Railway Accidents Scenarios Analysis: Human

Factor Based Review

Sanghamitra Poddar, Subhash Chandra Panja, Malay Gangopadhyaya

Abstract- Railways form the backbone of transportation about 3.7 % in the number of train accidents compared
network of any country and directly impacts it’s to that in 2014 got registered in EU-28 in 2015 [6].
economical growth and prosperity. Evidently, providing Commission Implementing Regulation (EU) 402/2013
the cheapest means of transport compared to other
means, railways undergo tremendous amount of load and
(the Regulation on a common safety method (CSM) for
pressure. Thus, Safety in railways play a very important risk evaluation and assessment [or “the CSM RA”]) is
role all over the world. A proper investigation of the an initiative of European Railway Agency (ERA) and
various factors mainly comprising of technological and the European Commission (the Commission) to
human factors involved in the maintenance in railways is establish an open and competitive market in railways
thus very cruicial. The main purpose of investigating an seeking to ensure satisfactory safety levels. The CSM
accident or incident is to improve the safety of railways
and to prevent further accidents. One of the most adopted
RA intends to harmonise processes to evaluate and
way to investigate an accident or incident is by identifying access risk and document the evidences produced
the prime causes of accidents and other causal factors during the application of these processes [7].
contributing to the event or made the outcome worse A framework for risk management that serves this
followed by recommendations on what could be done by purpose is introduced in [8]. Several activities such as,
government, the industry or others to improve railway analysis of risk, identification of hazard, it’s
safety in future. Based on the study of literature, this
paper reviews the human factors primarily responsible
evaluation, control and treatment are suggested. Next,
for accidents in railways all over the world. this paper shows how the framework can be applied
systematically to reduce risk at a given Moroccan level
Index Terms- Railway accidents, Human factors. crossing.

II. Literature Review of Relevant Cases

I. INTRODUCTION
A. Evaluators of sleepiness among railway control
The accidents in railways mainly occur due to two workers at work
factors: technological and human errors. The main Description of the main observed effects of the
focus in this review paper is on the human factors three principal shifts (night, morning and afternoon) on
involved in maintenance leading to accidents in patterns of sleep and wakefulness was reviewed [10].
railways. Developing risk assessment models hinged Another review demonstrates that stress is closely
on scenarios of the accidents is very important. Models related to impaired sleep in cross-sectional studies [11].
assessing the risk of accidents with an indication of Insufficient sleep leads to sleepiness. Using continuous
how those can be applied emphasizing on Slovakian encephalographical/electro-oculographical (EEG/EOG)
railways is discussed in [1]. Confidence of safety in , sleepiness can be evaluated in the laboratory and
industry possess three overlying scopes: confidence clinic. But, this approach cannot be implemented easily
inside company, across companies, and the whole in large real-life field studies. An easily applied
railway system [2]. To some extent, these three are method is the subjective ratings of sleepiness. But the
interdependent. But, separating them should make the quality of the measurement remains uncertain.
task of understanding and applying this guidance Focusing on the Karolinska Sleepiness Scale (KSS),
easier. the relevant research of these issues are brought out
Data Dependent Systems (DDS) in it’s basic form [12]. The prospective relationship between major work
can be used to describe and model stochastic process, characteristics (physical, psychosocial, scheduling) and
for example operational process [4]. This paper briefs disturbed sleep has limited Knowledge. Such
the fundamentals of time series theory along with an knowledge is sought by the study in [13]. A study
algorithm of statistically adequate discrete model of aimed to learn to counter the driver fatigue (which
stochastic dynamic system. Further, it gives a includes sleepiness) in rail, road, air and sea modes
developed strategy application for a real process. transportation was reported [14]. The study came up
Railway safety statistics in 2015 reported that 1739 with a suggestion that possibility to nap and schedules
persons got killed or injured seriouly in accidents in taking the humans limitations into consideration are
railways in the EU-28, which accounts to 14 % less effective ways to counterneasure the fatigue,
than that in 2014 excluding suicides. An increase of irrespective of the mode of transportation.
 A clear understanding is required to why some should consider aging workers as more vulnerable
workers tolerate shift work and others do not. This subjects in relation to shift and night work, and protect
intolerance leads to a sense of dissatisfaction among them by arranging shift schedules according to
the workers with their shift schedule. To analyze the ergonomic criteria and adopting specific supporting
same, two groups of shift workers were observed who measures for aging workers [22].
were either satisfied highly with their shift schedule or The main results of several studies carried out on
dissatisfied during an entire shift cycle [15]. A variety Italian workers using the work ability index as a
of organizational interventions have been suggested to complementary tool for workers' periodical health
mitigate the negative effects of shift work on work-life surveillance were reported. The work ability index
balance and health of employees [16]. shows a general decreasing trend over the years, but it
Based on a study which comprised of firstly a large changes differently according to working conditions
number of midwives and nurses and secondly a sample and personal health status [23]. A conceptual basis for
of service and industrial workers, a set of normative managing the first two levels of an error trajectory for
data were formed against which comparisons with fatigue was reviewed. The concept is based upon a
other shiftworking groups could be made; relationship prior sleep/wake model, which determines fatigue-risk
between modifier variables and outcome; thresholds by the amount of sleep individuals have
differentiating between group of shift workers based on acquired in the prior 24 and 48 h. In doing so,
the sensitivity of the scales with evidence were offered managing level 1 of the error trajectory involves the
[17]. Investigations suggested less tolerance of older implementation of systems that determine probabilistic
people to shift work. Field studies on age–shift work sleep opportunity, such as prescriptive HOS rules or
interaction effects on fatigue, sleep, performance, fatigue modelling. Managing level 2, requires
health and accidents were reviewed [18]. For 15 individuals to be responsible for monitoring their own
consecutive days, a study on quality of sleep and sleep prior sleep and wake to determine individual fitness for
length and different ratings of social and other duty. Existing subjective, neurobehavioral and
activities were conducted. Vigilance performance electrophysiological research is reviewed to make
(PVT) and Subjective sleepiness (KSS) were registered preliminary recommendations for sleep and wake
at work [19]. Major factors leading to impaired health thresholds. However, given the lack of task- and
at work are shift work especially night, work-related industry-specific data, any definitive conclusions will
stress and ageing. Such risk factors can result in rely in post-implementation research to refine the
increased harm for health as evident from the thresholds [24].
evaluation in a sample of 1842 hospital workers in The impact of rotation and timing of shifts on work –
Northern Italy [20].  home conflict, job attitudes, health and absenteeism
Working during ‘non-standard’ are recognized risk among the military police was tested using a study. A
factors for safety, health and social well‐being. Suitable total of 3122 employees participated in the study.
preventive and protective measures are required to Discriminant analysis was used to examine the
mitigate the adverse effects and ensure that the worker relationships between rotation and timing of shifts on
can cope satisfactorily. These are based mainly on the the one hand, and the outcome measures on the other
organization of shift schedules according to ergonomic [25]. A study estimated the prevalence, examined
criteria and on specific medical surveillance. associated impacts, and identified correlated factors of
Occupational medicine has to consider very carefully poor sleep quality among Japanese white-collar
the several factors (psycho‐physiological, pathological employees who were working in a labour market that
and social) that can influence tolerance and/or mal included extensive downsizing and restructuring [26].
adaptation [21]. Aging people often show a reduced Research suggests that less than 5 h sleep in the 24 h
tolerance to shift and night work related to (a) prior to work and/or more than 16 h of wakefulness can
chronobiological factors (weakening of the circadian significantly increase the likelihood of fatigue-related
system, earlier phasing of circadian rhythms); (b) impairment and error at work. Studies have also shown
psycho-physical conditions (physical fitness, sleep exponential safety declines with time on shift, with
efficiency, intervening illnesses); (c) social conditions roughly double the likelihood of accident or injury
(housing, commuting, social integration); (d) working after 10 h relative to the first 8 h. While it is
conditions (work load, specific task, stress and fatigue, acknowledged that reduced sleep, increased
human relations). According to our surveys in different wakefulness and longer work hours produce work-
work sectors, shift workers appear to have lower WAI related fatigue, few studies have examined the impact
than day workers in all decades, but more pronounced of workload on this relationship. Studies in the rail
in the older ones. However, it is not easy to distinguish industry have focused on drivers. This study
the influence of the different factors as aging and shift investigated fatigue in a large sample of Australian
work seniority usually are strictly correlated, and, on Rail Industry Employees [27].
the other hand, the “healthy worker effect” becomes Operators' mental workload while monitoring traffic
more evident with increasing age and years spent on density in a city traffic control center was evaluated.
shift work. Managers, ergonomists and occupational To determine the mental workload, physiological
health physicians should be aware of these aspects and signals (ECG, EMG) were recorded and the NASA-
Task Load Index (TLX) was administered for 16 of alertness. Paper examines whether similar results
operators. The results showed that the operators might be obtained using retrospective survey measures
experienced a larger mental workload during high of alertness obtained from shiftworking nurses at a
traffic density than during low traffic density. The single sitting [32]. On-site studies of shiftwork show
traffic control center stressors caused changes in heart considerable differences in sleepiness between
rate variability features and EMG amplitude, although workers. Variation between individuals depends on
the average workload score was significantly higher in different sleeping, coping and living habits of the
HTD conditions than in LTD conditions. The findings workers, on possible sleep disorders, but also on more
indicated that increasing traffic congestion had a ‘permanent’ inter-individual factors that may influence
significant effect on HR, RMSSD, SDNN, LF/HF sleepiness by changing either the circadian or
ratio, and EMG amplitude. The results suggested that homeostatic factors of sleep. The effects of circadian
when operators' workload increases, their mental phase, age, sex, physical fitness, domestic and
fatigue and stress level increase and their mental health personality factors on shiftworkers' sleepiness and
deteriorate. Therefore, it maybe necessary to work-related accidents are reviewed. It was concluded
implement an ergonomic program to manage mental that only a minor part of the variation in sleepiness
health. Furthermore, by evaluating mental workload, between individuals can be explained. The present
the traffic control center director can organize the knowledge is not sufficient to carry out reliable
center's traffic congestion operators to sustain the ‘selection’ of individuals suitable for shiftwork.
appropriate mental workload and improve traffic However, preventive advice and medical surveillance
control management [28]. focusing on individual differences in sleepiness are
Public authorities in European countries are paying recommended [33].
increasing attention to the promotion of work ability Sleepiness and fatigue are frequent problems in
throughout working life and the best method to monitor railway transportation with occasional monotony and
work ability in populations of workers is becoming a irregular shift schedules. Study aimed at (1) studying
significant question. The present study aims to the prevalence of severe sleepiness in shifts and (2)
compare the assessment of work ability based on the examining which shift and sleep-related factors were
use of the Work Ability Index (WAI), a 7-item associated with the occurrence of severe sleepiness in
questionnaire, with another one based on the use of an irregular shift system [34]. Shift work is related to
WAI’s first item, which consists in the worker’s self- problems in sleep/wakefulness and social life. The
assessment of his/her current work ability level as effects of a very rapidly forward rotating shift system
opposed to his/her lifetime best, this single question on sleep, health and well-being of young (− 45) and
being termed “Work Ability score” (WAS) [29]. elderly (45+) maintenance workers were studied by a
Variables associated with inadequate work ability controlled intervention study. In the beginning, all the
among nursing personnel at a public hospital, workers had a continuous backward rotating three-shift
considering factors related to socio‐demographic, system. A very quickly forward rotating shift system
lifestyles, working conditions, and health outcomes was developed, avoiding consecutive night shifts and
were identified [30]. Shiftwork researchers have with more free-time between the individual shifts. The
concentrated on the problems on the night shift, and effect of the new shift system on sleep/wakefulness and
have tended to neglect the truncation of sleep that can general well-being was studied by questionnaire and
occur before morning, or ‘early’, shifts. The sleep field studies including on-site registration of sleep
timing and duration between morning shifts of shift (actigraphy), subjective sleepiness (KSS) and
workers on a variety of shift systems were examined. psychomotor performance (PVT) [35].
The extent of morning shift sleep truncation depended In the last two decades the control of the Portuguese
very largely on the time at which individuals had to railway network has become much more centralized in
leave home. Further, there was some suggestive three centres, there integrating the functions of route
evidence that the failure of individuals to compensate flow management, electrical control and signalling.
by going to sleep earlier was not simply due to social Study aimed to investigate the influence of work and
pressures to stay up. Rather it would appear to have individual determinants in sleepiness among railway
depended, at least in part, on the biological clock control workers, namely socio-demographic factors,
determined ‘forbidden zone’ for sleep that occurs work ability, psychosocial factors, shiftwork
shortly before habitual sleep onset [31]. characteristics, fatigue perception, and sleep [36].
Current models of sleep regulation postulate both a Patterns and work schedules of employees were
homeostatic and circadian component and promise an evaluated in terms of need for recovery from work
understanding of disturbed and displaced sleep. [37]. The results of some intervention studies on the
Estimates of these components have traditionally effects of the change from weekly rotating to quicker
required relatively cumbersome and costly measures, rotating shift systems were presented [38].
including sleep electroencephalograms and Given the associations between poor quality sleep
continuously recorded rectal temperature, but it has and health, it is important to consider whether job
recently been demonstrated that they may successfully stressors are related to sleep-related outcomes. Studies
be based on frequent (e.g. 2-hourly) concurrent ratings from Europe and Japan suggest that these stressors
negatively impact sleep, but there are few studies of sleep, participants obtained less sleep and reported
job stressors and sleep quality that draw upon large lowered alertness on shift days. Tests near the
representative samples of workers in the USA. Using circadian nadir showed higher sleepiness and increased
data collected via telephone interviews from a performance deficits. While this schedule did not
nationally representative random sample of 1715 produce cumulative impairment, the performance
American full-time employees, this research considers deficits witnessed during the biological night are still
three dependent variables of past-month poor sleep of operational concern for industry and workers alike
quality: number of days the respondent had difficulty [48]. Lessons learned within the administration of the
initiating sleep, number of days of difficulty survey, the analysis of the data and dissemination of
maintaining sleep, and number of days of non- the findings will be of interest to researchers. An
restorative sleep [39]. By describing the dynamic overview of the scope of analyses of the database is
complexity of fatigue, it may help make explicit what introduced, in addition to proposals for further
different studies do or do not measure or account for in development of the survey instrument [49]. The
terms of the different aspects of fatigue. It was claimed hypothesis was largely not confirmed. Relatively few
that the proposed definition could be used to help associations were found, of which the majority was in
harmonise attempts to study and tackle fatigue in the direction as expected. In particular early starts of
transport health and safety contexts [42]. Another study morning shifts and many consecutive shifts seem to be
highlighted the value and difficulties of longitudinal avoided. The healthy worker effect, limited variation
nursing research [41]. between included schedules and the cross-sectional
The findings demonstrate marked differences in the design might explain the paucity of significant results
occurrence of severe sleepiness at the wheel, sleep [50].
preceding duty hours, and the use of sleepiness The objective of review was to examine the evidence
countermeasures between different shift types. In for the link between fatigue and safety, especially in
addition, although drivers slept reasonably well in transport and occupational settings [51]. Current major
connection with different shift types, the findings rail human factors programmes are summarised and
imply there is still room for improvement in alertness future research needs proposed. It was asserted that
management among this group of employees [42]. general human factors models and methods are being
Among men, ERI is a risk factor for the development re-assessed, and new ones developed, to meet the
of sleep disturbances in the Danish workforce. Among requirements of the railways [52].
women, an association between ERI and sleep Additional studies are needed for further
disturbances was restricted to the cross-sectional understanding of the mechanisms underlying the health
sample. Improving psychosocial working conditions risks of shift work, understanding which shift workers
might reduce the risk of sleep disturbances and are at most risk of SWD, to investigate treatment
subsequently also help to prevent clinical disorders options that address the health and safety burdens
related to sleep disturbances [43]. associated with shift work and SWD, and to further
Lessons learned within the administration of the develop and assess the comparative effectiveness of
survey, the analysis of the data and dissemination of countermeasures and treatment options [53].
the findings will be of interest to researchers. An
overview of the scope of analyses of the database is III. Literature Review on Techniques Used
introduced, in addition to proposals for further
A new useful analysis technique: timed fault tree
development of the survey instrument [44]. An expert
analysis extends traditional fault tree analysis with
group reviewed question wording, and the findings
temporal events and fault characteristics [3]. Timed
from the principal components analysis and other
Fault Trees (TFTs) can determine which faults need to
analyses of data from previous administrations of the
be eliminated urgentlyand it can also provide how
survey. Improvements were made to the design and
much time have been left at least to eliminate the root
layout of the questionnaire. The effectiveness of the
failure to prevent accidents.
process for the expert review of questions was
ETree is used to model a safety system (set of safety
discussed [45]. Shift work tolerance is a term
barriers) as an event tree. It produces simple models
describing the ability to adapt to shift work without
suited to all domains (aeronautics, automobile, rail, oil
adverse consequences. Paper systematically reviewed
& gas, etc.) and is an easy way to obtain risk matrices.
literature published investigating the relation between
The ETree module uses ALBIZIA, the BDD (Binary
individual differences such asage,gender, personality,
Decision Diagram) computation engine developed by
morningness/eveningness as well as biological
Total. ALBIZIA offers the advantage of
variables and different measures of shift work
running accurate analytical computations and
tolerance from 1998 till 2009 [46]. The results
providing extensive information on the system under
supported the view that the last hours of a single 12-h
study [5].
dayshift with frequent pauses are not associated with
Another approach, Fault tree analysis uses tree
an increase in sleepiness or performance errors [47].
structures to decompose system level failures into
Results showed that while there was not a
combinations of lower-level events, and boolean gates
cumulative cost to performance across days of splitting
to model their interactions [9]. [13] Torbjorn Akersted, et. al, “Work and Sleep—
A Prospective Study of Psychosocial Work
Factors, Physical Work Factors, and Work
Scheduling”.
[14] Anna Anund, et. al, “Countermeasures for
fatigue in transportation”.
IV. CONCLUSION [15] John Axelsson, et. al, “Tolerance to shift
The objective of this study is to provide an analysis on work—how does it relate to sleep and
railway accident scenarios based on human factors wakefulness?”.
responsible for railway accidents. In this review paper, [16] Bambra, Clare L, et. al, “Shifting Schedules”.
prime focus is given on the determinants of sleepiness [17] Barton, Jane, et. al, “The Standard Shiftwork
Index: a battery of questionnaires for
at work among railway control workers with an aim to assessing shiftwork-related problems”.
give a detailed insight on how to reduce accidents in [18] Blok MM, et. al, “What is the evidence for
railways. Railway accidents in India and abroad have less shift work tolerance in older workers?”.
been detailed. Further, to overcome such adversities, [19] A. Bonnefond, et. al, “Interaction of Age With
various techniques that have been developed or Shift-Related Sleep-Wakefulness,
Sleepiness, Performance, and Social Life”.
proposed are discussed. This review strongly aims at [20] P.M. Conway, et. al, “Main and interactive
helpimg railway authorities to prevent and effects of shiftwork, age and work stress on
subsequently reduce the number of accidents and hence health in an Italian sample of healthcare
improving the safety in railways. Safety engineering, workers”.
high quality maintenance, training and strict law [21] Giovianni Costa, “Shift work and
occupational medicine: an overview”.
enforcementare some steps to be focused on by the the [22] Giovianni Costa, “Some considerations about
railway authorities to prevent the accidents. Also, it not aging, shift work and work ability”.
only the responsibilty of railway authorities, but the [23] G. Costa, et. al, “Ageing, working hours and
people using the service must also be responsible and work ability”.
avoid incidents like travelling on foot board, boarding [24] Drew Dawson, et. al, “Managing fatigue: It's
about sleep”.
or de-boarding from running trains, over-crowding of [25] Evangelia Demerouti, et. al, “The impact of
trains to avoid any fatalities. shiftwork on work -home conflict, job
attitudes and health”.
REFERENCES [26] Yuriko Doi, et. al, “Impact and Correlates of
Poor Sleep Quality in Japanese White-Collar
[1] Bohus Lietner, “A General Model for Railway Employees”.
Systems Risk Assessment with the use of [27] Jillian Dorrian, et. al, “Work hours, workload,
Railway Scenarios Analysis”. sleep and fatigue in Australian Rail Industry
[2] Safety Assurance Guidance and examples of employees”.
good practices in safety assurance in Britain’s [28] Fallahi M, et. al, “Effects of mental workload
railway industry. on physiological and subjective responses
[3] Zhaoguang penga et. al., “Risk Assessment of during traffic density monitoring: A field
Railway Transportation Systems using Timed study”.
Fault Trees”. [29] Mehdi El Fassi, et. al, “Work ability
[4] Bohus Leitner, “Data Dependent Systems as a assessment in a worker population:
Tool for Modelling of Stochastic Processes in comparison and determinants of Work Ability
Transport Area”. Index and Work Ability score”.
[5] ETREE Module. [30] Frida Marina Fischer, et. al, “Work Ability of
[6] Railway safety statistics. Health Care Shift Workers: What Matters?”.
[7] Common Safety Method for risk evaluation [31] Simon Folkard, et. al, “Does the ‘forbidden
and assessment. zone’ for sleep onset influence morning shift
[8] Abdelaziz Berrado et. al. “A Framework for sleep duration?”.
Risk Management in Railway Sector: [32] Simon Folkard, et. al, “The Use of Survey
Application to Road-Rail Level Crossings”. Measures to Assess Circadian Variations in
[9] Fault Tree Analysis. Alertness”.
[10] Torbjorn Akersted , “Shift work and disturbed [33] Mikko Harma, “Sleepiness and shiftwork:
sleep/wakefulness”. individual differences”.
[11] Akersted T., “Psychosocial stress and [34] M. Harma, et. al, “The effect of an irregular
impaired sleep”. shift system on sleepiness at work in train
[12] Torbjorn Akersted, et. al, “Subjective drivers and railway traffic controllers”.
sleepiness is a sensitive indicator of [35] Mikko Harma, et. al, “A controlled
insufficient sleep and impaired waking intervention study on the effects of a very
function”. rapidly forward rotating shift system on
 sleep–wakefulness and well-being among
young and elderly shift workers”.
[36] Teresa Cotrim, et. al, “Determinants of
sleepiness at work among railway control
workers”.
[37] Nicole Jansen, et. al, “Need for recovery from
work: evaluating short-term effects of
working hours, patterns and schedules”.
[38] P. Kanuth, “Designing better shift systems”.
[39] Hannah K. knudsen, “Job stress and poor
sleep quality: Data from an American sample
of full-time workers”.
[40] Ross et. al, “A review of definitions of fatigue
– And a step towards a whole definition”.
[41] Anne pisarski et. al, “What roles do team
climate, roster control, and work life conflict
play in shiftworkers' fatigue longitudinally?”.
[42] Pylkkonen et. al, “Sleepiness, sleep, and use
of sleepiness countermeasures in shift-
working long-haul truck drivers”.
[43] Reiner Rugulies et. al, “Effort–reward
imbalance at work and risk of sleep
disturbances. Cross-sectional and prospective
results from the Danish Work Environment
Cohort Study”.
[44] Brendan Ryan et. al, “Attitudes and opinions
of railway signallers and related staff, using
the Rail Ergonomics Questionnaire
(REQUEST)”.
[45] Brendan Ryan et. al, “Developing a Rail
Ergonomics Questionnaire (REQUEST)”.
[46] Ingvild B. Saksvik et. al, “Individual
differences in tolerance to shift work – A
systematic review”.
[47] Mikael salliman et. al, “The effects of sleep
debt and monotonous work on sleepiness and
performance during a 12-h dayshift”.
[48] Michelle A. Short et. al, “The effect of split
sleep schedules (6h-on/6h-off) on
neurobehavioural performance, sleep and
sleepiness”.
[49] Brendan Ryan et. al, “Attitudes and opinions
of railway signallers and related staff, using
the Rail Ergonomics Questionnaire
(REQUEST)”.
[50] Hardy A. van de Ven et. al, “Associations
between shift schedule characteristics with
sleep, need for recovery, health and
performance measures for regular (semi-
)continuous 3-shift systems”.
[51] Ann Williamson, “The link between fatigue
and safety”.
[52] John R. Wilson et. al, “Rail human factors:
Past, present and future”.
[53] Kenneth P. Wright Jr et. al, “Shift work and
the assessment and management of shift work
disorder (SWD)”.