1. Background
It is increasingly acknowledged that drivers’ fatigue
is an important cause of road accidents. These accidents are mostly very
serious and often lead to severe driver injuries accompanied by high material
costs. The main cause of drivers’ fatigue is a lack of restful sleep. People
have to adapt to the 24/7 society that requires many hours of work, working on
shifts and short rest periods. In addition sleep disorders have become more
frequent in recent years. Also life style factors with unbalanced nutrition and
little physical activity may impair driver fitness and lead to fatigue as well.
The problem of fatigue will further increase in the next years due to
increasing mobility of people and goods 24 hours each day. Accordingly, the aim
of the “
The most comprehensive research undertaken regarding
the effects of drivers’ fatigue has been carried out in the
In
1.1 Hours of Service are not enough
Alertness management in transport operations has been
a neglected matter in the European Union. Hours-of-Service (HOS) regulations
for professional drivers do exist, but these rules are often not adhered to.
Monitoring nearly 600.000 motor trucks per year, the German Federal Office of
Commercial Transport found that 10.5% of the truck drivers did not comply with
the Hours-of-Service rules. The real number of cases is likely much higher.
The most conspicuous observation concerning the causes
of all fatigue-related accidents is that peak levels of road accidents are
about 10 times higher at night than at daytime. French research about truck
driver working times and habits showed that risk level varies with three key
factors: There is an increased risk of accidents 1) when it is night, 2) the
greater the length of the working day, and 3) the more irregular the working
hours. Driving time regulations do not take into account the time of the day, though
it is the most important parameter for the development of fatigue. Moreover,
resting times do not necessarily coincide with the best time to sleep, so that
a sleep deficit may develop. In addition, the competitive situation in industry
leads to stretching legal possibilities to the limits. Since so many road
traffic accidents have been happening as a result of fatigue in spite of Hours-of-Service
regulations, these regulations appear to be inadequate.
On the other hand, the problem of drivers’ fatigue has
been approached by technical means. For example, non-intrusive, vision-based
systems have been developed to monitor the driver’s eyes. If long periods of
eye closure are detected, a warning signal is produced. In addition, different
indicators of driving performance, like lane deviation, may be controlled. The
idea is that with increased fatigue and/or phases of microsleep subjects are
increasingly impaired in their ability to keep the lane. Such systems, however,
bear the disadvantage of being expensive and not yet sufficiently reliable. Moreover,
they produce the warning signal at a time when sleepiness has already exceeded
a critical threshold, such that further driving is impeded and subjects are
forced to pause and take a nap. Thus, to promote the highest level of safety on
national highways, more comprehensive programmes considering the causes of fatigue
must be established.
As there are no alertness management programmes in the
European Union yet, a great call for action exists. Alertness management deals
with all causes and consequences of drivers’ fatigue in transport operations
and aims at preventing fatigue. It presents an integrated approach combining
several methods to achieve higher transport safety.
1.2 Alertness management in Australia vs.
European efforts in alertness management
In
To close the gap of missing alertness management in
the European Union, alertness management concepts have to be developed and their
effects have to be evaluated on a short, medium and long time scale. Measures should
be specific to the needs and requirements of the different drivers such as truck
drivers, people at driving schools, bus drivers etc. In addition, they should be aimed at professional drivers
who are involved in high-risk transports, where the transported passengers or
goods or the surroundings are at high risk due to a possibly fatigued driver.
1.3 Advantages and disadvantages of alertness
management
Alertness management shows potential to reduce
accidents and casualties. Apart from the fact that there is always a high level
of human suffering, studies have shown that in
Both drivers and the employing companies may benefit
from alertness management programmes. Within the scope of the programme, drivers
get training on hazard identification and avoidance. Hence, an advantage from
the drivers’ point of view will be a decreased accident risk. Another essential
point is that the drivers’ state of health will improve. Moreover, training
truck drivers will enhance the image of professional drivers not only in their
own perception, but also in the public. Companies will also benefit from a
higher level of road safety, e.g. due to lower costs as a result of less
traffic accidents and lower insurance rates. There will be a higher efficiency
of work because of less employee turnover and times absent as well as a higher
job satisfaction of the truck drivers. Besides the company will be a precursor
regarding alertness management programmes.
Management of alertness in road traffic might also
pose problems. First of all, resources have to be provided for the drivers’
training. The most important aspect is the time which is needed for the
training. Guaranteeing successful training results requires an adequate number
of training units as well as a preliminary discussion and debriefing. That
means people trained at driving schools or truck drivers have to take working
time off to participate in the alertness management. An associated issue are the
costs of the training. They include preparational activities, implementation of
the alertness management with training materials and working time of the
trainers and the participating drivers. This might result in driving schools
getting more expensive. Another significant aspect is the sustainability of the
acquired knowledge. The question is how long the inputs of the alertness
management training are kept hold of by the participants and whether a special
follow-up training will be an essential part of it. Final points worth
mentioning are poor motivation of the training participants, a lack of risk
awareness and not enough time to implement the outputs of the training.
2. Alertness
Management
To combat drivers’ fatigue and to improve transport
safety, the Department of Traffic Safety at the
The main factors causing fatigue are the time of day
(i.e. the circadian rhythm), time since last sleep, and the time on task.
First, performance in tests varies over the course of the day in a
predetermined manner. This is called the „inner clock“ or circadian rhythm and
is synchroniced by daylight. Performance is especially impaired in the night
hours between 1 and
-----------------------------------
Insert Figure 1 about here
-----------------------------------
This three component model of fatigue is also implemented
in a computer program named “ALERT”. This programme can predict fatigue over
the course of a drive and generate suggestions as to when breaks should be made
to prevent phases of critically increased fatigue. With the help of ALERT
schedules and rosters can be compared regarding their effects on fatigue. The
programme is primarily used for the training programme “SAFE-T”, which takes both
the drivers and the transport company management into account. It teaches
drivers and schedulers about causes of fatigue and strategies of prevention.
-----------------------------------
Insert Figure 2 about here
-----------------------------------
Next to SAFE-T, a series of other stregies is also
applied. These include the counseling of individual people and companies, e.g.
regarding the chronobiological and other physiological and psychological
factors causing fatigue. An important point is that alertness management is
also a political topic, that is insights from sleep research and alertness
management programmes have to reach the persons in charge to be implemented in
adequate regularizations. Thus, counseling of persons from the political sector
are also included in the portfolio. In addition, campaigns are mounted to
increase public interest in the issue of fatigue (e.g. by means of the distribution
of brochures etc). Some final points worth mentioning involve the
implementation of alertness management strategies in the scope of driving
schools and their instructions. Moreover, reporting systems can be set up and
medical screening procedures with respect to daytime sleepiness are possible.
Another aspect of the holistic alertness management
approach are fit-for-driving tests. They assess fatigue, sleepiness and
performance and may be presented to the driver to prove if his status allows
for the beginning of duty. The implementation of these tests requires a basic
research programme which is in progress. First experiments deal with the
appropriateness of fit for driving tests, validation of a subjective fatigue
scale, and the time on task component of fatigue models.
Furthermore, ALERT can be combined with other services
such as routing on the internet. For example, recommendations for breaks during
a journey can be made on the basis of the information of the user. These tips
may be coupled with suggestions for rest areas, sights, hotels for overnight
stays etc. Recommendations can be
optimized to prevent times in which traffic jams are likely. Experiments are in
progress investigation the combination of ALERT and routing.
2.1 The training programme SAFE-T
The development of the training programme has been
supported by a grant of the German government (German Federal Ministry of
Education and Research, Bundeministerium für Bildung und Forschung, BMBF). An
evaluation study is just being conducted with drivers of the truck fleet of the
Ford AG in
2.1.1
Phase 1: Drivers’ training
The acronym SAFE-T stands for Self-responsibility, Advice,
Feedback, Evaluation and Training
and outlines the basic principles of the training program. It is based on a psychological
concept deduced from the social-cognitive process model of health-related
behaviour by Schwarzer (1996). Standard methods of behavioural therapy are
applied in the training programme, as for example self-observation, behaviour
modification and cognitive re-structuring according to Beck. Personal strategies
of alertness management are elaborated together with the drivers. The transfer
of these strategies to the daily routine is accompanied by a sleep and fatigue
diary, which the drivers have to fill in over several weeks. In addition,
drivers are supported by a telephone hotline. The training is tuned for the
daily routine of and the content of the training has been developed in close
cooperation with truck drivers.
However, even if a driver tries to control his
behaviour according to the best practice of alertness management, he remains
influenced by situational barriers. For example, familiar duties may hamper
adequate behavior. Therefore, the social support of a driver by his family and
his colleagues as well as the involvement of the company management are particularly
important.
The training consists of an introduction (about 45
minutes), two training units of about 90 minutes and a final review session. In
the introductory section the drivers are informed about the content and schedule
of the training. The drivers and the training personnel sign a contract that ensures
active participation and the general willingness to test new behavioural
strategies.
The first training unit is dedicated to exercises in
which drivers describe their individual signs, causes of, and reactions to
fatigue. After the training unit the trainers develop individual alertness
management tips based on the information given by the drivers. During the
second training unit implications of shift work for alertness as well as
frequently occurring sleep disorders and their therapy are dealt with. The
computer program ALERT is used to illustrate the negative effects of too little
sleep, long duty hours and night work on fatigue and the positive effect of
countermeasures, e.g. breaks, naps, and appropriate sleeping times.
-----------------------------------
Insert Figure 3 about here
-----------------------------------
In the final review session experiences with the alertness
management are discussed and additional advice is provided if necessary. The
drivers get an „Alertness Kit“ containing individual alertness management tips,
a brochure „Awake at the Wheel“, ear plugs, a sleeping aid and individual
information on cognitive restructuring.
First results of the pilot study are already available.
Assessing the training efficiency with appropriate questionnaires and the sleep
and fatigue diary, it was shown that the alertness management training has the
potential to improve the driver’s knowledge regarding basic principles of sleep,
fatigue and shift work. Furthermore, statistically significant changes in the
drivers’ attitudes were found. Data
analysis will presuably be completed in September 2005.
2.1.2 Phase 2: Scheduler training
Moreover, as mentioned above, not only drivers could
benefit from alertness management strategies. Thus, another aim is to actively
involve the schedulers in the company’s attempt to cope with fatigue. An understanding
of driver behaviour under the influence of fatigue and knowledge about
fit-for-driving tests and other means of alertness management has to be
established.
The scheduler training is primarily based in the
computer programme ALERT to predict phases of decreased alertness. With the aid
of this programme, schedules and rosters can be optimized, that is, fatigue can
be considered in the planning of driving, breaking and loading times. The aim
is to invent a system which helps the scheduler to react to fatigue preceding
the start and during a tour. Furthermore, companies can be advised regarding
the optimal equipment of the trucks concerning fatigue-related countermeasures.
This consulting service requires a supervision of the company over the course
of several weeks.
Phase 2 involving the scheduler training has just
started. Initial results are expected for xxxx.
2.2 Assessing Fitness-for-Driving
Fit-for-driving tests are means to avoid that fatigued
drivers start duty. They are presented to the driver before driving and have to
be short to keep the costs low and to increase acceptance. Some measures are
not suitable as fit for driving tests, e.g. since they require a laboratory
equipment and cannot be used in work environments (e.g Multiple Sleep Latency
Test, pupillography). Other tests are too long (Mackworth clock vigilance
test), and some are not validated. Even if all these conditions are fullfilled,
they might not be accepted by the drivers.
At least three different formats of fit-for-driving
tests can be distinguished, 1) checklists, 2) standardised subjective fatigue
and sleepiness scales, and 3) performance tests. They can be presented on a
hand-held computer, which makes data storage and evaluation convenient and
increases the acceptance by the drivers.
A fit-for-driving checklist is a questionnaire to
document information relevant to drivers’ fatigue. The most important data in
fatigue assessment is the duration of the last sleep period and the time since
last sleep. Also quality of sleep and activities in the time since the last
sleep is helpful information. Possible questions for checklists are: When did
you fall asleep? When did you wake up? How long have you been awake in between?
How refreshing was your sleep? How demanding was the time since your last
sleep? Do you think you will become too tired during driving? Answers to the
questions can be marked on a five point scale: not at all, a little, somewhat,
quite a bit, very much.
The main problem with such a checklist is that it can
be manipulated by the drivers. Nevertheless, the questions given have an
educational value. They can make drivers aware of the issues that affect their
performance. Moreover, filling in a checklist is not time-consuming.
Standardised subjective fatigue and sleepiness scales
can be used to reveal fatigue or sleepiness of the driver. The Karolinska
sleepiness scale is a semi-qantitative standardised ten-point scale, on which
the driver has to mark his sleepiness during the previous ten minutes. The idea
is that if the driver’s sleepiness is beyond a certain threshold, he should not
start his duty. First attempts have been made to validate the threshold by
simultaneous recordings of EEG and EOG signals from which microsleep can be
assessed. However, the research base is still weak and further experiments are
needed. In the Samn-Perelli fatigue scale the driver has to mark his fatigue in
relation to the statements given on the display. Fatigue assessed by this scale
has a score between 0 and 20. The Samn-Perelli scale is standardised and has
been used for years mainly in aviation.
Both the Karolinska sleepiness and the Samn-Perelli
fatigue scales can be manipulated by the drivers towards better results,
especially when it is known which result can restrain the driver from starting
his duty. However, an advantage is that subjective scales are usually completed
in a shorter time than fit-for-driving checklists.
Performance measures should involve tests that are
especially sensitive to the effects of fatigue. They may consist of a simple
reaction time test or an unstable tracking task. A simple reaction time test
measures the response time to the occurrence of a visual signal. That is the
user is asked to react as quickly as possible to a circle that appears on the display.
The unstable tracking task is a test in which a cursor moves horizontally on the
display according to an internal disturbance signal and according to the user’s
input. The user has to keep the cursor in the center of the screen. The average
distance of the cursor from the middle of the screen and the number of so
called “control losses” when the cursor reaches the margins of the screen are
measured.
-----------------------------------
Insert Figure 4 about here
-----------------------------------
Both performance tests are more demanding than the
fit-for-driving checklist and the fatigue and sleepiness scales. Unlike the
subjective scales and the fit-for-driving checklist, the performance tests
cannot be willfully manipulated by the driver towards better results, but only
towards worse ones. One disadvantage is that the duration of the tests has to
be quite long, since it is known that the initial results can be influenced by
the user’s short-time motivation. From an economical point of view the duration
of the test should be kept as short as possible. Moreover, due to large interindividual
differences in the results of the test, it is necessary to evaluate results of
each driver separately. Current results have to be compared with previous “baseline”
results of the same subject in an awake state to reveal differences in
performance due to fatigue. Experiments are in progress addressing these
issues.
3. Conclusions
and perspective
Alertness management as presented here is a useful
means to fight the adverse effects of fatigue on road safety. Further endeavors
will be made to continuously improve its efficacy. To implement alertness
management area-wide in other truck fleets, the evaluated training SAFE-T will
be refered to the association of commercial and industrial workers’ compensation
insurance carriers. Furthermore, the training will be adapted to other target
groups such as bus and train drivers. Future projects involve the expansion of
the combination of ALERT and routing on the internet and experiments on basic
research questions regarding the factors causing fatigue. Furthermore, the
portfolio shall be expanded integrating other important aspects influencing
traffic safety, e.g. age and gender. The latter issues are based on theoretical
assumption regarding the emerging concept of “Adaptive Automation”. Adaptive
Automation provides a framework for the meaningful use of technical systems to
support the individual in everyday activities.
Essential
references:
Akerstedt T, Folkard S, Portin C. Predictions
from the three process model of alertness. Aviation, Space and Environmental Medicine
2004; 75(3): A75-A83.
Borbély AA. A two-process model of sleep
regulation. Human Neurobiology 1982; 1: 195-204.
Bundesanstalt
für GüterverkehrBei Straßenkontrollen nach den VO’en (EWG) Nrn. 3820/85,
3821/85 und dem AETR festgestellte Verstöße nach Verkehrsarten. Stand 21.02.01,
2000.
Commonwealth
of Australia. Beyond the
European Commission. Saving 20.000 lives on our
roads – a shared responsibility.
Maycock G. Sleepiness and driving: the experience of
Moore-Ede M, Heitmann A, Guttkuhn R, Trutschel
U, Aguirre A, Croke D. Circadian alertness simulator for fatigue risk
assessment in transportation: application to reduce frequency and severity of
truck accidents. Aviation, Space and Environmental Medicine 2004; 75(3):
A107-A118.
ten
Thoren C, Gundel A. Müdigkeit als Unfallursache im Stadtbereich - eine
Befragung von Unfallbeteiligten. Somnologie 2003; 7(4): 125-133.
Samn SW, Pirelli LP. Estimating aircrew fatigue:
a technique with implications to airlift operations (Technical Report No
SAM-TR-82-21). USAF
Schwarzer
R. Psychologie des Gesundheitsverhaltens. Göttingen, Hogrefe, 1996.
Spencer MB, Gundel A. A PC-based programm for
the assessment of duty schedules in civil aviation: The way forward. (Report
DERA/CHS/PP5/CR/980069/1.0). DERA,
Wierwille, W.W., Ellsworth,
Figure legends:
Figure 1:
The main factors causing fatigue are the time of day
(i.e. the circadian rhythm), time since last
sleep (sleep-related component), and the time on task. Fatigue models generally
assume that the circadian component has a sinusoidal and the sleep related
component an additive exponential effect on fatigue. The time on task effect
superimposes circadian and sleep-related fatigue. A break is supposed to “set
back” fatigue to the value which is caused by the combined influence of the
time of the day and sleep related factors.
Figure 2:
Computer programme ALERT predicting phases of reduced
alertness. The display shows alertness over the course of a day (x-axis: time)
for a total of 25 days. Red bars indicate phases of critically decreased
alertness. Times indicated by a blue bar represent sufficient alertness. Black
bars stand for suggestions for naps, grey for sleep phases. Free fields
represent times off duty.
Figure 3:
Truck drivers in the first training unit trying to
figure out individual signs of sleepiness. The photograph shows that the
training requires an active participation of the drivers.
Figure 4:
Unstable tracking task implemented on a hand held
computer for easy use in a driving context.
Figure 1
Figure 2
Figure 3
Figure 4
