Fatigue and workload in short and long-haul train driving

Abstract

BACKGROUND: Little has been investigated regarding the role of shift schedule on fatigue and workload among Iranian train drivers.

OBJECTIVE: This study sought to compare train drivers’ fatigue and workload between a long-haul and a short-haul train trips.

METHODS: One-hundred train drivers, in two routes of Tehran-Mashhad (long-haul) and Tehran-Semnan (short-haul), were asked to complete the Samn-Perelli Fatigue Scale prior to departure, immediately after ending driving duty on the way going, and immediately after ending driving on the way back. Moreover, they were asked to complete NASA-TLX at the end of their shift.

RESULTS: Accordingly, understudy train drivers stated relatively similar levels of fatigue and workload for the two trips. Furthermore, fatigue scores were significantly higher at the end of shifts in both routes.

CONCLUSION: Overall, train drivers in long-haul trips had longer rest hours between the ways going and coming back, which seems to compensate for the side effects of longer driving durations.

Keywords

Railway industry NASA-TLX Samn-Perelli Fatigue Scale work hours

1 Introduction

Fatigue has long been recognized as a factor which can influence safety relevant/critical tasks among shift workers in various industries including rail transportation system [1 –6]. Train driving is among jobs within this industry, which has always been the target of elevated fatigue. Fatigue can influence train driving task in different ways. Kecklund et al. [7] undertook an analysis of 79 train accidents, in which the train driver was involved, and found that fatigue or sleepiness was a contributory factor in about 17% of the incidents [7]. According to accident analyses in Japan [8] and China [9], fatigue and sleepiness were among the causes of railway accidents. Moreover, the survey conducted by Härmä et al. showed that train drivers reported severe fatigue in over half of all night shifts and in about one-fifth of all the morning shifts [2].

Train driving involves many cognitive tasks including alertness, attention, working memory, long-term memory recall, situation awareness, judgment, and decision-making [10, 11]. These cognitive functions can be adversely influenced by fatigue. A decrease in alertness and reaction time have been shown to occur following an increase in fatigue level [12]. Additionally, fatigue is considered as a precondition for slips in attention (skill-based errors) among train drivers [13]. Dorrian et al. investigated the effects of fatigue on train drivers performance (n = 50). Fatigue magnitude was surveyed using the Fatigue Audit InterDyne (FAID) software, taking into account duration, circadian timing and recency of work and rest periods. The software was developed based on the Dawson and Fletcher’s fatigue model. A data logger also was fitted to the locomotive in order to record performance indicators including fuel consumption, fuel rate, braking behavior, and train speed. It was observed that drivers with high fatigue tend to be more reactive and their efficiency reduced through a decrease in train control and an increase in fuel consumption. Moreover, these high fatigue drivers engaged in heavy brake and speeding violations [14]. In another study, conducted by Dorrian et al. it was revealed that high fatigue (calculated with Dawson and Fletcher’s fatigue model [15]) influenced train drivers’ performance (n = 20) by braking less (indicated by brake pipe pressure) and driving at faster speeds [16].

Of the many factors which affect train drivers’ fatigue, a number of them are pertinent to the nature of their task which include high level of workload, inappropriate work-rest schedule, and excessive and irregular working hours [7 , 18].

Workload (both high and low level of workload) is amongst the key factors which impact fatigue negatively. In a study related to railway industry, workload was defined as the individuals’ effort to meet the demands of their work [19]. Several studies emphasized the importance of workload in rail industry and particularly for train drivers [17 , 20–22] Train drivers should be able to efficiently monitor and control incoming information and other environmental changes [23]. Porter specified the required skills of a train driver as the ability to remember and recall information, the ability to think ahead and assessing the impact of different factors affecting the train, fast reaction time, control skills, and the ability to maintain vigilance [24].

According to Zhang et al. continuous workload accompanied by subjective exhaustion can cause reduction in performance through fatigue [10]. The amount of workload the train driver is exposed to influences the way in which fatigue manifests its consequences [16]. Therefore, recognition of workload and its dimensions is an important issue.

Consecutive hours of train driving, excessive working hours, and long commutes had been shown to increase risk of accidents by generating fatigue and increase in workload [18, 25]. Chang et al. investigated the effects of consecutive driving on accident risk and found that accident risk grew with increased consecutive driving hours for both passenger and freight trains. Compared with the accident risk during the first hour, the accident risk doubles after four consecutive hours of driving [26]. Dorrian et al. reported duration of shift and workload as predictors of high level of fatigue among rail employees including train drivers [17].

Despite the high level of safety in rail system, the accidents negatively affect the organization, victims, and surrounding communities and result in loss of lives and severe adverse economic effects [27]. Fatigue, workload and prolonged working hours are among key causes of human errors, and consequent accidents in rail industry [28]. Elevated fatigue and workload decrease the route knowledge and deteriorate the performance [29]. Despite the critical role of work hours in issues such as fatigue and workload among train drivers, no study has addressed Iranian train drivers. Hence, the aim of the present studywas to

Measure fatigue and workload among train drivers.

Compare workload and fatigue between two kinds of train driving schedules including a long-haul and a short-haul passenger train trips.

2 Materials and methods

2.1 Participants

One-hundred volunteered train drivers, 71 in Tehran-Semnan route (short-haul trip) and 29 in Tehran-Mashhad route (long-haul trip) met the inclusion criteria including being healthy, lack of sleep disorders and medications or other medical conditions which interfere with sleep and alertness. Two train drivers were excluded because of missing data. Semnan station connects several base stations to each other and the number of trips to this station is more in comparison with Mashhad station; therefore, a higher number of drivers was selected from this route.

Participants, who drove passenger trains and two types of locomotives, including Simens and GM, were chosen randomly from two operating companies under the aegis of Keshesh section of Iran railway. The study was conducted during spring-summer in 2012.

2.2 Train drivers’ job task description

There are about 300 railway stations in Iran and the railway network converges in Tehran. Generally, according to the current schedule determined for train drivers, to navigate a passenger train, a complete train trip starts and ends at the origin depot and each work shift consists of five stages (Fig. 1). In the first stage, the train driver must attend the office of depot moderator about two hours before train departure and carry out several checks (i.e. equipment, switches, engine, etc.). Following this, the driver navigates the locomotive to the origin station. At the station, the passenger carrying coaches would be attached to the locomotive and after boarding passengers, train driver starts the trip. In the on-board stage the train driver is responsible for the operation of trains on the block (the distance between the origin station and destination station). At this point, the train driver is responsible for the safe control of the train (by controlling speed, monitoring signs and signals, monitoring the track and environment, etc.) in order to reach the destination station on time.

After approaching the destination station, the driver can take rest for a certain time. The length of rest time depends on the shift length. Then, in the fourth stage, the driver has to steer the train from the destination station to the origin station (the way back). In the last stage, after disembarking the passengers and detaching the wagons, the driver leads the locomotive to the depot and completes the reports before going off-duty.

In the present study two routes of Tehran-Semnan (for short-haul trip, with a length of about 240 kilometers) and Tehran-Mashhad (for long-haul trip, with a length of about 900 kilometers) were chosen for the purpose of comparison. For Tehran-Semnan the shift schedule includes four hours on-board stage (the way going), four hours rest, four hours on-board stage (the way back) and for Tehran Mashhad it consists of twelve hours on-board stage (the way going, including two groups of crew, each crew for six hours), a 12 hour rest, 12 hours on-board stage (the way back, include two groups of crew, each crew for six hours) (Fig. 1). Overall, the nature of tasks performed by train drivers in the selected short-haul and long-haul trips are the same. However, they differ in terms of shift schedule including consecutive driving hours and amount of rest, and also the number of drivers in the cockpit. Regarding short-haul trips, a train driver and an assistant are in charge of locomotive and train; and driving lasts consecutively for about 4 hours. In the selected long-haul trip there are two groups of crew, each taking the duty of driving task for six hours during the way going and six hours during the way back. The non-duty crew can take rest or sleep while the other crew does the driving.

2.3 Assessment tools and study procedure

2.3.1 Fatigue

In the present study, fatigue was assessed using the 7-point Samn-Perelli Fatigue Scale ranging 1 = Fully alert, wide awake, 2 = Very lively, responsive, but not at peak, 3 = Okay, Somewhat fresh, 4 = A little tired, less than fresh, 5 = Moderately tired, let down, 6 = Extremely tired, very difficult to concentrate, 7 = Completely exhausted, unable to function effectively) [30]. The tool was initially developed to assess fatigue in airlift operations [30] and then has been widely applied in different fatigue assessment topics across various transportation modes with an acceptable validity and reliability level [5 , 31–33].

2.3.2 Workload

NASA Task Load Index (TLX) was used to assess workload of train drivers in this research. NASA-TLX is a well-known subjective workload assessment tool which has been widely applied in different domains [34, 35]. It has been proven to be a valid and reliable tool in several studies [36, 37]. The rating consisted of six subscales including: mental, physical, and temporal demands, performance, effort, and frustration level. The calculation of this scale was done according to the method presented by Miyake and Kumashiro [38, 39]. In this way, the participants rated their level of workload for each subscale on a 10-cm visual-analog scale and then these scores were transformed to a 0–100. In this sense, two scores can be calculated: 1) Raw-TLX which is the arithmetic mean of six scores and 2) Adaptive Weighted Workload (AWWL) calculated by multiplying the highest score by 6 and the remaining five ordered scores by 5 to 1 respectively, and then averaging the obtained scores.

2.3.3 Study procedure

Participants were asked to complete NASA-TLX at the end of the shift (immediately after approaching origin station). Also, they were asked to rate their level of fatigue three times: prior to departure (in the depot), immediately after approaching destination station (for drivers in Tehran-Mashhad it was immediately after ending the driving duty on the way going), prior to approaching origin station (for drivers in Tehran-Mashhad it was prior to ending the driving duty on the way back). The examiner attended the depot and after gathering background (type of locomotive and time of departure) and demographic (age, work experience, weight, and height) data, the whole procedure and the exact time for rating scales were completely explained to volunteered train drivers. Since the shift schedule was determined by the relevant department, text messages were sent to participants on certain times in order to remind them to answer the scales.

3 Results

3.1 Participants’ demographics

All the participants were male (n = 97), aged between 23–59 years (mean age 37.35). On average, they had worked as a train driver for 14.18 years. Table 1 presents descriptive data related to demographic variables of train drivers in two understudy routes, separately. Results of independent samples t-test analyses revealed no significant differences between age, work experience and BMI of train drivers in these two routes (P > 0.05) (Table 2).

3.2 Fatigue

The trend of fatigue in two types of the studied trips (short-haul and long-haul) is displayed in Fig. 2. It can be seen that train drivers’ fatigue (prior to approaching destination station and prior to approaching origin station) obtained a higher score in Tehran-Mashhad route but these differences were not statistically significant (P > 0.05) (Table 2). Furthermore, Samn-Perelli Fatigue scale revealed that fatigue had significant higher score at the end of shifts in both routes (Table 3). However, the Samn-Perelli scores showed similar trends in both routes and there were no significant differences.

3.3 Workload

Effort and mental workload obtained the highest values amongst dimensions of NASA-TLX in both routes and Performance was rated as the least important subscale (Table 1). Figure 3 depicts the overall score for train drivers workload (AWWL and RTLX) and also its six subscales, separately in two routes. Train drivers’ workload between long-haul and short-haul trips was compared using Mann-Whitney statistical test. RTLX and AWWL were rated higher in Tehran-Semnan route. However, there were no statistically significant differences between Adaptive Weighted workload, RTLX, and workload subscales between train drivers in the studied routes.

3.4 Fatigue and workload

Correlation between the Samn-perelli Fatigue scale and NASA Task Load Index were assessed using Partial correlation coefficient, controlling for fatigue prior to departure and type of locomotive, see Tables 4 and 5. The results indicate that for Tehran-Semnan route, there is a positive correlation between fatigue prior to destination station and prior to approaching origin station (at the end of the shift) and mental demand (P < 0.05), when controlling for locomotive type. For Tehran-Mashhad positive correlations were shown between mental demand and RTLX and fatigue at the end of the shift.

4 Discussion

The present study compared fatigue and workload between two types of shift regime: a long-haul and a short-haul. Similar scores were reported for fatigue and total workload by train drivers in short-haul and long-haul trips. Our assumption was that the longer duration of shift in Tehran-Mashhad induces more fatigue and workload. However, taken together, it seems that fatigue and workload affect train drivers in both schedules similarly.

A body of research has compared factors such as fatigue and sleepiness among 12-hour and 8-hour shift systems across different jobs. Using BMS scales, Kallus et al. compared fatigue among rail traffic controllers (n = 18) and reported a higher increase of fatigue in the 8-hour compared to the 12-hour shift regime [40]. Lowden et al. investigated the effect of a change from a rotating 8-hour to 12 hour schedule on sleep, sleepiness, performance, perceived health, and well-being among control room operators of a chemical plant (n = 72). The authors observed that this change increased satisfaction with working hours, sleep, and also alertness [41]. It has been argued that flexible working conditions (e.g. longer pauses between shifts and extra off-days) may compensate for disadvantages of prolonged working hours [40 –42]. Nevertheless, the pattern of the work and the work-rest schedule in the mentioned studies differ from those in our research. As already noted, in the present study train drivers in long-haul trip could take rest for 12 hours between the ways go and back, which may cause fatigue to be not significantly higher than those of short-haul trip. This program may have less impact on circadian rhythm of train drivers. In Tehran-Semnan route drivers can take rest for four hours which sometimes reduced due to some irregularities which results in high level of fatigue and workload. Additionally, it is worthwhile to note that there are other factors, besides the total shift duration, which differ between the two understudy trips; each may widely affect train drivers alertness status independently. These factors include time for duty, number of drivers in cockpit, sleep conditions and facilities during the journey, types of locomotives run on rails, etc. which are recommended to be investigated in future.

Furthermore, the results revealed that fatigue reached a high level at the end of shifts. This result is consistent with those of Baulk et al. and Powell et al. who assessed fatigue, using Samn-Perelli Fatigue Scale, and found an elevated trend for fatigue at the end of each shift [31, 43]. This may indicate the chronic nature of fatigue [44] which mean that fatigue can accumulate over a period of time (e.g. at the end of the shift) and shows its signs and effects. Moreover, the significant difference of fatigue between approaching to destination station and origin station manifested that other factors apart from workload, such as quality and quantity of train drivers sleep during the shift, impact train drivers’ fatigue. Lamond et al. evaluated the sleep quantity and quality of train drivers (n = 40) during long-haul trips, using wrist activity monitor and sleep/wake diary. The results manifested the sleep to be lower and poorer during the relay, in comparison to sleep at home. Sleeping conditions for train drivers is likely to be an important factor which influences their sleep and should be considered [45]. Regarding train drivers’ workload, it was observed that effort and mental workload were the most critical dimensions of workload from train drivers’ point of view. In the study conducted by Dorrian et al. two dimensions of mental workload and temporal demand was rated highly by train drivers [17]. The higher effort can be attributed to the old cabins with poor physical conditions (including noise, vibration, etc) which impose physical strain to the train drivers’ body. Railway in Iran has been established since many decades ago and locomotives and drivers’ cabs are old and not ergonomically well-designed (with lifespan of averagely 40 years). Thus, exposure to various physical agents are expected which impose extra physical and mental demands to train drivers. Additionally, in our study, a positive correlation was observed between workload and fatigue at the end of the shift, when controlling for locomotive type, in the long-haul trip. This is in accordance with the results of Dorrian et al. [17] but this relationship is inconsistent with Popkin et al. [19].

5 Limitations

There are some limitations with the present study that should be noted. The first limitation is that as it was described in the method section, the examiner attended the depot and explained the whole process of the study and the exact time for rating scales for participants. Moreover, text messages were sent to participants on certain times in order to remind the subjects to answer the scales. Nevertheless, there is the possibility of train drivers’ forgetfulness in completing the questionnaires or even the possibility of internal or external distractions which made subjects to answer the questionnaires later.

Finally, it has been previously mentioned that fatigue is a complex and multidimensional phenomenon which can be investigated relying on different approaches including behavioral, psychosocial, physiological, and psychological approach [44, 46]. Accordingly, it is recommended to investigate fatigue and its consequences more specifically and objectively in future studies. In the present study we just relied on subjective fatigue and workload and it would be beneficial to apply objective measurements in order to examine fatigue and workload and evaluate the effects of these phenomena on performance outcomes such as errors, incidents, accidents etc.

Conflict of interest

The authors have no conflict of interest to report.

Footnotes

Acknowledgments

This study was financially supported by the Iranian Railway Research Center. The authors would like to thank Seyed Hossein Ghotb Hosseini and Hamid Mir Aali for their invaluable helps in data collection and all participants for their cooperation.

References

Åkerstedt

. Shift work and disturbed sleep/wakefulness. Occ Med 2003;53(2):89–94.

Härmä

, Sallinen

, Ranta

, Mutanen

, Müller

. The effect of an irregular shift system on sleepiness at work in train drivers and railway traffic controllers. J Sleep Res 2002;11(2):141–51.

Hockey

, Carrigan

. Human Factors in Railway Systems: Implications for Safety: Rail Research UK; 2004 [cited 2013 October 16]. Available from: http://portal.railresearch.org.uk/Shared%20Documents/Reports/rssbb2.pdf.

Ingre

, Kecklund

, Åkerstedt

, Kecklund

. Variation in sleepiness during early morning shifts: A mixed model approach to an experimental field study of train drivers. Chronobiol Int 2004;21(6):973–90.

Jay

, Dawson

, Ferguson

, Lamond

. Driver fatigue during extended rail operations. Appl Ergon 2008;39(5):623–9.

Stone

, McGuffog

, Spencer

, Turner

, Mills

. Fatigue and shift work in UK train drivers. In: Wilson

, Norris

, Clarke

, Mills

, editors. People and Rail Systems. Hampshire, UK: Ashgate Publishing Ltd; 2007. pp. 565–72 .

Kecklund

, Åkerstedt

, Ingre

, Söderström

. Train drivers’working conditions and their impact on safety, stress and sleepiness: A literature review, analyses of accidents and schedules: National Institute for Psychosocial Factors and Health. Stress Research Report; 1999. Available from: https://it.uu.se/research/project/train/papers/SRR288etapp1.pdf.

Kogi

, Ohta

. Incidence of near accidental drowsing in locomotive driving during a period of rotation. J Hum Ergol 1975;4(1):65.

Zhou

. Epidemiological features and causes of railway traffic accidents. Zhonghua yu fang yi xue za zhi [Chinese Journal of Preventive Medicine] 1991;25(1):26–9.

10.

Zhang

, Zheng

, Yu

. Automatic recognition of cognitive fatigue from physiological indices by using wavelet packet transform and kernel learning algorithms. Expert Syst Appl 2009;36(3):4664–71.

11.

Pilcher

, Coplen

. Work/rest cycles in railroad operations: Effects of shorter than 24-h shift work schedules and on-call schedules on sleep. Taylor & Francis;; 2000. pp. 573–88

12.

Dorrian

, Roach

, Fletcher

, Dawson

. Simulated train driving: Fatigue, self-awareness and cognitive disengagement. Appl Ergon 2007;38(2):155–66.

13.

Baysari

, McIntosh

, Wilson

. Understanding the human factors contribution to railway accidents and incidents in Australia. Accident Anal Prev 2008;40(5):1750–7.

14.

Dorrian

, Hussey

, Dawson

. Train driving efficiency and safety: Examining the cost of fatigue. J Sleep Res 2007;16(1):1–11.

15.

Dawson

, Fletcher

. A quantitative model of work-related fatigue: Background and definition. Ergonomics 2001;44(2):144–63.

16.

Dorrian

, Roach

, Fletcher

, Dawson

. The effects of fatigue on train handling during speed restrictions. Transp Res Part F Traffic Psychol Behav 2006;9(4):243–57.

17.

Dorrian

, Baulk

, Dawson

. Work hours, workload, sleep and fatigue in Australian Rail Industry employees. Appl Ergon 2011;42(2):202–9.

18.

Sussman

, Coplen

. Fatigue and alertness in the United States railroad industry part I: The nature of the problem. Transp Res Part F Traffic Psychol Behav 2000;3(4):211–20.

19.

Popkin

, editor An examination and comparison of workload and subjective measures collected from railroad dispatchers. Proceedings of the Human Factors and Ergonomics Society Annual Meeting; 1999: SAGE Publications.

20.

Coplen

, Sussman

. Fatigue and alertness in the United States railroad industry part II: Fatigue research in the Office of Research and Development at the Federal Railroad Administration. Transp Res Part F Traffic Psychol Behav 2000;3(4):221–8.

21.

Pickup

, Wilson

, Norris

, Mitchell

, Morrisroe

. The Integrated Workload Scale (IWS): A new self-report tool to assess railway signaller workload. Appl Ergon 2005;36(6):681–93.

22.

Pickup

, Wilson

, Sharpies

, Norris

, Clarke

, Young

. Fundamental examination of mental workload in the rail industry. Theor Issues Ergonomics Sci 2005;6(6):463–82.

23.

Zhao

, Lian

, Dang

, Tong

. Classification of driver fatigue expressions by combined curvelet features and gabor features, and random subspace ensembles of support vector machines. Journal of Intelligent and Fuzzy Systems 2014;26(1):91–100.

24.

Whitlock

, Pethick

, Mills

. Driver vigilance devices. In: Wilson

, Norris

, Clarke

, Mills

, editors. Rail human factors: Supporting the integrated railway. London Ashgate Publishing; 2005. pp. 120–30.

25.

Pollard

. Issues in locomotive crew management and scheduling. 1990.

26.

Chang

, Ju

. Effect of consecutive driving on accident risk: A comparison between passenger and freight train driving. Accident Anal Prev 2008;40(6):1844–9.

27.

Mazloumi

, Kazemi

, Nasl Saraji

, Hosseini

, Barideh

, editors. Performance Obstacles Associated with Train Drivers. Proceedings of the 5th International Conference on Applied Human Factors and Ergonomics AHFE; 2014; Kraków, Poland.

28.

Costa

. Shift work and health: Current problems and preventive actions. Safety and health at Work 2010;1(2):112–23.

29.

Naweed

, Balakrishnan

. Understanding the visual skills and strategies of train drivers in the urban rail environment. Work: A Journal of Prevention, Assessment and Rehabilitation 2014;47(3):339–52.

30.

Samn

, Perelli

. Estimating aircrew fatigue: A technique with application to airlift operations. DTIC Document, 1982.

31.

Baulk

, Fletcher

, Kandelaars

, Dawson

, Roach

. A field study of sleep and fatigue in a regular rotating 12-h shift system. Appl Ergon 2009;40(4):694–8.

32.

Powell

, Spencer

, Holland

, Petrie

. Fatigue in two-pilot operations: Implications for flight and duty time limitations. Aviat Space Envir Md 2008;79(11):1047–50.

33.

Powell

, Spencer

, Petrie

. Fatigue in airline pilots after an additional day’s layover period. Aviat Space Envir Md 2010;81(11):1013–7.

34.

Hart

, editor NASA-task load index (NASA-TLX); 20 years later. Proceedings of the Human Factors and Ergonomics Society Annual Meeting; 2006 Sage Publications.

35.

Hart

, Staveland

. Development of NASA-TLX (Task Load Index): Results of empirical and theoretical research 1988 [cited 2013 October 16]. 139–83]. Available from: http://humanfactors.arc.nasa.gov/groups/TLX/downloads/NASA-TLXChapter.pdf.

36.

Rubio

, Díaz

, Martín

, Puente

. Evaluation of subjective mental workload: A comparison of SWAT, NASA-TLX, and workload profile methods. Applied Psychology 2004;53(1):61–86.

37.

Xiao

, Wang

, Lan

. [The appraisal of reliability and validity of subjective workload assessment technique and NASA-task load index]. Zhonghua lao dong wei sheng zhi ye bing za zhi=Zhonghua laodong weisheng zhiyebing zazhi=Chinese Journal of Industrial Hygiene and Occupational Diseases 2005;23(3):178–81.

38.

Mazloum

, Kumashiro

, Izumi

, Higuchi

. Quantitative overload: A source of stress in data-entry VDT work induced by time pressure and work difficulty. Ind Health 2008;46(3):269–80.

39.

Miyake

, Kumashiro

. Subjective mental workload assessment technique-An introduction to NASA-TLX and SWAT and a proposal of simple scoring methods. Jpn J Ergon 1993;29(6):399–408.

40.

Kallus

, Boucsein

, Spanner

. Eight-and twelve-hour shifts in Austrian rail traffic controllers: A psychophysiological comparison. Psychol Sci Quart 2009;51(3):283–97.

41.

Lowden

, Kecklund

, Axelsson

, Åkerstedt

. Change from an 8-hour shift to a 12-hour shift, attitudes, sleep, sleepiness and performance. Scand J Work Environ Health 1998:69–75.

42.

Williamson

, Gower

CGI

, Clarke

. Changing the hours of shiftwork: A comparison of 8-and 12-hour shift rosters in a group of computer operators. Ergonomics 1994;37(2):287–98.

43.

Powell

, Spencer

, Holland

, Broadbent

, Petrie

. Pilot fatigue in short-haul operations: Effects of number of sectors, duty length, and time of day. Aviat Space Envir Md 2007;78(7):698–701.

44.

Shen

, Barbera

, Shapiro

. Distinguishing sleepiness and fatigue: Focus on definition and measurement. Elsevier; 2006. pp. 63–76.

45.

Lamond

, Darwent

, Dawson

. How well do train driver’s sleep in relay vans? Ind Health 2005;43(1):98–104.

46.

Korunka

, Kubicek

, Prem

, Cvitan

. Recovery and detachment between shifts, and fatigue during a twelve-hour shift. Work: A Journal of Prevention, Assessment and Rehabilitation 2012;41, 3227–33.