The working time of taxi drivers in large
cities of Japan is characterized by the fact
that they work on alternate days for as long as
16 hr or more. The extent of prolonged hours of
work by taxi drivers is relatively well
documented. A statistical survey by the Ministry
of Transportation in 1973 revealed that the
average monthly working hours of taxi drivers
were more than 210 hr, the mean monthly sum of
portal-to-portal hours amounting to nearly 270
hr. While it has been suggested that their
workload has close relation to the incentive
wage system on the basis of the total of fares,
there are few reports making refèrence to
the interaction between the wage system and the
actual mode of work by taxi drivers. Further
studies seem necessary in this respect, since
the prolonged hours of work alone cannot be
easily rectified so long as the wage incentive
continues.
The long working hours of taxi drivers bring
with thera frequent night work. Special emphasis
bas been laid on fatigue and unhealthy working
conditions of taxi drivers by some field
reports. It is generally agreed that the usual
16-hr shift system produces marked changes of
physiological parameters. It has been further
pointed out that many drivers who are involved
in frequent night work, normally 13 nights per
month per taxi driver, are subject to diseases
of digestive and other organs. Thus the
Committee for Driving Work and Safety of the
Japan Association of Industrial Health submitted
a report in 1974 reviewing results of field
investigations and recommended intensified
occupational health measures for vehicle
drivers. The report further stated that the
working hours of occupational drivers should be
reduced to 8 hr or less within the 24-hr period
beginning from the shift start. In fact,
however, the 16-hr shift system is still widely
employed among taxi drivers, except for a small
number of companies where rotation systems with
a shift duration of 8-12 hr have been
introduced.
The above report also outlines the effects
of continued driving on the drivers'
performance, which seem as important as are
the environmental factors accounting for the
traffic accidents. It is the purpose of this
study to examine the effects of the long shifts
of alternate days on the taxi drivers'
performance. It is felt that the direct
observations of the manner of work of drivers
during the whole shift period may contribute to
the understanding of the behavioral changes
according to the shift system. Daily reports by
drivers, readings from individual tachograms,
and time records of daily living by means of the
diary method are also referred to.
[...]
BEHAVIORAL
CHANGES
Changes of driving performance by
time of the day
The taxi driving performance changed in a
very characteristic way according to the time of
the day. The major factors of the changes were
the rate of meeting passengers, the mean
distance per hire, and the general traffic
conditions. As shown by Fig. 3, which is based
on reports of 100 shifts, the hourly income was
not proportional to the rate of the duration of
running on hire and fluctuated greatly during
the day. The hourly income was relatively high
in the morning due to a high percentage of the
hired running. In later periods, the rate of
carrying passengers remained almost constant at
around 50%. The drop of income at the lunch time
and in the evening was due both to taking meals
and to smaller fares per hire in the evening. A
sharp rise in the last periods later than 22:00
is seen both for the fare per hire and for the
hourly income. The high levels continued even
after midnight. The income per hire was only Y
340 in the first period of the shift, a little
more than the minimum fare of Y 280, indicating
that most passengers used the car for short
distances. The mean income per hire was
maintained at around Y 450 to Y 550 from 9:00 to
22:00, but it rose to Y 600 to Y 700 in the
hours later than 22:00. It was mainly because
passengers at night usually hired the taxis for
middle- or long-distance drives.
In addition to this, the difference in
traffic conditions between daytime and night
also favored the income increase at night. It is
shown by Fig. 4 which indicates the mean driving
speed and the mean fraction of the time when the
car remains stationary for different periods of
the shift. The data for 2 whole shifts are
shown. Both curves were almost symmetrical. The
mean speed was only 21-22 km./hr in the morning
and evening rush hours when the car was brought
to a halt for about 35 %. of the time. The
average speed in the middle of the day was about
30 km/hr, while it increased to 34-39 km/hr in
the periods later than 22:00, when the car
stopped remarkably less for only 15 % of the
running time.
Thus, the increase of the gross income at
night was evidently the result of the combined
effects of longer distances travelled and more
favorable traffic conditions in late hours than
in the other periods. Such a situation is
characteristic of large cities like Kawasaki.
These facts explain the reason why the late
night hours are very important for taxi drivers
in order to increase the total amount of income
for the day's work. It was common to, all the
drivers inquired that they ,waited for the night
hours and tried to increase the income at night
rather than in the daytime. Very unfortunately,
these relatively gainful periods began at
around 22:00, when nearly 15 hr had past since
the shift start and when the drivers were
influenced by the prolonged work and the
circadian rhythm.
Changes of subsidiary behavior by
time of the day
The taxi drivers' subsidiary behavior which
is apparently irrelevant to the driving
performance was recorded by an observer who sat
in the assistant's seat during the whole shift.
Table 3 summarizes the results of observations
on two shifts for six different categories of
subsidiary activities in four periods of driving
interspersed by pauses for breakfast (around
9:00), lunch (around 13:00), and supper (around
19:00). The mean rate per 5 min of each kind of
activities is indicated. The mean rate was 4.32,
2.54, 3.81, and 4.43 in periods I,II, III, and
IV, respectively. The changes were mainly due to
changes in the rate of the upper limb
movements and that of changing sitting
positions. The rate of looking aside while
driving decreased slightly as the time
progressed. More remarkable changes were found
for those activities which had relatively low
rates at the beginning of the shift. Compared
with the first period, the subsidiary
shoulder-neck movements became 1.5 times
more frequent in period IV, lower limb
movements 2.2 times, and
yawning 3.4
times.
The cumulative frequency distribution of
those subsidiary activities are illustrated in
Fig. 5 for six different kinds and for the four
periods. The distribution pattern differed
greatly among the periods. In the case of the
upper limb movements, they tended to repeat more
than twice per 5-min interval in more than a
quarter of the observations of periods III and
IV, while lower limb movements, shoulder-neck
movements, and yawning were liable to
repeat twice or more when they occurred in
period IV. Such a tendency was most
conspicuous for yawning which was specific to
period IV, often appearing in bursts in drowsy
conditions.
A certain relation between the rate of
subsidiary behavior and the driving performance
is suggested by Fig. 6 showing changes in the
total rate of such activities separately for
driving with frequent stopping for 30% or more
of the time and that with stopping for less than
30% of the time. Driving with less frequent
stopping produced less frequent activities in
the first period, but in later periods,
especially in period IV, the subsidiary
activities were seen to increase greatly also in
driving without frequent stopping.
Figure 7 gives recordings of subsidiary
activities on three occasions within a shift of
a driver. It is demonstrated that some of
subsidiary activities became correlated with
the driving performance as fatigue
developed. At 11:30-12:00, in the 5th hr of
work, the driver was in good humor and was
attempting to locate passengers utilizing
wireless communication with the service office.
At this time of the day, his subsidiary
activities were least frequent, and each of such
movements appeared rather sporadically without
any distinct interrelations with the others. At
22:00-22:30 when he had been out for more than
14 hr, subsidiary movements were very frequent
and multifarious. Yawning was seen in bursts
together with some sitting position changes and
with the suppression of upper limb
movements. Repeated changes of the sitting
posture seemed to have resulted not only from
fatigue but from drowsiness. This was more
clearly confirmed by similar recordings at
2:00-2:30, nearly 19 hr from. the departure
time. When a passenger got off at 2:30, the
driver confessed that he had been very drowsy
some time ago. Actually at about 2:10, the
driver had a very sleepy look and even closed
his eyes when the car stopped for a while at a
red signal. The combination of yawns with
sitting posture changes was typical for that
period, the other subsidiary movements being
temporarily infrequent. Later on, the latter
increased again.
CONCLUSIONS
Based on the results of this study and our
previous report, it can be concluded that the
alternate-day shift systerm of taxi drivers
virtually influences the daily course of their
driving performance as well as significantly
degrade their performance capacity. It has been
reported by SAKAI and TAKAHASHI (1975) that 49%
of the inquired taxi drivers had to go to the
next shift without sufficient recovery from the
fatigue of the previous shift, and that the
alternate-day shifts of 16 hr produced a
significantly more frequent rate of subjective
fatigue symptoms and sleep deficit than did the
12-hr shift system. Although the present study
was not specifically concerned with
physiological changes of drivers, these
observations lend support to the assuraption
that the taxi drivers are working beyond the
physiologically tolerable limit suggested by
many investigations and by standards of road
traffic.
Studies on effects of prolonged driving have
been generally concentrated along two lines of
examination, one of these has been concerned
with changes in driving behavior, and the other
has examined the physiological and psychological
correlates of driving. There are enough
evidences indicating a demonstrable fatigue
effect of a long automobile drive, while the
relevance of behavioral changes to driving
situations has been suggested by some
authors
The examinations of the drivers' subsidiary
activities may also be useful, especially when
the intensity of driving work changes towards
the end of a shift, as is the case for the city
taxi drivers. Although the observed changes of
subsidiary behavior in the present study may not
be directly relevant to those in monotonous
factory work reported by KISHIDA (1973) and
others, effects of monotonic driving in a fixed
posture should also be taken into account. It
would be reasonable to assume that the
remarkable changes in the taxi drivers'
subsidiary behavior were produced not only by
prolonged working hours but also by degradation
due torather monotonic driving situations. Night
work seemed to accelerate both of those
effects.
The effects of the alternate-day shift
system have a close relation with the incentive
wage system. The intensity of work in terms of
the driving speed and distance is enhanced
towards the end of the long shift seemingly
'voluntarily.' This is because intensive work
during late night hours secures the daily
income, since the running distance on hire can
be increased more easily in these hours than in
the other periods of the day. The same reason
also accounts for the prolongation of working
hours and reduction of resting periods. Results
of our previous study has indicated that a
majority of the taxi drivers are conscious of
the harmful effects of long hours of work, 51 %
demanding a reduction of the shift length, 35%
urging a reduction in the number of days of work
instead, and only 9 % favoring the present
system (. Although the situation is rather
complex due to the interactions of the incentive
wage system. with traffic circumstances in
cities, effective means should be explored to
eliminate the existing alternate-day shift
system of the taxi drivers of this country.
