, the fast backward-rotating shift system is linked with reduced physical and psychological health and higher fatigue.16
“Night shift should not last more than 10 hours including overtime and two consecutive night shift blocks in a row should not be allowed; with not more than eight night shifts in every month,” says Giovannoli. “Furthermore, night shift should not be extended beyond 0800 and after a night shift block, two days off should be granted, to provide enough rest time before the next shift, which in the ‘forward’ rotation is a morning shift.”
The ideal of a totally non-fatigued technician may not be achievable in the actual workplace, yet avoiding adverse outcomes of fatigue on safety and productivity should be the objective while implementing an FRMS. Empowering fatigue reporting, using bio-mathematical fatigue models and properly managing work shifts are valuable strategies for targeted fatigue risk management in maintenance organizations.
Mario Pierobon works in business development and project support at Great Circle Services in Lucerne, Switzerland.
- U.K. Air Accidents Investigation Branch (AAIB), AAIB Bulletin 6/2011.
- Hobbs, Alan. “An Overview of Human Factors in Aviation Maintenance.” Australian Transport Safety Bureau, AR 2008 055. December 2008.
- Australian Civil Aviation Safety Authority (CASA). “Bio-Mathematical Fatigue Modelling in Civil Aviation Fatigue Risk Management — Application Guidance.” 2010.
- Dawson, D.; Reid, K. (1997). “Equating the performance impairment associated with sustained wakefulness and alcohol intoxication.” Journal of the Centre for Sleep Research, 2, 1-8. Cited in Hobbs (2008).
- Hobbs, Alan. “An Overview of Human Factors in Aviation Maintenance.” ATSB Transport Safety Report — Aviation Research and Analysis Report AR 2008 055, December 2008.
- Hobbs, A.; Williamson, A. (2003). “Associations between errors and contributing factors in aircraft maintenance.” Human Factors, 45, 186-201. Cited in Hobbs (2008).
- Hobbs, Alan. “An Overview of Human Factors in Aviation Maintenance.” ATSB Transport Safety Report — Aviation Research and Analysis Report AR 2008 055, December 2008.
- Vangelova, Katia. “The Effect of Shift Rotation on Variations of Cortisol, Fatigue and Sleep in Sound Engineers.” Industrial Health, 46, 490–493 (2008); cited in Giovannoli, Marco, “Fatigue Monitoring to Improve Productivity and Safety in Aviation Maintenance.” City University London (2008).